研究者詳細 - 神谷　亨

論文 - 神谷　亨

分割表示 >> ／全件表示757 件中 1 - 757 件目

Automatic Identification of CTC in Fluorescence Microscopy Images Based on A Lightweight Hybrid Network Model 査読有り国際誌

Nakamichi K., Tsuji K., Yoneda K., Kamiya T.

Journal of Advances in Information Technology 16 ( 5 ) 648 - 654 2025年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

—Circulating Tumor Cells (CTC) are expected to be a useful biomarker for cancer metastasis. CTC analysis can be used to assess cancer status and the therapeutic effects of anticancer drugs. Pathologists analyze blood samples from images taken with a fluorescence microscope. However, the number of CTCs in blood is very small, which is a time-consuming task. In this paper, we propose an automatic CTC identification method from fluorescence microscopy images. In the proposed method, we detect cell regions using a selective enhancement filter and blob analysis. Then, we identify the CTC using a SqueezeNetbased Convolutional Neural Network (CNN) model with Spatial Pyramid Pooling (SPP). We apply the proposed method to 5040 microscopy images and evaluate its effectiveness. The experimental results show that the proposed method has a true positive rate of 97.30%, a false positive rate of 2.069%, and an Area Under the Curve (AUC) of 0.991.

DOI： 10.12720/jait.16.5.648-654

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=105005978773&origin=inward

Automatic Classification of Respiratory Sounds Based on Convolutional Recurrent Neural Network and Bagging k-Nearest Neighbour 査読有り

Minami, Lu, Kamiya, Kido

Journal of Image and Graphics 13 ( 1 ) 46 - 51 2025年01月

Foreword 査読有り国際誌

Kamiya T.

ICBSP 2024 - Proceedings of the 2024 9th International Conference on Biomedical Imaging, Signal Processing 2025年01月

Retinal Layer 3D Segmentation in Wide-Field OCT images using Self-Supervised Pre-Training of Swin Transformers 査読有り国際誌

Yiqing Ye, Guangxu Li, Yining Dai, Kailu Wang, Kang Wang, Zhina Wang, Jiajia Liu, and Tohru Kamiya

International Conference on Computing and Pattern Recognition 38 - 43 2025年01月

Detection of Lung Regions from LDCT Images and 3D Image Registration Using FFD 査読有り

Tanaka C., Kamiya T., Terasawa T., Aoki T.

Proceedings 2025 9th International Conference on Biomedical Engineering and Applications Icbea 2025 26 - 30 2025年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, lung cancer is a serious disease worldwide. To overcome the disease, early detection and early treatment of lung cancer are important tasks. In Japan, CT (Computed Tomography) examination is performed in many medical facilities for visual screening. However, there is a problem that huge number of images taken by CT is a burden to the doctor. Therefore, the CAD (Computer Aided Diagnosis) system is in the spotlight to reduce the burden. On the other hand, in CT screening, LDCT (Low Dose Computed Tomography) is desirable considering radiation exposure. However, the LDCT image is characterized by lower image quality at lower dose. Therefore, a CAD that can be applied to LDCT is needed. In this paper, we propose a lung field detection method and a three-dimensional (3D) registration method to generate temporal subtraction images that can be applied to LDCT images. Our method consists of the following steps: detection of lung regions using an active contour model, correction of corresponding slices between images, global matching based on the center of gravity, setting the VOI (volume of interest), and local matching on the VOI. In this paper, we apply our method to LDCT images of 7 cases. In addition, we conducted a comparison experiment with other registration methods. In the proposed method, a decrease of 7.55% in FWHM and 34.4% in sum of histogram of temporal subtraction images compared to the other method, and its usefulness was verified.

DOI： 10.1109/ICBEA66055.2025.00013

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=105008497197&origin=inward

Recognition of Plastic Bottles Region Using Improved DeepLab v3+ 査読有り国際誌

Murata Y., Kamiya T.

Proceedings of International Conference on Artificial Life and Robotics 819 - 822 2025年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Factory automation is one solution to the labor shortage. We focus on the sorting of plastic bottles in waste disposal plants and try to automate the process using robotic arms. In this paper, we propose an image analysis method for the recognition of plastic bottles limited to 500ml capacity. The method is semantic segmentation, and the deep learning model is DeepLab v3+. Modifications using ECA Block and Mish function show improvements at the points of misrecognition with the base model.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85219582103&origin=inward

Non-Invasive Classification of EGFR Mutation from Thoracic CT Images Using Radiomics Features and LightGBM 査読有り国際誌

Takahashi R., Kamiya T., Terasawa T., Aoki T.

Proceedings of International Conference on Artificial Life and Robotics 823 - 826 2025年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Cancer caused 9.7 million deaths in 2022, including 1.8 million from lung cancer the leading cause of cancer death. EGFR mutation testing is essential for lung cancer treatment planning, but it is invasive and visual identification from chest CT images is difficult. This paper proposes a computer-aided diagnosis system to identify EGFR mutation status. Lung tumor regions were automatically extracted and radiomics features were obtained. Dimensionality reduction was performed using null importance, variance inflation factor, and recursive feature elimination. The method was applied to 143 cases and achieved an accuracy of 59.1%, a true positive rate of 54.3% and a false positive rate of36.1%. The results suggest that CAD (Computer-Aided Diagnosis) systems can improve the non-invasive detection of EGFR mutations in lung cancer.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85219512260&origin=inward

Identification of lung nodules based on combining multi-slice CT images and clinical information 査読有り国際誌

Nishitaki Y., Kamiya T., Kido S.

Proceedings of International Conference on Artificial Life and Robotics 844 - 847 2025年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Although chest CT scans are an effective means of diagnosing lung cancer, there are still problems such as the heavy burden on physicians. To solve this problem, computer-aided diagnosis (CAD) systems are being introduced. Conventional CAD systems are based on a method that uses only image information. In this study, we propose a method for identifying nodular shadows that integrates a composite image created from multi-slice CT images and clinical information such as the patient's age, sex, and medical history in the medical record. The proposed method extracts features from the multi-section CT and clinical information, respectively, integrates the features, and then performs binary classification of nodules and vessels using a classifier. The proposed method achieved a very high accuracy of Accuracy=0.983, TPR=0.987, and FPR=0.018.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85219552429&origin=inward

Detection of Lung Nodules from Temporal Subtraction CT Image Using Elastic Net-Based Features Selection 査読有り国際誌

Baba N., Kamiya T., Terasawa T., Aoki T.

Proceedings of International Conference on Artificial Life and Robotics 827 - 830 2025年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

CT (computed tomography) is mainly used to diagnose lung cancer. Many CT images impose a heavy burden on visual screening, so a CAD (computer-aided diagnosis) system is expected to reduce the burden. In this paper, we propose an image analysis method to detect lung nodules from chest CT images using machine learning techniques. The best results were obtained for the method using Light GBM with feature reduction by Elastic Net.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85219531661&origin=inward

Detection of Lung Nodules from CT Image Based on Ensemble Learning 査読有り国際誌

Baba N., Kamiya T., Terasawa T., Aoki T.

Proceedings of International Conference on Artificial Life and Robotics 831 - 834 2025年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Lung cancer is the most frequently diagnosed cancer worldwide and the leading cause of cancer-related deaths, making early detection and treatment crucial. Temporal subtraction system, one of the CAD, emphasize the differences between the current and previous images. In this study, radiomics features are extracted as explanatory variables from the temporal subtraction images. Feature selection is performed using Elastic Net, followed by the application of machine learning methods. Finally, ensemble learning is applied to classify unknown data into two categories: positive and negative lung nodules.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85219582462&origin=inward

Classification of Heat Transfer Coefficient Using Deep Learning Incorporated Boiling Images 査読有り国際誌

Mitsuyama F., Umeno R., Takakuma K., Yabuki T., Kamiya T.

Proceedings of International Conference on Artificial Life and Robotics 797 - 801 2025年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Boiling cooling has been used as a cooling method for electronic devices due to its high heat transfer coefficient(HTC). The regularity of the boiling phenomenon is a crucial factor in the development of more efficient cooling systems. To develop such systems, it is essential to accurately measure the HTC, which is closely related to the boiling phenomenon. In this paper, we propose a method for predicting the HTC of two different heat transfer surfaces using deep learning with boiling sound and boiling images as inputs. The proposed method achieves an accurate improvement of 2.0% and 16.7% compared to models using only boiling sound and boiling images as input, respectively.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85219565340&origin=inward

Automatic Classification of Respiratory Sounds by Improving the Loss Function of ResNet 査読有り国際誌

Oshima R., Kamiya T., Kido S.

Proceedings of International Conference on Artificial Life and Robotics 793 - 796 2025年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Respiratory diseases cause 8 million deaths annually, and this number is expected to increase. Breath auscultation, a primary diagnostic method, is non invasive, repeatable, and immediate, but faces challenges such as reliance on skilled practitioners, difficulty in quantitative assessment, and limited accessibility in developing regions or disaster sites. To address these issues, we developed a deep learning-based breath sound classification system using the ICBHI 2017dataset. Our method classifies breath sounds into four categories: Normal, Crackle, Wheeze, and Crackle and Wheeze. We use ResNet-34 as the base model, which is enhanced with CBAM for better spatial and channel feature extraction. To deal with class imbalances, we incorporate Focal Loss. The system achieves Accuracy of 0.732, SE of 0.607, SPof 0.843, and ICBHI Score of 0.725.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85219504987&origin=inward

Automatic Classification of Respiratory Sounds Based on Convolutional Recurrent Neural Network and Bagging k-Nearest Neighbor 査読有り国際誌

Minami K., Lu H., Kamiya T., Kido S.

Journal of Image and Graphics (United Kingdom) 13 ( 1 ) 46 - 51 2025年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

Respiratory diseases or lung diseases such as asthma bronchiectasis cystic fibrosis are a serious disease. Approximately 8 million people died in each year by chronic obstructive pulmonary disease, lower respiratory tract infections, trachea, bronchial and lung tumors. In addition, COVID-19 is prevalent worldwide in recent years. To analyze these symptom, auscultation of respiratory sounds is very important for screening the respiratory disease. However, there is no quantitative evaluation method for the diagnosis of respiratory sounds until now. To overcome this problem, it is necessary to develop a system to support the diagnosis of respiratory sounds. In the development of support system for auscultation, research by a large-scale, open database used in ICBHI (The International Conference on Biomedical and Health Informatics) 2017 Challenge is in progress. It is expected that a general purpose and highly accurate system will be developed using this dataset. We describe an algorithm for the automatic classification of the respiratory sounds as crackles, wheeze, both, and normal. We improve the classification rates compared with other ICBHI 2017 Challenge teams based on three components. First, we generate the spectrogram images by short-time Fourier transformation. We also extract features using a convolutional recurrent neural network. Third, we classify unknown respiratory sounds by bagging k-nearest neighbor algorithm. In the experiment, we applied our proposed method to 920 respiratory sound data which is obtained by the ICBHI Challenge data sets, and achieved Sensitivity with 0.670, Specificity with 0.863, ICBHI Score with 0.766 respectively. Also, area under the curve based on receiver operating characteristic curve of normal class with 0.892, crackle with 0.891, wheeze with 0.874, both with 0.883 were obtained respectively.

DOI： 10.18178/joig.13.1.46-51

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85217135665&origin=inward

Classification of Histological Types of Primary Lung Cancer from CT Images Using Clinical Information 査読有り国際誌

Honda, Kamiya, Kido

International Conference on Biomedical Engineering and Technology 128 - 132 2024年12月

Retinal Layer 3D Segmentation in Wide-Field OCT images using Self-Supervised Pre-Training of Swin Transformers 査読有り国際誌

Yiqing Ye, Guangxu Li, Yining Dai, Kailu Wang, Kang Wang, Zhina Wang, Jiajia Liu, Tohru Kamiya

International Conference on Computing and Pattern Recognition ( International Conference on Computing and Pattern Recognition ) 38 - 43 2024年10月

A Surface Projection Based Medical Virtual Teaching Platform 査読有り国際誌

Wang C., Li G., Kihara N., Kamiya T., Zhou J.

ACM International Conference Proceeding Series 98 - 102 2024年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Augmented reality (AR) technology is rapidly growing and widely applying in medical education. In this paper, a projection platform with RGB-D camera feedback is proposed for atomy and clinic education. In terms of the alignment of visual image and training model, we introduced an automatic calibration algorithm; To correct the distortion of projected image, a real-time correction method was proposed. According to the experimental results, the shape fidelity and time-efficient was confirmed, which meet the requirements of medical training applications.

DOI： 10.1145/3691521.3691526

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85212835774&origin=inward

Rigid Image Registration for Head MRI Based on 3DCNN Incorporated Global Information 査読有り国際誌

Baba S., Li G., Kamiya T.

ACM International Conference Proceeding Series 62 - 65 2024年10月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Since Alzheimer's disease is incurable nowadays, early detection and treatment of Alzheimer is extremely important. One of the diagnostic methods is to evaluate the brain atrophy with the progression of Alzheimer's disease according to MRI image. Computer-Aided Diagnosis (CAD) system with medical image processing and artificial intelligence provides an automatic way to analyze the degree of brain atrophy to physicians. As one of the critical steps, three-dimensional (3D) image registration method is utilized to analyze temporal variation of the brain volume. In this paper, a 3D convolutional neural network (3DCNN) based method with the blocks of 3D Grid Loss, Stem Block, Vision Transformer, and s decoupled fully connected (DFC) Attention is proposed. We applied our method to the ADNI dataset, which includes 328 head MRI cases. The results showed that the Normalized Cross-Correlation (NCC) was 0.952, with more than 0.13 improvement comparing the up-to-date methods.

DOI： 10.1145/3691521.3691538

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85212858679&origin=inward

Classification of Histological Types of Primary Lung Cancer from CT Images Using Clinical Information 査読有り国際誌

Honda N., Kamiya T., Kido S.

ACM International Conference Proceeding Series 128 - 132 2024年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Identification of primary lung cancer is very important because it influences the course of treatment, especially for small cell carcinomas, which metastasize rapidly and have to be detected at an early stage. In addition to imaging, clinical information is often used in CAD (computer aided diagnosis) systems. Especially, clinical information such as smoking history, which is considered to be important in the diagnosis of lung cancer. In this paper, we propose a method to identify primary lung cancer by adding clinical information from medical records in addition to images to improve the accuracy of diagnosis. We use tumor images surrounded by rectangular regions from CT images in an open dataset as input images and train the method based on deep learning technique. We evaluate the proposed method by discriminating tumors from unknown data. In our experiments, we found that the accuracy was improved about 6% when clinical information was added to 604 images, which included four classes of cancer; adenocarcinoma, small cell carcinoma, squamous cell carcinoma, and large cell carcinoma.

DOI： 10.1145/3678935.3678969

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85209395567&origin=inward

改良型U-Net を用いた低線量CT 画像の画質改善法

澤田, 村上, 李, 神谷

FIT2024（第23 回情報科学技術フォーラム） 231 - 234 2024年09月

カルテ情報を加味した多断面CT 画像からの結節状陰影の識別

西滝、神谷、木戸

FIT2024（第23 回情報科学技術フォーラム） 227 - 239 2024年09月

色彩の分離に基づいた3次元点群顔データからの口唇領域の抽出

木原，木村（野元），大河内，李, 中村, 神谷

第43回日本医用画像工学会大会 278 - 279 2024年08月

Ensemble Multiscale Attention UNetによる胸部CT 画像からの肺腫瘍領域の自動抽出

高橋神谷，寺澤，青木

第43回日本医用画像工学会大会 216 - 217 2024年08月

Multi-robot navigation based on velocity obstacle prediction in dynamic crowded environments 査読有り国際誌

Chen Y., Wang Y., Li B., Kamiya T.

Industrial Robot 51 ( 4 ) 607 - 616 2024年07月

記述言語：英語掲載種別：研究論文（学術雑誌）

Purpose: The purpose of this paper is to propose a new velocity prediction navigation algorithm to develop a conflict-free path for robots in dynamic crowded environments. The algorithm BP-prediction and reciprocal velocity obstacle (PRVO) combines the BP neural network for velocity PRVO to accomplish dynamic collision avoidance. Design/methodology/approach: This presented method exhibits innovation by anticipating ahead velocities using BP neural networks to reconstruct the velocity obstacle region; determining the optimized velocity corresponding to the robot’s scalable radius range from the error generated by the non-holonomic robot tracking the desired trajectory; and considering acceleration constraints, determining the set of multi-step reachable velocities of non-holonomic robot in the space of velocity variations. Findings: The method is validated using three commonly used metrics of collision rate, travel time and average distance in a comparison between simulation experiments including multiple differential drive robots and physical experiments using the Turtkebot3 robot. The experimental results show that our method outperforms other RVO extension methods on the three metrics. Originality/value: In this paper, the authors propose navigation algorithms capable of adaptively selecting the optimal speed for a multi-robot system to avoid robot collisions during dynamic crowded interactions.

DOI： 10.1108/IR-12-2023-0337

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85189666801&origin=inward

A Temporal Subtraction Technique for Phalange CR Image Using CNN 査読有り国際誌

Ono H., Kamiya T., Aoki T.

Journal of Advances in Information Technology 15 ( 7 ) 873 - 878 2024年07月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

X-ray examinations are widely used in the diagnosis of Rheumatoid Arthritis (RA). However, the condition of many phalanges and joints must be evaluated visually, which causes a lack of objectivity due to subjective evaluation by the physician and an increased workload for the physician in reading the images. In this paper, we propose an image analysis method for hand Computed Radiography (CR) images based on the temporal subtraction method to support the diagnosis of rheumatoid arthritis. The proposal method consists of three steps. First, a Convolutional Neural Network (CNN) model for semantic segmentation, which is efficient in terms of computational complexity and accuracy, is proposed to extract phalangeal regions. Second, a geometric-matching CNN with instance-specific optimization is used to align the phalangeal regions. Finally, the current image and the aligned past image are subtracted to visualize the temporal changes. We applied the proposed method to hand CR images and confirmed its effectiveness.

DOI： 10.12720/jait.15.7.873-878

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85212714373&origin=inward

Detection of mutation of driver gene from the CT images based on ensemble learning 査読有り国際誌

Kamiya, Y. Yoshifuku, T. Terasawa, T. Aoki

Computer Assisted Radiology and Surgery Proceedings of the 38th International Congress and Exhibition ( S97-S98 ) 2024年06月

Optimization of Corneal Parameters in Refractive Surgery 査読有り国際誌

Wang L., Li G., Liu Z., Huang Y., Shi B., Kamiya T.

ACM International Conference Proceeding Series 144 - 148 2024年04月

担当区分：最終著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This study establishes a personalized human eye model for application in refractive surgery and proposes a method to optimize corneal parameters during surgery. The input data includes preoperative spherical equivalent, corneal curvature, and corneal thickness of the patient. Linear regression analysis and neural network methods are employed to predict corneal curvature and thickness, respectively, during the surgery based on the preoperative examination results. The results demonstrate that the neural network method performs accurately in predicting corneal ablation curvature, while the linear regression analysis method is effective in predicting ablation thickness. Utilizing these methods to predict corneal parameters in refractive surgery can assist surgeons in achieving precise corneal ablation and significantly reduce surgical risks.

DOI： 10.1145/3669754.3669776

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85203828242&origin=inward

Deep Learning Based Prediction of Heat Transfer Coefficient Using Spectrogram Images from Boiling Sound 査読有り国際誌

Mitsuyama, Umeno, Yabuki, Kamiya

29th International Conference on Artificial Life and Robotics 914 - 917 2024年02月

Unsepervised image registration based on Residual-connected DRMINE for diagnostic metastatic bone tumors 査読有り国際誌

Baba S., T. kamiya, T. Terasawa, T. Aoki

29th International Conference on Artificial Life and Robotics 1016 - 1020 2024年02月

Classification of circulating tumor cell clusters by morphological characteristics using convolutional neural network-support vector machine 査読有り国際誌

Park J., Ha S.M., Kim J., Song J.W., Hyun K.A., Kamiya T., Jung H.I.

Sensors and Actuators B: Chemical 401 2024年02月

記述言語：英語掲載種別：研究論文（学術雑誌）

Metastasis is the leading cause of cancer-associated deaths, and the circulating tumor cell (CTC) cluster plays a significant role as a precursor to metastasis. Thus, there is a great demand for high-throughput identification of rare CTC clusters for prognostic diagnosis. Immunofluorescence staining is considered the gold standard for identifying CTCs. However, as CTC clusters are extremely heterogeneous cells, multiple staining markers are required for accurate discrimination. Additionally, the staining procedure is tedious and the analysis of large amounts of stained images is labor-intensive and error-prone. Recently, machine learning-based identification has been introduced to achieve accurate discrimination, but they still rely on immunofluorescence staining for dataset preparation. In this study, we developed a hybrid algorithm, a convolutional neural network support vector machine (CNN-SVM), for the accurate classification of CTC clusters without immunofluorescence staining. In dataset preparation, the Wright–Giemsa staining was used to highlight the morphological features of the cells. Four morphological characteristics that display the unique traits of cells were drawn with each eigenvector, as a result of learning, the algorithm classified CTC clusters of various configurations with a sensitivity and specificity of > 90%. Therefore, our algorithm is expected to be a powerful tool for cancer diagnosis and prognosis.

DOI： 10.1016/j.snb.2023.134896

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85176212805&origin=inward

ランドマーク点の時系列変化と3次元構造に基づいた顔の非対称度評価法査読有り

木原，野元，大河内，李（天津工大），中村，神谷

バイオメディカル・ファジィ・システム学会誌 ( バイオメディカル・ファジィ・システム学会誌 ) 25 ( 2 ) 57 - 64 2024年02月

Three-dimensional quantification of postoperative facial asymmetry in patients with unilateral cleft lip and palate using facial symmetry plane 査読有り国際誌

Shiigi, Okawachi, Kamiya, Hosoki, Nomoto, Ratman, Amir, Ishihata, Nakamura

The Cleft Palate Craniofacial Journal 61 ( 2 ) 209 - 218 2024年02月

Unsupervised image registration based on Residual-connected DRMINE for diagnostic metastatic bone tumors 査読有り

Baba S., Kamiya T., Terasawa T., Aoki T.

Proceedings of International Conference on Artificial Life and Robotics 1016 - 1020 2024年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Metastatic tumors are frequently identified through follow-up surveillance using computed tomography (CT) scans. However, CT scans produce more than 100 images in an examination, which imposes a significant burden on radiologists and entails a potential risk of misdiagnosis. Temporal subtraction is utilized in Computer-Aided Diagnosis (CAD) and proves to be an effective technique in aiding image interpretation process for the radiologists. In this study, we focus on the preliminary stage of CAD development specialized in bone metastasis extraction, with a particular emphasis on rigid registration. We propose a novel rigid registration technique by augmenting DRMINE, which estimates mutual information using neural networks, with skip connections and normalization. From the three datasets, ten images were selected randomly from the cervical, thoracic, and lumbar regions. These images were then augmented with rotation as well as horizontal and vertical translations to create modified versions. The registration accuracy was assessed based on the Full Width at Half Maximum (FWHM) of the difference images. In the proposed method, FWHM values for the thoracic and lumbar regions of the spine exhibited a maximum reduction rate of 2.8% and a minimum reduction rate of 0.533%. However, the cervical spine region exhibited superior FWHM results with DRMINE compared to the proposed methodology. The proposed method was influenced by the capture area, but it indicated the potential to provide stable registration as the standard deviation decreased for all FWHM values.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85190241162&origin=inward

Physics-based Optical Coherence Tomography Angiography (OCTA) Image Correction for Shadow Compensation 査読有り国際誌

Li G., Wang K., Dai Y., Zheng D., Wang K., Zhang L., Kamiya T.

IEEE Transactions on Biomedical Engineering 2024年01月

担当区分：最終著者記述言語：英語掲載種別：研究論文（学術雑誌）

Optical coherence tomography (OCT) is being widely applied in clinical studies to investigate insight into the retina under the retinal pigment epithelium. Optical coherence tomography angiography (OCTA) is one of the functional extensions of OCT, for visualizing retinal circulation. Due to obstruction of light propagation, such as vitreous floaters or pupil boundaries, OCTA remains challenged by shadow artifacts that can disrupt volumetric data. Detecting and removing these shadow artifacts are crucial when quantifying indicators of retinal disease progression. We simplified an optical attenuation model of shadow formation in OCTA to a linear illumination transformation. And learn its parameters using an adversarial neural network. Our framework also consists of a sub-network for shadows automatic detection. We experimented our method on 28 OCTA images of normal eyes and compared the non-perfusion area (NPA), an index to measure retinal vascularity. The results showed that the NPA adjusted to a reasonable range after image processing using our method. Furthermore, we tested 150 OCTA images of synthesis artifacts, and the mean absolute error(MAE) values reached 0.83 after shadow removal.

DOI： 10.1109/TBME.2024.3478384

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85206991320&origin=inward

Super Resolution of Remote Sensing Image Using Improved SwinIR 査読有り国際誌

Matsumoto S., Kamiya T.

2024 Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2024 2024年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, the number of satellite launches has been increasing. One of the reasons for this is the growing market for space development, which has led to the release of a large amount of satellite data. Among them are satellite images, which capture a wide range of ground conditions and are used in various fields, such as the extraction of building areas. However, human analysis of these images is time-consuming and inefficient, so deep learning is expected to be used for image analysis. The images that can be used to recognize objects in satellite images are high-resolution images taken by high-performance satellites, but such images are expensive, and the cost of acquiring a large number of images is high. In this paper, we propose a method to reconstruct a high-resolution satellite image by super-resolution from a low-resolution satellite image using the Vision Transformer model. Image Restoration Using Transformer (SwinIR) is adopted as the base model. To improve the accuracy, Knowledge Distillation and Multi-Layer Perceptron with Gating Unit are incorporated into the base model, and good results are obtained.

DOI： 10.1109/SCISISIS61014.2024.10759940

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85214686521&origin=inward

Registration of Spinal Region on CT Images Based on SuperPoint with Convolutional Block Attention Module 査読有り国際誌

Nabeshima K., Terasawa T., Kamiya T., Aoki T.

2024 Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2024 2024年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Cancer is currently the leading cause of death in Japan, and the number of cancer patients is increasing every year. The number of patients with bone metastases from cancer is also increasing. The spine is a common site of bone metastasis, and cancer metastasis to the spine can cause serious symptoms such as hypercalcaemia and nerve paralysis due to spinal cord compression. As a result, it can affect patients' quality of life. Early detection and treatment of bone metastases is therefore important. To analyse the image information, CAD (Computer Aided Diagnosis) is introduced in image registration, which is obtained by temporal subtraction technique. One of them is CT (Computed Tomography) image-based registration technique by taken at different times of the same subject and performs subtraction to enhance changes in the lesion over time. In this approach, the registration of the same slice is important for the generation of temporal subtraction image. In this paper, we propose a method to register both images obtained different time series based on SuperPoint algorithm incorporated CBAM (Convolutional Block Attention Module) as a method to select the same slice and register the spinal region on the CT image.

DOI： 10.1109/SCISISIS61014.2024.10759972

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85214674575&origin=inward

Registration Method for Head CT and MR Images Based on TransMorph Incorporated ASPP 査読有り国際誌

Murakami T., Yamamoto A., Kamiya T.

2024 Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2024 2024年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, with the development of diagnostic imaging equipment, image fusion, which fuses medical images from multiple modalities, has been proposed as a diagnostic imaging application. This has the advantage of integrating images with different characteristics and resolutions into a single composite image, which compensates for the missing information of each and facilitates the reading of lesions. Examples of the use of fusion images include surgical simulation and CyberKnife-based surgery. To create a fusion image, the two images must be aligned. Currently, this is often done manually, which increases the burden on the surgeon and introduces problems such as variations in accuracy depending on the surgeon's skill. Therefore, this paper describes the development of an automatic registration method for creating a fusion image between head CT and MRI, which is necessary for CyberKnife-based surgery of brain tumors. The method uses TransMorph, a registration model, as the base model, and a new model is constructed by adding ECAM and ASPP. As a result, the proposed method with the introduction of ASPP to the base model shows an improvement in the mutual information value and succeeds in improving the accuracy.

DOI： 10.1109/SCISISIS61014.2024.10760204

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85214692089&origin=inward

Detection of Temporal Changes from Satellite Images Based on Improved SNUNet-CD 査読有り国際誌

Watanabe Y., Kamiya T.

2024 Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2024 2024年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Change Detection (CD) technology is an integral part of Earth observation in remote sensing (RS) image analysis. It is widely used in areas such as natural disasters, urban development monitoring, land resource management, and surveillance. As a pixel-to-pixel prediction task, the CD technique is location sensitive and requires more attention to feature extraction. Recently, deep learning (DL)-based CD methods have made various efforts to extract deep semantic features, but they ignore the importance of shallow layer information, including high-resolution and fine-grained features, which leads to uncertainty in the change-ground entities. To address these problems, SNUNet-CD is a model that mitigates the loss of localization information in the deeper layers and improves detection accuracy by using dense connections. Although it shows higher detection accuracy than other methods and provides a good trade-off between accuracy and computational complexity, it has difficulty in exploiting the different semantic and detailed information contained in deep and shallow features, which often blurs the target boundaries of the identified changes. Therefore, in this paper, we propose a model that incorporates Attention Gates, an attention mechanism that selectively highlights and updates salient regions of the feature map, into the skip-connection part to effectively link features of different levels. The performance evaluation results showed that Precision=0.9073, Recall=0.9127, and Fl-score=0.9099, confirming the effectiveness of introducing Attention Gates.

DOI： 10.1109/SCISISIS61014.2024.10760239

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85214693174&origin=inward

A Denoising Method from Low-dose CT Images Based on Improved U-Net with Attention Mechanism 査読有り国際誌

Sawada S., Li G., Murakami S., Kamiya T.

2024 Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2024 2024年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Japan has the highest number of CT scanners per million population among developed countries, and the annual medical radiation dose is about four times the world average. Screening of high-risk populations with low-dose CT has been shown to reduce lung cancer mortality. However, in order to obtain image quality comparable to that of normal-dose CT, it is necessary to accurately estimate the image noise of low-dose CT. In this paper, we apply a noise reduction method using convolutional neural networks: the CNN model is based on U-Net, a fully convolutional network with encoder-decoder structure, with two convolutional layer modifications and three attention mechanisms. Experimental results using whole body slice CT images of piglets show that the proposed method is useful by comparing with CT images of normal dose and by evaluating image quality using peak signal-to-noise ratio (PSNR).

DOI： 10.1109/SCISISIS61014.2024.10759999

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85214672500&origin=inward

呼吸音から生成した複数画像による呼吸器疾患の自動分類

田端, 陸, 神谷, 間普, 木戸

電子情報通信学会技法IE2023-32 2023年12月

4次元データからの口唇領域の抽出に基づいた顔の対称性解析

木原、木村、大河内、李(天津工大)、中村、神谷

信学技法 ( 信学技法 ) MI2023-18 1 - 4 2023年11月

Facial Symmetry Analysis Using Temporal Change in Landmark from A Video Image of 3D Point Cloud 査読有り国際誌

Kihara, Nomoto, Okawachi, Li（天津工大）, Nakamura, Kamiya

International Conference on Biomedical and Bioinformatics Engineering 11 - 16 2023年11月

Facial Symmetry Analysis Using Temporal Change in Landmark from A Video Image of 3D Point Cloud 査読有り国際誌

Kihara N., Kimura-Nomto N., Okawachi T., Li G., Nakamura N., Kamiya T.

ACM International Conference Proceeding Series 11 - 16 2023年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In this paper, we propose two methods to evaluate the facial symmetry from a video image of 3D point clouds. One is the analysis based on the movement of face landmarks. First, we perform the face landmark matching using the FPFH value. It describes the local geometric feature, and hence landmarks in different frames can be matched by comparing FPFH values. After the matching process, we evaluate the facial symmetry from the left-right difference of the temporal changes in landmarks. The other is the analysis based on the center of gravity. We extract the lip region from a point cloud and divide it into subregions. The positions of the center of gravity are calculated for all the subregions, and we evaluate the facial symmetry from the left-right difference of them. As an experimental result, it is shown that the proposal method improves the matching accuracy and evaluates the symmetry like our sense.

DOI： 10.1145/3637732.3637771

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85187678842&origin=inward

Noise2Noiseを用いた低線量CT画像のデノイジング法

澤田，神谷，村上, 李（天津工大）

北九州医工学術者協会誌 ( 北九州医工学術者協会誌 ) 63 5 - 8 2023年11月

改良型U-Netを用いた胸部CT画像からの肺腫瘍領域の自動識別

高橋，神谷，寺澤, 青木

北九州医工学術者協会誌 63 1 - 4 2023年11月

Pose Estimation of Point Sets Using Residual MLP in Intelligent Transportation Infrastructure 査読有り国際誌

Yujie Li, Zhiyun Yin, Yuchao Zheng, Huimin Lu, Tohru Kamiya, Yoshihisa Nakatoh, Serikawa Seiichi

IEEE Intelligent Transportation Systems Transactions 24 ( 11 ) 13359 - 13369 2023年11月

Symmetry Analysis of Face from A Video Image of 3D Point Cloud 査読有り国際誌

Narumi Kihara, Tohru Kamiya, Namiko Kimura-Nomoto, Takako Okawachi, Guangxu Li and Norifumi Nakamura

The 8th International Symposium on Artificial Intelligence and Robotics 2023年10月

Noise-robust Grasp Position Estimation Using PointNet 査読有り国際誌

Ryoto Fujikame, Yuchao Zheng, Shin Ishiyama, Huimin Lu, Kamiya Tohru, Yoshihisa Nakatoh, Seiichi Serikawa

The 8th International Symposium on Artificial Intelligence and Robotics 2023年10月

Extraction of Lung Tumor Regions from Thoracic CT Images Using an Improved U-Net 査読有り国際誌

Takashi, Kamiya, Terasawa, Aoki

Proc. of the 23nd International Conference on Control, Automation and Systems 1489 - 1493 2023年10月

Identification of Root Resorption on Panoramic X-ray Image Based on EfficientNet 査読有り国際誌

Sakata, Kamiya, Oda, Morimoto

Proc. of the 23nd International Conference on Control, Automation and Systems 1749 - 1752 2023年10月

Remote Sensing Image Registration Based on Improved Geometric-matching CNN 査読有り国際誌

Morishima, Lu, Kamiya

Proc. of the 23nd International Conference on Control, Automation and Systems 1745 - 1748 2023年10月

A Detection Method for Nodular Shadows from Temporal Subtraction Images Using A Machine Learning Technique Incorporated Radiomics Features 査読有り国際誌

Baba, Kamiya, Terasawa, Aoki, Kido

Proc. of the 23nd International Conference on Control, Automation and Systems 1741 - 1744 2023年10月

An Denoising Method Low-Dose CT Image Using Image Restoration Convolutional Neural Network 査読有り国際誌

Sadamatsu, Murakami, Li, Kamiya

Proc. of the 23nd International Conference on Control, Automation and Systems 1737 - 1740 2023年10月

Classification of Driver Gene Mutations from 3D-CT Images Based on Radiomics Features 査読有り国際誌

Watanabe, Kamiya, Terasawa, Aoki

Proc. of the 23nd International Conference on Control, Automation and Systems 1733 - 1736 2023年10月

Identification of Nodular Shadows from CT Images Using Improved CoAtNet Incorporated Clinical Recording 査読有り国際誌

Nishitaki, Kamiya, Kido

Proc. of the 23nd International Conference on Control, Automation and Systems 1727 - 1732 2023年10月

Facial Symmetry Analysis for Cleft Lip Patients from 4D Point Cloud Data 査読有り国際誌

Kihara, Kimura-Nomoto, Okawachi, Li, Nakamura, Kamiya

Proc. of the 23nd International Conference on Control, Automation and Systems 1502 - 1505 2023年10月

Automatic Segmentation of Finger Bone Regions from CR Image Using the Improved HRNet 査読有り国際誌

Hiraoka, Kamiya, Aoki

Proc. of the 23nd International Conference on Control, Automation and Systems 1498 - 1501 2023年10月

A Method for Denoising from Low-Dose CT images Based on Noise2Noise 査読有り国際誌

Sawada, Murakami, Li, Kamiya

Proc. of the 23nd International Conference on Control, Automation and Systems 1494 - 1497 2023年10月

Classification of histological types of primary lung cancer from CT images using clinical information 査読有り国際誌

Honda, Kamiya, Kido

Proc. of the 23nd International Conference on Control, Automation and Systems 1753 - 1757 2023年10月

An Image Registration Technique for Brain MR Images Using Linear Transform by 3DCNN 査読有り国際誌

Baba, Kamiya

Proc. of the 23nd International Conference on Control, Automation and Systems 1762 - 1765 2023年10月

Recognition of Specific Parts of Plastic Bottles Using Improved DeepLab v3+ 査読有り国際誌

Ideta, Kamiya

Proc. of the 23nd International Conference on Control, Automation and Systems 1758 - 1761 2023年10月

FPA LRSE U-Netを用いたCT画像からの脊椎領域の自動抽出

木佐貫　航平，神谷　亨，寺澤　岳，青木

第42回日本医用画像工学会大会予稿集 72 - 73 2023年09月

AI技術とコンピュータ支援診断への応用招待有り査読有り

神谷

画像ラボ 3 ( 8 ) 29 - 34 2023年08月

3次元点群動画像を用いた顔の時系列変化の検出および非対称度評価

木原　成海，野元　菜美子，大河内　孝子，李　光旭, 中村　典史, 神谷:

第42回日本医用画像工学会大会予稿集 217 - 218 2023年07月

3DCNNを用いた脳MR画像の3次元位置合わせ

馬場　星太郎，神谷

第42回日本医用画像工学会大会予稿集 227 - 228 2023年07月

非侵襲的なドライバー遺伝子変異有無の識別のためのRadiomics特徴量に基づく2クラス分類

渡邊，神谷，寺澤，青木

第42回日本医用画像工学会大会予稿集 219 - 220 2023年07月

Res BCDU-SCB Net: A Lung CT Image Segmentation based on SC-BLSTM 査読有り国際誌

Wang T., Li G（天津工大）., Tang C（天津工大）., Kamiya T.

ACM International Conference Proceeding Series 42 - 46 2023年07月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Due to the unique physiological structure of human lung tissue, segmentation of the lung parenchymal regions is a very challenging task. Traditional image segmentation methods have certain limitations on practical applications. Therefore, lung tissue segmentation methods based on deep learning have become a research hotspot in recent years. Res BCDU-Net may overcome the problem of high false positives in other semi-automatic segmentation, but it still has the problem of low accuracy in edge segmentation, which is probably caused by the unclear input of a single CT image. Therefore, we propose to utilize three consecutive CT images as the input, which employs inter-slice context to restrain the false positive. When a single input CT image is not clear, the accuracy of lung segmentation results can be improved by referring to its previous and next slices. Based on the input of three continuous images, the spatial information in the process of lung segmentation is added in this paper. We build a Res BCDU-SCB network and collect and integrate spatial information through SC-BLSTM module. An SC-attention module is designed to take both finer-grained spatial information and rich semantic information into account. In this paper, we append BLSTM (Bidirectional Convolutional Long Short-term Memory) module with SC-attention module, named SC-BLSTM, to decrease performance degradation caused by ambiguous boundaries and variable lung parenchymal shapes. The results of extensive experiments on lung CT images (LIDC-IDRI database) show that it may improve the final segmentation accuracy and reduce the false positives, which confirms the effectiveness of our proposed method.

DOI： 10.1145/3613307.3613316

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85175976936&origin=inward

Temporal subtraction technique for phalanges CR images using U-Net and Geometric-matching CNN 査読有り国際誌

Ono, Kamiya, Aoki

Intl. Journal of Computer Assisted Radiology and Surgery 2023 S84 - S85 2023年06月

Multidimensional Deformable Object Manipulation Based on DN-Transporter Networks 査読有り国際誌

Teng, Huimin Lu, Yujie Li, Tohru Kamiya, Yoshihisa Nakatoh, Serikawa Seiichi, Pengxiang Gao

IEEE Transactions on Intelligent Transportation Systems 24 ( 4 ) 4532 - 4540 2023年04月

Research on adaptive formation control of mobile robot based on improved virtual spring method 査読有り国際誌

Chen Y(天津工大)., Li B., Kamiya T., Shi X.

ACM International Conference Proceeding Series 775 - 780 2023年03月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

A robot formation control algorithm based on virtual spring model is proposed to solve the problem of path planning and formation keeping in unknown environment. By improving the design of target attraction, obstacle repulsion and interaction force between robots, the formation control is more secure and stable. Furthermore, the leader-follower formation transformation problem is studied. According to the objective function of minimum completion time, minimum total energy consumption and minimum total spring deformation, an adaptive formation transformation strategy is designed. This strategy can choose formation transformation mode adaptively according to the environment characteristics, such as adjust the parameters of the formation and reconfigure the formation, so that the robot formation can complete the task in a short time and energy consumption. Finally, the effectiveness of the proposed method is verified by simulation and real experiment.

DOI： 10.1145/3594315.3594404

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85168241761&origin=inward

Volumetric choroidal segmentation using 3D residual U-Net 査読有り国際誌

Li G(天津工大)., Wang K(天津工大)., Wang X(天津工大)., Sun B(天津工大)., Wang K(天津工大)., Gao Y(天津工大)., Sun S(天津工大)., Kamiya T., Dai Y(天津工大).

ACM International Conference Proceeding Series 145 - 149 2023年03月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Estimating dimensional measurements of the choroid provides diagnostic values which can be used to assess choroidal health. In this paper we describe a methodology of calculating measurements from choroid segmentations automatically generated using convolutional neural network (CNN). We use a three-dimensional (3D) U-Net architecture built from residual units to segment the choroid. A surface fitting phase is jointed to the main process to compensate segmented defects at the area of Optic Nerve Hypoplasia (ONH). Consequently, we process these segmentations to estimate the mean choroidal thickness(MCT). The model is evaluated on volumetric scans from 183 subjects, approximately half of which are thyroid eyes. In the choroidal layer segmentation experiment, the accuracy of the automatic segmentation algorithm proposed in this paper was 98.25% when comparing the manual segmentation results masked by doctors. It showed that the MCT in thyroid eyes were higher than those in normal eyes.

DOI： 10.1145/3594315.3594337

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85168249566&origin=inward

Research on Semantic segmentation co-evolutionary method for Automated Driving Haze Images 査読有り国際誌

Tang C（天津工大）., Wang Y., Kamiya T., Li Y.

ACM International Conference Proceeding Series 157 - 163 2023年03月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In this paper, we address the problem of semantic segmentation of advanced automatic driving image in hazy weather. We mainly adopt the co-evolution method of image defogging model and semantic segmentation model, the defogging model adopts generative adversarial GAN network, the semantic segmentation model adopts BiSeNet,and the training data is a public self-driving data set Cityscapes. The co-evolution of these two models is used to improve the accuracy of semantic segmentation of fog maps. The experimental results show that the semantic segmentation of fog map by co-evolution has a more obvious improvement effect, and to a certain extent, it can meet the target extraction and recognition of autonomous driving requirements.

DOI： 10.1145/3594315.3594339

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85168236599&origin=inward

Automatic Classification Method for Plastic Bottles and Caps Using Multi Attention Eff-UNet 査読有り国際誌

Moritsuka, Kamiya

International Conference on Artificial Lefe and Robotics 980 - 93 2023年02月

CNNを用いた磁粉探傷画像の分類法査読有り

森塚、神谷

バイオメディカル・ファジィ・システム学会誌 24 ( 2 ) 35 - 42 2023年02月

Atomatic Identification of Tumor Cells for Circulating Tumor Cells by Convolutional Neural Netwokrs 査読有り国際誌

Hashimoto, Kamiya, Li（天津工大）, Yoneda, Tanaka

International Journal of Innovative Computing, Information and Control 19 ( 1 ) 1 - 14 2023年02月

AI技術の医用画像解析への応用招待有り査読有り

神谷

計測と制御 62 ( 2 ) 95 - 97 2023年02月

AUTOMATIC IDENTIFICATION OF TUMOR CELLS FOR CIRCULATING TUMOR CELLS BY CONVOLUTIONAL NEURAL NETWORKS 査読有り国際誌

Hashimoto K., Kamiya T., Li G（天津工大）., Yoneda K., Tanaka F.

ICIC Express Letters 19 ( 1 ) 1 - 14 2023年02月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

Liquid biopsy allows non-invasive collection of circulating tumor cells (CTCs) in blood without the need for sampling and can clearly demonstrate their presence in many types of cancer. In this study, we propose a method to automatically identify CTCs from fluorescence microscopy images and enable quantitative analysis based on convolutional neural networks (CNN). In this paper, a cell nucleus region cropping algorithm is applied in addition to a filtering process centered on a selective enhancement filter. Next, identification by SqueezeNet is performed. We performed the proposed method to 5,040 images of 6 samples and conducted experiments to identify CTCs. The number of detected CTCs was 148 (TPR = 100%), and the number of over-detected non-CTCs was 925. For the identification, TPR = 88.51% and FPR = 5.102% for the CNN model using SqueezeNet. The proposed method successfully reduced the number of detections by about 71.4% without missing any correct answers, but the proposed method did not show good results in any of the evaluation metrics.

DOI： 10.24507/ijicic.19.01.1

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85148695370&origin=inward

深層学習による経時的差分像からの結節状陰影の検出

神谷，芳野，寺澤，青木

第１４回呼吸機能イメージング研究会学術集会 70 - 70 2023年01月

Automatic Classification Method for Plastic Bottles and Caps Using Multi Attention Eff-UNet 査読有り国際誌

Moritsuka S., Kamiya T.

Proceedings of International Conference on Artificial Life and Robotics 980 - 983 2023年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In Japan, increasing amounts of waste are becoming a social problem. One of approaches to solve the problem is recycle of the plastic bottles. However, they are thrown away with their caps still attached, and it should be removed by hand. To solve this problem, we developed a method for automatic identification of plastic bottles and caps using deep learning technique. In this paper, we propose a method that combines different numbers of Efficient blocks and adds an attention structure and verify its usefulness through experiments.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85149463467&origin=inward

Squeeze-and-Excitation Block Based Mask R-CNN for Object Instance Segmentation 査読有り

Nagasawa K., Ishiyama S., Lu H., Kamiya T., Nakatoh Y., Serikawa S., Li Y.

Communications in Computer and Information Science 1732 CCIS 56 - 64 2023年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Deep learning-based methods have taken center stage in image recognition, such as AlexNet and deep learning-based method. At present, Image recognition based on deep learning has been widely used in agriculture, factory automation, automated driving, medical fields and so on. In the fields of automated driving and medical care, the accuracy of the image recognition directly affects human lives. For these reasons, the importance of improving the accuracy of image recognition is clear. In this paper, we focus on instance segmentation tasks. The method used is Mask R-CNN, which is the basis of current state-of-the-art methods. The network structure based on ResNet, and we tried to improve the accuracy by adding Squeeze-and-Excitation Block (SE Block). According to the result of experiments, it is proved that this method has certain advantages for object instance segmentation.

DOI： 10.1007/978-981-99-2789-0_5

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85163323022&origin=inward

Pose Estimation of Point Sets Using Residual MLP in Intelligent Transportation Infrastructure 査読有り国際誌

Li Y., Yin Z., Zheng Y., Lu H., Kamiya T., Nakatoh Y., Serikawa S.

IEEE Transactions on Intelligent Transportation Systems 2023年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

6D pose estimation of arbitrary objects is a crucial topic for intelligent transportation infrastructure measurement. However, some external environmental factors and the characteristics of the object itself impact the accuracy of the object’s pose estimation in practical applications. In this paper, we propose a new multi-class dataset ICD-4 (Industrial car Components Dataset) for 6D object pose estimation, which mainly includes four component categories, and every category takes 20,000 different scenarios. ICD-4 dataset delivers quite a few research challenges involving the range of object pose transformations and has significant research value for small-scale pose estimation tasks. We also propose an innovative method PoseMLP, a pose estimation network that uses residual MLP (multilayer perceptron) modules to predict the 6D pose estimation directly. Simultaneously, the experimental results demonstrate the effectiveness and reliability of the proposed method.

DOI： 10.1109/TITS.2023.3250604

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85153361050&origin=inward

Preface for ACPR 2023 Proceedings 査読有り国際誌

Liu C.L., Yagi Y., Kamiya T., Blumenstein M., Lu H., Yang W., Cho S.B.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 14408 LNCS v - vi 2023年01月

Remote Sensing Image Registration Based on Improved Geometric-Matching CNN 査読有り国際誌

Morishima F., Lu H., Kamiya T.

International Conference on Control, Automation and Systems 1745 - 1748 2023年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Environmental change detection is one of the uses of satellite images. This process is performed by subtracting image pairs obtained by different time series or sensors. Therefore, image registration is an important pre-processing step in detection of environmental changes. Currently, image registration methods based on deep learning are gaining attention. In general, higher satellite image resolution results in more accurate registration. However, the increase in image size leads to higher computational costs during training and estimation of deep learning models. Then, we propose a method that reduces the number of parameters of the model to lower the computational cost while maintaining the accuracy. This method makes it easier to handle high-resolution satellite images. The proposed method modified the GMCNN (Geometric-matching CNN) architecture by adding CSA (Cosine Similarity Attention) and SE (Squeeze-and-Excitation) layers to enhance the feature map, and point-wise convolution to reduce the number of parameters. The improved GMCNN decreases the grid MSE by 0.0037 compared to the conventional GMCNN. It also reduces the number of parameters by 49.6%.

DOI： 10.23919/ICCAS59377.2023.10316818

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179178272&origin=inward

Recognition of Specific Parts of Plastic Bottles Using Improved DeepLab v3+ 査読有り国際誌

Ideta D., Kamiya T.

International Conference on Control, Automation and Systems 1758 - 1761 2023年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In this paper, we focus on the manpower shortage in factories and conduct an experiment to try to automate the process. Factory automation has many advantages, such as reducing labor costs and improving production efficiency. Among them, we focused on the sorting of plastic bottles at a waste disposal plant. Currently, garbage is sent to the landfill without being sorted, and it is sorted by hand. Unlike cans, plastic bottles cannot be easily automated because they cannot be separated using magnets. Therefore, we propose an image processing technique to recognize where plastic bottles are located, and to use a robot arm to sort them, thereby realizing automation. In this paper, we focus on image processing technique based on deep learning. As basic research, we conducted an experiment to see how well the robot can identify a single plastic bottle in an image. We attempted to use semantic segmentation methods for detection, using DeepLabv3+ as the basic model, and improved it. scSE Block and fine-tuning were introduced to improve the accuracy significantly compared to previous studies.

DOI： 10.23919/ICCAS59377.2023.10316854

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179179143&origin=inward

Identification of Root Resorption on Panoramic X-ray Image Based on EfficientNet 査読有り国際誌

Sakata K., Kamiya T., Oda M., Morimoto Y.

International Conference on Control, Automation and Systems 1749 - 1752 2023年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Root resorption is a disease in which the cells of the root of the tooth are destroyed. This disease may be harmless, depending on the cause. When the root resorption has progressed, early detection and treatment is necessary. If root resorption is left untreated, the lesion can progress to pulp exposure, necrosis, sinus and periodontal defects, and perforation. Diagnosis of root resorption is made by clinical examination and a panoramic x-ray machine, however accurate diagnosis requires computed tomography (CT). CT allows full-plane examination of the area of interest and provides an accurate diagnosis. However, CT is less widely used than panoramic radiography in private dental office. Therefore, there is a need for a method to detect root resorption from panoramic radiographs with a high reliability comparable to that of CT images. In this paper, we propose an image analysis method to identify the presence or absence of root resorption from panoramic radiographs. In particular, we propose a method that uses EfficientNet to extract features with fine-tuning and learns them using L2-softmax loss as an error function. We applied the proposed method to panoramic X-ray images of 59 actual cases and obtained TPR: 71% and FPR: 43% respectively.

DOI： 10.23919/ICCAS59377.2023.10316940

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179183951&origin=inward

Identification of Nodular Shadows from CT Images Using Improved CoAtNet Incorporated Clinical Recording 査読有り国際誌

Nishitaki Y., Kamiya T., Kido S.

International Conference on Control, Automation and Systems 1727 - 1732 2023年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

When lung cancer is suspected, a CT scan of the chest is widely used as a method of precise examination. However, the number of CT images obtained in a single examination is enormous, placing a heavy burden on the physician who reads the images. In addition, there is concern about the possibility of undetected lesions due to differences in the skills and experience of the reader. Therefore, computer-aided diagnosis systems have been introduced to reduce their workload and undetected lesions. When physicians make a diagnosis, they consider not only CT images but also information about the patient. Therefore, attempts are being made to improve the accuracy of diagnosis by mimicking this process using artificial intelligence. In this paper, we propose a model for identifying nodular shadows using deep learning, aiming to improve the accuracy of diagnosis by introducing medical record information in addition to image information. Normal tissue is divided into three classes: branched vessels, thin vessels, and round vessels, and a total of four classes are classified, including abnormal tissue. Experimental results show that the accuracy of nodular shade discrimination is improved when medical record information is added.

DOI： 10.23919/ICCAS59377.2023.10316787

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179182295&origin=inward

Facial Symmetry Analysis for Cleft Lip Patients from 4D Point Cloud Data 査読有り国際誌

Kihara N., Kimura-Nomoto N., Okawachi T., Li G（天津工大）., Nakamura N., Kamiya T.

International Conference on Control, Automation and Systems 1502 - 1505 2023年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Cleft lip is one of the most common birth defects. Several operations are performed to form a natural lip. A problem with these operations is that the criteria for the facial symmetry are unclear. Based on this background, we propose a method for evaluating the facial symmetry using 4D point cloud data. We evaluate the facial symmetry in two ways. One is based on the temporal changes in the face landmarks. Corresponding points are searched using two local geometric descriptors. The results are approximated to the 3D lines and these slopes are compared to obtain the left-right difference of the movement. The other is based on the center of gravity. The lip part is extracted from the point cloud and divided into subregions. The centroid coordinates are then calculated for each subregion and compared to obtain the left-right difference of the facial structure. The experiment is performed using artificially generated 4D data. As an experimental result, it is shown that our method can find point correspondences with smaller error than comparative methods.

DOI： 10.23919/ICCAS59377.2023.10316967

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179178953&origin=inward

Extraction of Lung Tumor Regions from Thoracic CT Images Using An Improved U-Net 査読有り国際誌

Takahashi R., Kamiya T., Terasawa T., Aoki T.

International Conference on Control, Automation and Systems 1489 - 1493 2023年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Lung cancer currently kills the largest number of people in the world, with lung cancer accounting for the highest number of deaths among men and women. Therefore, early detection and treatment are important issues to reduce the number of lung cancer deaths. Genetic testing can confirm the presence or absence of driver gene mutations involved in cancer cell growth and other factors. In the process of identifying the presence or absence of driver gene mutations, lung cancer regions are extracted manually in cooperation with a radiologist. One of the concerns is that manual lung cancer extraction places a heavy burden on the physician who reads the images. Therefore, we propose an automatic extraction of lung tumor regions from chest CT images using deep learning, with the goal of developing a CAD system that reduces the workload of the physician and prevents lesions from being overlooked. We constructed a new model based on U-Net with the addition of CBAM and MultiRes Block. Experimental results on the model using CT images of the chest showed a 4.2% improvement in accuracy for Dice and a 4.6% improvement for IoU compared to the original U-Net model.

DOI： 10.23919/ICCAS59377.2023.10316783

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179183296&origin=inward

Classification of Histological Types of Primary Lung Cancer from CT Images Using Clinical Information 査読有り国際誌

Honda N., Kamiya T., Kido S.

International Conference on Control, Automation and Systems 1753 - 1757 2023年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Identification of primary lung cancer is very important because it influences the course of treatment, especially for small cell carcinomas, which metastasize rapidly and must be detected at an early stage. In addition to imaging, clinical information is often used in CAD (computer aided diagnosis) systems. In addition to images, clinical information is often used in CAD systems, especially information on smoking history, which is considered to be important in the diagnosis of lung cancer. In this paper, we propose a method to identify primary lung cancer by adding clinical information from medical records in addition to images in order to improve the accuracy of diagnosis. We use tumor images surrounded by rectangular regions from CT images in an open dataset as input images and train the method by deep learning. We evaluate the proposed method by discriminating tumors from unknown data. In our experiments, we found that the accuracy was improved by about 5% when clinical information was added to 655 images, which included four classes of cancer: adenocarcinoma, small cell carcinoma, squamous cell carcinoma, and large cell carcinoma.

DOI： 10.23919/ICCAS59377.2023.10316865

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179178408&origin=inward

Classification of Driver Gene Mutations from 3D-CT Images Based on Radiomics Features 査読有り国際誌

Watanabe S., Kamiya T., Terasawa T., Aoki T.

International Conference on Control, Automation and Systems 1733 - 1736 2023年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Cancer is a leading cause of death worldwide, accounting for nearly 10 million deaths in 2020. Lung cancer is the most common cause of death from this cancer. Diagnosis is made mainly by biopsy, which checks for mutations in the driver genes of lung cancer. If there is a mutation, molecularly targeted drugs with higher therapeutic efficacy can be used. However, it is difficult for physicians to make a decision and places a heavy burden on patients. To solve this problem, a computer aided diagnosis (CAD) system is needed to identify the presence or absence of driver gene mutations from CT images. In this paper, Radiomics is used to extract features from 3D lung cancer in CT images and analyze them using machine learning to classify the presence or absence of mutations. Because the features obtained is huge, dimensionality reduction is performed using Null Importance. Furthermore, the accuracy is improved by adding gender as clinical information. The proposed method was applied to a dataset consisting of 175 cases. As a result, we obtained an AUC (Area Under the Curve) of 0.985, accuracy of 92.0%, true positive rate of 85.7%, and false positive rate of 2.20%.

DOI： 10.23919/ICCAS59377.2023.10316810

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179182718&origin=inward

Automatic Segmentation of Finger Bone Regions from CR Image Using the Improved HRNet 査読有り国際誌

Hiraoka T., Kamiya T., Aoki T.

International Conference on Control, Automation and Systems 1498 - 1501 2023年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In Japan, it is serious problems such as increasing nursing care of handicapped people requiring and the aging of caregivers which is caused in a super-aging society. One of the main causes of being classified as persons requiring nursing care include joint diseases such as rheumatoid arthritis and bone fractures due to osteoporosis. Early detection and treatment of these diseases are considered important. Rheumatoid arthritis and osteoporosis are generally diagnosed by simple X-ray examination. However, there are problems with radiographic diagnosis by physicians, such as lack of objectivity and reproducibility of diagnosis, and increased workload on the radiologists. To solve these problems, a Computer-Aided Diagnosis (CAD) system is being developed. Because the CAD system may use the results of quantitative computer analysis, it is expected to improve the reproducibility and accuracy of diagnosis and reduce the burden on physicians. Therefore, this paper proposes a segmentation method of phalanx region from CR images to develop a CAD system. The proposed method is based on HRNet + JPU + U-Net. The proposed method was applied to 101 cases of X-ray images, and mIoU=0.897 was obtained. Experiments for segmentation from images confirmed the usefulness of the proposed method by improving the extraction accuracy of the boundary of the phalanx region.

DOI： 10.23919/ICCAS59377.2023.10317032

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179183712&origin=inward

An Image Registration Technique for Brain MR Images Using Linear Transform by 3DCNN 査読有り国際誌

Baba S., Kamiya T.

International Conference on Control, Automation and Systems 1762 - 1765 2023年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In this paper, we focus on brain atrophy in Alzheimer's disease and propose a 3D image registration method of brain MR images for the purpose of calculating the atrophy rate by temporal subtraction technique. The nonlinear 3D image registration model of Stergios et al. is recombined into a linear image registration model, and image registration is performed with isotropic affine and rigid body transformations. In addition, the accuracy is improved by the Instance-specific Optimization, which performs the optimization process for each case. For the training of the model, we used pseudo-data obtained by arbitrarily linear transformation of the ADNI brain MR images. Synthetic data and actual ADNI time series data (AD, MCI) were used for the test. The experimental results show that the addition of the optimization process improves the normalized cross-correlation (NCC) by more than 0.1. The loss function of the model is compared between Grid loss and MSE, and it is confirmed that MSE is effective for NCC. Furthermore, the output images confirm the usefulness of the image registration of the skull in the linear transformation.

DOI： 10.23919/ICCAS59377.2023.10316755

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179178497&origin=inward

An Denoising Method Low-Dose CT Image Using Image Restoration Convolutional Neural Network 査読有り国際誌

Sadamatsu Y., Murakami S., Guangxu L（天津工大）., Kamiya T.

International Conference on Control, Automation and Systems 1737 - 1740 2023年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Radiation is widely used in medicine to diagnose and treat disease. CT (Computed Tomography) scans allow early detection of externally invisible diseases and appropriate treatment. However, radiation exposure from the examination may result in a future risk of cancer. Efforts are therefore being made to reduce radiation exposure. During the examination, noise is generated in the image when the dose is reduced. Noise reduces the visibility of the image and may cause lesions to be missed. In this study, we focus on Convolutional Neural Networks (CNNs), a type of deep learning model that has recorded high accuracy in image processing tasks. The proportion of frequency components in an image has more low-frequency components and fewer high-frequency components. Since image features are treated equally across channels, important information such as noise and edges are easily lost. To solve this problem, we propose CNN with channel attention module. In addition, we employ MAE as the loss function to enable effective learning. Using whole body slice CT images of pigs, we evaluate the image quality by Peak Signal-to-Noise Ratio (PSNR) and show that the proposed method is effective.

DOI： 10.23919/ICCAS59377.2023.10317050

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179178606&origin=inward

A Method for Denoising from Low-Dose CT Images Based on Noise2Noise 査読有り国際誌

Sawada S., Murakami S., Guangxu L., Kamiya T.

International Conference on Control, Automation and Systems 1494 - 1497 2023年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Japan has the highest number of CT scanners per million people among developed countries, and the annual medical radiation dose is about six times higher than the world average. Screening of high-risk populations by means of low-dose CT has been verified to be able to reduce lung cancer mortality. However, the low-dose CT image noise should be accurately estimated in order to achieve good image quality as normal dose CT. In this paper, we apply a denoising method using convolutional neural networks without use of the clean images during learning phase. The CNN model is based on SRResNet, which has achieved high performance in super-resolution tasks, and has two attention mechanisms, Spatial Attention and Channel Attention, in the Residual Block. Experiments using whole body slice CT images of pigs showed that the proposed method is useful by comparison with normal dose CT images and by evaluation of image quality using peak signal-to-noise ratio (PSNR).

DOI： 10.23919/ICCAS59377.2023.10317004

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179178164&origin=inward

A Detection Method for Nodular Shadows from Temporal Subtraction Images Using A Machine Learning Technique Incorporated Radiomics Features 査読有り国際誌

Baba N., Kamiya T., Terasawa T., Aoki T., Kido S.

International Conference on Control, Automation and Systems 1741 - 1744 2023年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Lung cancer progresses rapidly, and early detection and treatment are important. Computed tomography (CT) equipment is mainly used for the examination. However, the number of CT images is huge and places a heavy burden on physicians. Therefore, a computer aided diagnosis (CAD) system is necessary to overcome these problem. One of the CAD is temporal subtraction technique. The technique is that performs a subtraction operation using the current and past images of the same patient to remove normal structures and emphasize newly appearing areas that have changed over time. However, because two images taken at different dates are used, artifacts due to misalignment often occur. Therefore, it is necessary to classify partial images with high false-positive rates and enhanced temporal changes into lesions and normal tissue. In this paper, the temporal subtraction image generation method is used to extract candidate areas of abnormal shadows as initial shadows. The proposed method was applied to 22 cases of chest CT images, and the results showed that the proposed method achieved a discrimination accuracy of Accuracy=74.54, TPR=72.65, FPR=22.59, and AUC=79.16 in XGBoost, which is a kind of machine learning method.

DOI： 10.23919/ICCAS59377.2023.10316767

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85179180349&origin=inward

機械学習による胸部CT画像からのドライバー遺伝子情報変異有無の識別法

吉福，神谷，寺澤，青木

バイオメディカル・ファジィ・システム学会年次大会講演論文集 2022年12月

Recognition of sidewalk environment based on WideSegPlus 査読有り国際誌

Sakai, Lu, Yujie Li, Kamiya

Proceedings Volume 12508, International Symposium on Artificial Intelligence and Robotics 2022 ( SPIE ) 2022年12月

Grasping position estimation method using depth image for thin objects 査読有り国際誌

Yoshihara, Koga, Lu, Kamiya

Proceedings Volume 12508, International Symposium on Artificial Intelligence and Robotics 2022 ( SPIE ) 2022年12月

Underwater image super-resolution using improved SRCNN 査読有り国際誌

Horimachi, Lu, Zheng, Kamiya, Nakatoh, Serikawa

Proceedings Volume 12508, International Symposium on Artificial Intelligence and Robotics 2022 ( SPIE ) 2022年12月

Underwater video networking and targe tracking 査読有り国際誌

Ota, Lu, Jianru Li, Yuchao Zheng, Kamiya, Nakatoh, Serikawa

Proceedings Volume 12508, International Symposium on Artificial Intelligence and Robotics 2022 2022年12月

Automatic Classification of Respiratory Sound Considering Hierarchical Structure 査読有り国際誌

Tabata M., Lu H., Kamiya T., Mabu S., Kido S.

International Conference on Control, Automation and Systems 2022-November 537 - 541 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Respiratory diseases are one of the leading causes of death worldwide. Approximately 8 million people die annually from respiratory diseases. Diagnosis is made primarily by auscultation using a stethoscope. The lack of quantitative criteria makes diagnosis difficult in the field where physicians are in short supply. To solve this problem, a computer aided diagnosis (CAD) system that quantitatively analyzes and classifies respiratory sounds and outputs them as a 'second opinion' is needed. In this paper, HPSS (Harmonious/Percussive Sound Separation) is used to separate abnormal respiratory sound features. Images are generated from the spectral envelopes obtained by linear prediction coefficients (LPC) for each of the three types of respiratory sound data before separation. The CNN (convolutional neural networks) framework based on hierarchical structure of the correct labels is introduced. The proposed method was applied to the dataset used in the International Conference on Biomedical and Health Informatics (ICBHI) 2017 Challenge. As a result, we obtained a sensitivity of 63.5%, specificity of 85.1%, average score of 74.3%, harmonic score of 72.7%, area under the curve of 87.8%, and false negative rate of24.5%, respectively.

DOI： 10.23919/ICCAS55662.2022.10003771

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146537757&origin=inward

Recognition of Plastic Bottle Using Improved U-Net 査読有り国際誌

Ideta D., Kamiya T.

International Conference on Control, Automation and Systems 2022-November 200 - 203 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper focuses on separation rubbish, which is one of the causes of global warming. Currently, recyclable waste including garbage is sent to the waste disposal site in an unsorted state, and it is separated by human source. This is difficult work and a cause of manpower shortage. To overcome those problem, it is necessary to introduce automatic sorting algorithms on separation rubbish process. In this study, we focused on separation rubbish process especially plastic bottles based on image recognition technique using a deep learning approach. We implement an improved U-Net as the deep learning scheme to increase segmentation accuracy. As a result, we obtained the segmentation accuracy with 0.789 of cap part and 0.972 of body part on plastic bottle respectively.

DOI： 10.23919/ICCAS55662.2022.10003819

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146598114&origin=inward

Identification of abnormal tissue from CT images using improved ResNet34 査読有り国際誌

Honda N., Kamiya T., Kido S.

International Conference on Control, Automation and Systems 2022-November 532 - 536 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, CT examinations have been widely used as a screening method to detect lung cancer. However, reading enormous CT images become a heavy burden to the physician. To avoid this problem, computer-aided diagnosis systems have been introduced on CT screening. In general, physicians consider patient information in addition to image information when they make a diagnosis, new efforts are being made to improve the accuracy of diagnosis by mimicking this information with a machine. In this paper, we propose a method for identifying pulmonary nodules by adding medical record information to images to improve the accuracy of diagnosis. We classify nodules from unknown data by assigning branching information of vascular opacities, straight vascular shadows, and nodular shadows as labeled image, which are a cause of misrecognition based on image features in machine learning. In the experiment, the classification accuracy of the nodule class was improved by adding clinical information to 644 images including 161 nodal images.

DOI： 10.23919/ICCAS55662.2022.10003937

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146569950&origin=inward

Identification for Drinking Spout of Plastic Bottle Using Multi Eff-UNet 査読有り国際誌

Moritsuka S., Kamiya T.

International Conference on Control, Automation and Systems 2022-November 209 - 212 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Currently, the increase of general waste is a problem in Japan. It is garbage and human waste. One solution to this problem is to reuse the bodies of used plastic bottles to reduce the amount of plastic waste and the increase in general waste. For this purpose, some municipalities provide bags for disposing of plastic bottle bodies. However, in many cases, the bottles are not only discarded in these bags, but also with their caps attached. In such cases, the caps must be found and removed by hand in the sorting and processing facility at the waste treatment plant. This means that the same work is performed by person for a long time, causing problems such as overlooking caps due to fatigue. To solve these problems, we develop a method for automatic identification of plastic bottle bodies and caps using deep learning technique. In this paper, we propose a model that combines multiple Eff-UNets. Specifically, we combine EfficientNetB4 for local segmentation and EfficientNetB5 for global segmentation. By using our method, we conducted an experiment on images of plastic bottles collected from the internet and other sources. We obtained segmentation results of 97.9% for plastic bottle bodies and 86.0% for plastic bottle caps.

DOI： 10.23919/ICCAS55662.2022.10003920

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146540644&origin=inward

Generation of Super-Resolution Images from Satellite Images Based on Improved RCAN 査読有り国際誌

Morishima F., Lu H., Kamiya T.

International Conference on Control, Automation and Systems 2022-November 213 - 216 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Satellite images can be analyzed and used for a variety of purposes. In the future, satellite image analysis will become more important since the number of satellites launches, and the amount of satellite data increase every year. Under these circumstances, there are some problems to be solved. One is the existence of low-resolution satellite images. To analyze the lower resolution of satellite images there are some technical issues such as reduction of noise, misclassification of object recognition. Therefore, high-resolution images are necessary. However, high-resolution satellite images are expensive, and its images may not be available in the past satellite images. Super-resolution which is introduced in image processing is a method to solve these problems. Convolutional neural network (CNN)-based methods are effective, and there is a need for models that can perform super-resolution with higher accuracy. In this paper, we propose a method for super-resolving satellite images, based on the improved the RCAN (residual channel attention network) model with SRM (style-based recalibration module). The proposed method improves the PSNR (peak signal to noise ratio) by 0.0181 dB compared to the conventional RCAN model.

DOI： 10.23919/ICCAS55662.2022.10003856

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146557504&origin=inward

Extraction of Cervical Lymph Nodes using Improved U-Net++ 査読有り国際誌

Shime N., Kamiya T., Ishida T.

International Conference on Control, Automation and Systems 2022-November 562 - 565 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Early detection and treatment of the lymph node are important since swelling of the neck is a likely factor in systemic metastasis of cancer. One of the diagnoses of cervical swelling is a CT scan, which has a beneficial influence on the diagnosis of the disease. However, the reading of CT images is burdened by the large number of images, which increases the physician's workload. In addition, since it is based on the subjective judgment of the physician, there may be discrepancies in diagnostic results and undetected cases due to differences in experience. A means of solving these problems requires a CAD system that provides a second opinion to the physician. Therefore, this paper proposes a segmentation method of cervical lymph node region for the purpose of developing a CAD system for the diagnosis of cervical lymphadenopathy from CT images. The proposal method is a CNN model with U-Net++ as the backbone, introducing CBAM (convolution block attention module) and dual-branch multi-scale attention module. The proposed method was applied to CT images of 11 cases, yielding IoU of 62.29, confirming its validity.

DOI： 10.23919/ICCAS55662.2022.10003916

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146553025&origin=inward

Environment Recognition from Spherical Camera Images Based on Multi-Attention DeepLab 査読有り国際誌

Nishida Y., Guangxu L., Lu H., Kamiya T.

International Conference on Control, Automation and Systems 2022-November 204 - 208 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Electric wheelchair is an easy-to-operate means of transportation that does not require physical strength. With the number of electric wheelchair users increasing in recent years, the increase in traffic accidents becomes a problem. Therefore, by developing an autonomous electric wheelchair, it is expected that the risk of accidents will be reduced and the convenience of the electric wheelchair will be improved. Environment recognition is indispensable for the development of autonomous electric wheelchairs. We propose a semantic segmentation method for recognizing 16 objects in traffic environment. This paper examines the improvement of problems such as the high price of autonomous electric wheelchairs due to the increase in the number of sensors used, which has been a concern in related research. Therefore, we use panoramic images acquired by a spherical camera as input data, and extern the Multi-Attention Deep Lab algorithms fitting for the recognition of distorted images. A new CNN model is constructed using Deep Lab v3+, scSE Block, Pairwise Self-Attention, and Joint Pyramid Up-sampling. We conducted a recognition experiment using images taken on campus and verified its effectiveness. (Comparing to DeepLab v3+, IoU and Dice showed a 3.5% and 3.6% improvement in accuracy, respectively.)

DOI： 10.23919/ICCAS55662.2022.10003689

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146621100&origin=inward

Development of Temporal Subtraction Technique for Phalanges CR Image using Geometric-matching CNN 査読有り国際誌

Ono H., Kamiya T., Aoki T.

International Conference on Control, Automation and Systems 2022-November 558 - 561 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

We are developing a computer-aided diagnosis system for rheumatoid arthritis. X-rays images are widely used to diagnose the rheumatoid arthritis. However, it is difficult for physicians to read minute changes from the images. Therefore, we propose a method to visualize lesions in the phalangeal region by comparing past and current images of the same subject using a temporal subtraction technique. The proposed method consists of three steps: segmentation of phalanges, registration, and generation of subtraction images. First, the phalangeal region is extracted from the hand CR image using DeepLabv3+. Next, the past and current phalangeal region images are aligned by geometric-matching based on a CNN (convolutional neural networks) with instance-specific optimization. Finally, we apply the temporal subtraction technique to those images. We confirmed the effectiveness of the proposed registration method in an experiment using synthetic data. Also, the proposed method was applied to a pair of past and current image sets on same subject to generate a subtraction image. As a result, we confirmed that the proposed method can visualize changes between past and current images.

DOI： 10.23919/ICCAS55662.2022.10003828

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146633083&origin=inward

Detection of Driver Gene Mutations from Thoracic CT Images Based on LightGBM with Radiomics Features 査読有り国際誌

Watanabe S., Kamiya T., Terasawa T., Aoki T.

International Conference on Control, Automation and Systems 2022-November 542 - 545 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Lung cancer is one of the most common cancers worldwide and has become a general medical problem. To lessen the risk of death, early detection and treatment is particularly required. The patients can use molecularly targeted drugs when the driver gene mutations of the cancer are detected, but invasive biopsies are required. So, development of new methods to detect it noninvasively and in a short time are expected. we propose a new machine learning method for identifying the presence or absence of driver gene mutations of lung cancer on Thoracic CT Images that is a non-invasive, in a short time, and low-cost CAD (Computer Aided Diagnosis) system. In the proposed method, radiomics features are given as explanatory variables in addition to Thoracic CT Images, and supervised learning using LightGBM is performed to conduct binary classification with/without driver gene mutations.

DOI： 10.23919/ICCAS55662.2022.10003871

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146575218&origin=inward

Denoising on Low-Dose CT Image Using Deep CNN 査読有り国際誌

Sadamatsu Y., Murakami S., Li G., Kamiya T.

International Conference on Control, Automation and Systems 2022-November 546 - 549 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Computed Tomography (CT) scans are widely used in Japan, and they contribute to public health. On the other hand, there is also a risk of radiation exposure. To solve this problem, attempts are being made to reduce the radiation dose during imaging. However, reducing the radiation dose causes noise and degrades image quality. In this paper, we propose an image analysis method that efficiently removes noise by changing the activation function of Deep Convolutional Neural Network (Deep CNN). Experimental tests using full-body slice CT images of pigs and phantom CT images of lungs with Poisson noise show that the proposed method is helpful by comparing them with normal-dose CT images and evaluating image quality using peak signal-to-noise ratio (PSNR).

DOI： 10.23919/ICCAS55662.2022.10003847

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146610706&origin=inward

Classification the Root Resorption from Panoramic X-ray Image Using Center Loss Redefined in Angle Space 査読有り国際誌

Tamura K., Kamiya T., Oda M., Morimoto Y.

International Conference on Control, Automation and Systems 2022-November 570 - 573 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Root resorption is a pathological condition which is characterized by the loss of the tooth root. Root resorption is not painful in its early stages. As a result, many people who are potentially affected and the condition are often left untreated until it is detected during regular check-ups. If detected early, good treatment results can be achieved, whereas failure to treat the condition properly can lead to tooth extraction. However, the root resorption is currently difficult to detect on panoramic radiographs and may be treated as caries after it becomes painful. The aim of this paper is to identify root resorption from panoramic X-ray images using a deep metric learning algorithm. As a loss function for distance learning, it is known that the loss function in angle space is consistent. Therefore, a loss function is defined and trained using the cosine value of the angle between the feature and the center position to improve the discrimination performance. We obtained experimental results based on 150 image sets with 0.80 of accuracy, 0.62 of TPR, 0.19 of FPR and 0.78 of AUC, respectively.

DOI： 10.23919/ICCAS55662.2022.10003752

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146551688&origin=inward

Classification of CTC on Fluorescence Image Based on Improved AlexNet 査読有り国際誌

Kisanuki K., Guangxu L., Kamiya T.

International Conference on Control, Automation and Systems 2022-November 550 - 553 2022年11月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

These days, cancer has been the most primary cause of death in Japan. Cancer often progresses by repeating metastasis, so early detection and early treatment are important. Analysis of Circulating Tumor Cells (CTCs) has come to gather attention as a new biomarker that CTCs can detect primary cancer in human body. However, the number of CTCs in a billion blood cells is only a few, and detecting CTCs is very hard. Accordingly, we propose an automatic detection method of CTCs from fluorescence microscopy images to enable quantitative analysis by computer. This method consists of two parts. The first part, we use some series of filtering to the images and, new dividing method some overlapping nucleus then, from the images cut out the region of interest (ROI). The second part is distinguishing images by using CNN. We applied the proposed method to 5040 images of 6 samples. As a result, we obtained TPR:94.59%, FPR:6.544% by using AlexNet based model.

DOI： 10.23919/ICCAS55662.2022.10003905

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146616673&origin=inward

Automatic Detection of Cervical Lymph Nodes from Non-Contrast CT Images Using the SSD 査読有り国際誌

Baba, Ishida, Kamiya

International Conference on Control, Automation and Systems 566 - 569 2022年11月

Automatic Identification of CTCs in Fluorescence Microscope Images Using Morphological Filtering to Detect Cell Nuclei 査読有り国際誌

Hashimoto, Park, Ha, Jung, Kamiya

International Conference on Control, Automation and Systems 554 - 557 2022年11月

Multi Eff-UNet＋を用いたペットボトルとキャップの識別

森塚、神谷

第8回画像関連学会連合会秋季大会 11 - 12 2022年11月

距離学習とアンサンブル学習を併用した歯根吸収の判別法

田村晃聖，神谷亨，小田昌史*,森本泰宏

北九州医工学術者協会誌 ( 北九州医工学術者協会誌 ) 62 11 - 14 2022年11月

Geometric Matching CNNによる指骨CR画像の位置合わせとその評価査読有り

小野、神谷、青木

Medical Imaging Technology ( Medical Imaging Technology ) 40 ( 5 ) 226 - 232 2022年11月

Three-dimensional quantification of postoperative facial asymmetry in patients with unilateral cleft lip and palate using facial symmetry plane 査読有り国際誌

Shiigi, Okawachi, Kamiya, Hosoki, Nomoto, Ratman, Amir, Ishihata, Nakamura

The Cleft Palate Craniofacial Journal 2022年09月

3次元点群データを用いた口唇裂の対称性解析

細木大祐, 神谷亨, 野元菜美子, 椎木綾乃, 大河内孝子, 中村典史

第21回情報科学技術フォーラム 525 - 526 2022年09月

Multi-organ Statistical Shape Model Building Using a Non-rigid ICP Based Surface Registration 査読有り国際誌

Wu J., Li G., Kamiya T.

Journal of Image and Graphics(United Kingdom) 10 ( 3 ) 95 - 101 2022年09月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

Landmark correspondence is one of the key steps in Statistical Shape Model (SSM) building. In this paper, a non-rigid iterative closest point surfaces registration method is introduced to seek proper corresponded landmarks in the multi-organ surface meshes. Surfaces of four abdominal organs are used in the experiment to build SSM from landmarks corresponded by five different registration strategies. The proposed method of individually non-rigid registration of single organs shows the least errors measured by Hausdorff distance and the best model quality of generalization ability, specificity, and compactness.

DOI： 10.18178/joig.10.3.95-101

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85138705827&origin=inward

Temporal Subtraction Technique for Thoracic MDCT Based on Residual VoxelMorph 査読有り国際誌

Miyake N., Lu H., Kamiya T., Aoki T., Kido S.

Applied Sciences (Switzerland) 12 ( 17 ) 2022年09月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

The temporal subtraction technique is a useful tool for computer aided diagnosis (CAD) in visual screening. The technique subtracts the previous image set from the current one for the same subject to emphasize temporal changes and/or new abnormalities. However, it is difficult to obtain a clear subtraction image without subtraction image artifacts. VoxelMorph in deep learning is a useful method, as preparing large training datasets is difficult for medical image analysis, and the possibilities of incorrect learning, gradient loss, and overlearning are concerns. To overcome this problem, we propose a new method for generating temporal subtraction images of thoracic multi-detector row computed tomography (MDCT) images based on ResidualVoxelMorph, which introduces a residual block to VoxelMorph to enable flexible positioning at a low computational cost. Its high learning efficiency can be expected even with a limited training set by introducing residual blocks to VoxelMorph. We performed our method on 84 clinical images and evaluated based on three-fold cross-validation. The results showed that the proposed method reduced subtraction image artifacts on root mean square error (RMSE) by 11.3% (p < 0.01), and its effectiveness was verified. That is, the proposed temporal subtraction method for thoracic MDCT improves the observer’s performance.

DOI： 10.3390/app12178542

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85137919955&origin=inward

SEblockとDeformable Convolutionを加えたAlexNetによる胸部CT画像からの結節状陰影の検出

玉井，神谷，青木，木戸

2022年度電気学会電子・情報・システム部門大会 1262 - 1267 2022年08月

Geometric Ｍatching CNNによる指骨CR画像の位置合わせ手法

小野、神谷、青木

第41回日本医用画像工学会大会予稿集 84 - 85 2022年07月

臨床情報を加えた改良型 ResNet による CT 画像からの結節状陰影の識別

本田直也，神谷亨，木戸尚治

第41回日本医用画像工学会大会予稿集 176 - 177 2022年07月

LPCとHPSSを用いた呼吸音の自動分類

田端、陸、神谷、間普、木戸

第41回日本医用画像工学会大会予稿集 180 - 181 2022年07月

LightGBMにRadiomics特徴量を加味した胸部CT画像からのドライバー遺伝子変異有無の識別

渡邊、神谷、寺澤、青木

電子情報通信学会医用画像研究会、信学技報MI2022-11 61 - 66 2022年05月

Deep CNNを用いた低線量CT画像のデノイジング処理

貞松、村上、李、神谷

電子情報通信学会医用画像研究会、信学技報MI2022-11 67 - 70 2022年05月

Segmentation of Lung Nodules on CT Images Using a Nested Three-Dimensional Fully Convolutional Network 査読有り国際誌

Shoji Kido, Shunske Kidera, Yasushi Hirano, Shingo Mabu, Tohru Kamiya, Nobuyuki Tanaka, Yuki Suzuki, Masahiro Yanagawa, Noriyuki Tomiyama

Frontiers in Artificial Intelligence 5 2022年02月

記述言語：英語掲載種別：研究論文（学術雑誌）

In computer-aided diagnosis systems for lung cancer, segmentation of lung nodules is important for analyzing image features of lung nodules on computed tomography (CT) images and distinguishing malignant nodules from benign ones. However, it is difficult to accurately and robustly segment lung nodules attached to the chest wall or with ground-glass opacities using conventional image processing methods. Therefore, this study aimed to develop a method for robust and accurate three-dimensional (3D) segmentation of lung nodule regions using deep learning. In this study, a nested 3D fully connected convolutional network with residual unit structures was proposed, and designed a new loss function. Compared with annotated images obtained under the guidance of a radiologist, the Dice similarity coefficient (DS) and intersection over union (IoU) were 0.845 ± 0.008 and 0.738 ± 0.011, respectively, for 332 lung nodules (lung adenocarcinoma) obtained from 332 patients. On the other hand, for 3D U-Net and 3D SegNet, the DS was 0.822 ± 0.009 and 0.786 ± 0.011, respectively, and the IoU was 0.711 ± 0.011 and 0.660 ± 0.012, respectively. These results indicate that the proposed method is significantly superior to well-known deep learning models. Moreover, we compared the results obtained from the proposed method with those obtained from conventional image processing methods, watersheds, and graph cuts. The DS and IoU results for the watershed method were 0.628 ± 0.027 and 0.494 ± 0.025, respectively, and those for the graph cut method were 0.566 ± 0.025 and 0.414 ± 0.021, respectively. These results indicate that the proposed method is significantly superior to conventional image processing methods. The proposed method may be useful for accurate and robust segmentation of lung nodules to assist radiologists in the diagnosis of lung nodules such as lung adenocarcinoma on CT images.

DOI： 10.3389/frai.2022.782225

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85125736621&origin=inward

Automatic Segmentation of Finger Bone Regions from CR Images Using Improved U-Net and MSGVF Snakes 査読有り国際誌

Kawagoe, Murakami, Kamiya, Aoki

ICIC Express Letters Part B: Applications 13 ( 2 ) 155 - 160 2022年02月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

Bone disease including rheumatoid arthritis and osteoporosis is one of se-rious diseases especially elderly people. Diagnostic imaging using computed radiography (CR) imaging is an effective method for diagnosing the bone disease. However, accurate diagnosis should be made within a limited time, increasing the burden on doctors. There-fore, the goal of this paper is to develop a computer aided diagnosis (CAD) system for automatically diagnosing bone disease from CR images. The proposed method based on an improved U-Net and multiscale gradient vector flow snakes (MSGVF Snakes) extracts each phalange region from the CR image and creates an image for each phalange. In the experiment, the proposed method was applied to CR images of 101 cases, and the performance was evaluated. As a result, our proposed model based on improved U-Net for the segmentation obtained intersection over union (IoU) of 0.939, true positive (TP) of 0.984, and false positive (FP) of 0.068. The initial segmentation accuracy was higher than the conventional approach such as U-Net, Residual U-Net. Final segmentation performance of the phalanges based on MSGVF Snakes are TP of 0.945 and FP of 0.030.

DOI： 10.24507/icicelb.13.02.155

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85123095886&origin=inward

深層距離学習における誤差関数の改良によるパノラマエックス線画像からの歯根吸収の検出

田村、神谷、小田、森本

北九州医工学術者会議 2022年01月

画像処理技術による肺癌のコンピュータ支援診断招待有り

神谷

第13回呼吸機能イメージング研究会学術集会抄録集 27 2022年01月

Liver Segmentation in CT Images Using Deep-Learning and 3D CRF 査読有り国際誌

Nagano, Kamiya, Li

International Conference on Artificial Life and Robotics 540 - 53 2022年01月

Segmentation of liver tumors in multiphase computed tomography images using hybrid method 査読有り国際誌

Wu, Furuzuki, Li, Kamiya, Mabu, Tanabe, Ito, Kido

International Journal of Computer and Electrical Engineering 97 1 - 11 2022年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

The multiphase scan has enabled an improved detection of liver tumors. However, tumor regions and peripheral tissues are difficult to distinguish and delineate owing to their highly similar image features. Moreover, their characteristics vary significantly in different phases. This is challenging when using segmentation methods that are based on unique training models. Herein, a hybrid framework is proposed for liver tumor segmentation in multiphase images. We first develop a cascade region-based convolutional neural network with refined head to locate the tumors. Meanwhile, phase-sensitive noise filtering is introduced to refine the segmentation conducted by a level-set-based framework. This method is sensitive to the intensity contrast but not to the regions of interest, thereby affording better performance in delineating adjacent tumors. In our experiment, the average precision and recall rates are 76.8% and 84.4%, respectively. The intersection over union, true positive rate, and false positive rate are 72.7%, 76.2%, and 4.75%, respectively.

DOI： 10.1016/j.compeleceng.2021.107626

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85120725147&origin=inward

Underwater Video Networking and Target Tracking 査読有り国際誌

Ota H., Lu H., Li J., Zheng Y., Kamiya T., Nakatoh Y., Serikawa S.

Proceedings of SPIE - The International Society for Optical Engineering 12508 2022年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

IoT technology has made remarkable progress in recent years, and the world is full of IoT devices that continue to evolve every day. From smartphones, personal computers, and smartwatches to home appliances such as refrigerators and washing machines, and even indoor lights and house keys, IoT devices have become an inseparable part of our lives. In addition to devices used by individuals, IoT technology supports our daily lives from both front and back sides, such as IoT-enabled industrial equipment and satellite positioning systems. Japan has been making a national push to shift to IoT in industries that reduce the burden on workers and have recently been promoting a plan called Smart Agriculture, Forestry, and Fisheries. Among these three types of industries, the agricultural sector is slightly ahead of the others, with the Smart Agriculture Demonstration Project starting in 2019, and 182 districts in Japan are implementing the project by FY2021. The forestry and fisheries industries are also developing daily to become next-generation industries based on the program established in December 2019, although they are behind agriculture. However, the examples mentioned so far are those that have been promoted with the help of companies and the national government, although everyone has benefited from them. In this paper, we propose an IoT device that can be used as an IoT buoy by using a microcomputer, Raspberry Pi, to create a camera device that can distribute underwater images in real-time and track its location.

DOI： 10.1117/12.2655052

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146731793&origin=inward

Underwater Image Super-Resolution Using Improved SRCNN 査読有り国際誌

Horimachi R., Lu H., Zheng Y., Kamiya T., Nakatoh Y., Serikawa S.

Proceedings of SPIE - The International Society for Optical Engineering 12508 2022年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, industrialization and economic development in countries around the world have led to an ever-increasing demand for energy. Renewable energies are attracting attention, but they still often use mineral resources such as coal, petroleum, and natural gas, and onshore resources are depleting day by day. These energy and metal resources, such as copper, support Japan's industries and affluent lifestyle, and if Japan continues to rely on imports for most of these resources, it will become difficult for Japan to secure a stable supply of these energies and resources. Therefore, mining of mineral resources on the seafloor is essential to solve these problems, and research on seafloor resource surveys and mining is underway. Because direct human exploration and mining of seafloor resources are naturally dangerous, underwater robots are used to explore and mine seafloor resources. However, due to light absorption and turbidity in water, the underwater image of an underwater robot is sometimes less visible, making exploration unsatisfactory. Therefore, there is a need for higher-resolution underwater images of underwater robots. In this study, we perform super-resolution of underwater images using an improved SRCNN to support research on underwater images of underwater robots. The conventional SRCNN method uses the ReLU function as the activation function, but the improved SRCNN uses the PReLU function and FReLU function, which are extended activation functions of the ReLU function, to improve accuracy.

DOI： 10.1117/12.2655051

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146731084&origin=inward

Three-Dimensional Quantification of Postoperative Facial Asymmetry in Patients With Unilateral Cleft lip and Palate Using Facial Symmetry Plane 査読有り国際誌

Shiigi A., Okawachi T., Kamiya T., Hosoki D., Nomoto N., Ratman M.F., Amir M.S., Ishihata K., Nakamura N.

Cleft Palate-Craniofacial Journal 2022年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

Objective: To quantitatively assess three-dimensional (3D) soft tissue facial asymmetry in patients with unilateral cleft lip and palate (UCLP) who have undergone primary lip repair. Design: Clinical, retrospective, comparative, methodological study. Patients/Participants: Twenty patients with UCLP were selected after a review of the records. Inclusion Criteria: Complete UCLP; surgically treated without secondary repair. An age-matched and sex-matched Control group was employed. Main Outcome Measures: A 3D facial symmetry plane (FSP) was obtained by superimposing the point clouds of the original 3D facial image excluding the surgical site and including lip and nose areas and those of a mirrored facial image using the iterative closest point (ICP) adjustment method. The discrepancies in the depth and angle of the normal vector of the facial surface of each point cloud between right and left sides (cleft and non-cleft sides in the UCLP group, respectively) based on FSP were calculated. Results: Facial asymmetry in the UCLP group was significantly greater than in the Control group regarding both the discrepancies in the depth (1.34 ± 0.62, 0.73 ± 0.32 pixels, respectively) (P =.0004) and surface angle (18.0 ± 5.88, 12.8 ± 4.0°, respectively) (P =.0024). Biaxial assessment of the discrepancies in the depth and surface angle allowed us to visually extract UCLP patients with greater facial asymmetry. Conclusions: Facial asymmetry analysis based on 3D FSP effectively facilitates the facial asymmetry quantification and soft tissue surgical outcome evaluation in patients with UCLP.

DOI： 10.1177/10556656221123276

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85138316191&origin=inward

Multidimensional Deformable Object Manipulation Based on DN-Transporter Networks 査読有り国際誌

Teng Y., Lu H., Li Y., Kamiya T., Nakatoh Y., Serikawa S., Gao P.

IEEE Transactions on Intelligent Transportation Systems 24 ( 4 ) 4532 - 4540 2022年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

In the process of transportation, the handling and loading methods of rigid objects are becoming more and more perfect. However, whether in today's transportation system or in daily life, such as packing objects or sorting cables before transportation, the manipulation of deformable objects has been always inevitable and has attracted more and more attention. Due to the super degrees of freedom and the unpredictable physical state of deformed objects. It is difficult for robots to complete tasks under the environment of the deformable object. Therefore, we present a method based on imitation learning. In the generated expert demonstration, the agent is offered to learn the state sequence, and then imitate the expert's trajectory sequence which avoid the above-mentioned difficulties. In addition, compared with the baseline method, our proposed DN-Transporter Networks are more competitive in a simulation environment involving cloth, ropes or bags.

DOI： 10.1109/TITS.2022.3168303

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85129430885&origin=inward

Grasping Position Estimation Method Using Depth Image for Thin Objects 査読有り

Yoshihara T., Koga S., Lu H., Kamiya T.

Proceedings of SPIE - The International Society for Optical Engineering 12508 2022年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

The robot market in Japan is gradually expanding due to increasing demand. Industrial robots are being actively introduced in the manufacturing industry. The introduction of robots in the industry has three advantages: 1) securing labor, 2) increasing productivity, and 3) improving quality. The robot can run for a long time with constant work efficiency, thus achieving stable production. In addition, by replacing human labor, robots can reduce labor costs and reduce human error. The downside of introducing a robot is that the robot has to be told where to grab, which takes time, and a technician with specialized knowledge. Furthermore, this method cannot perform grasping when the grasping object is not in the specified position. However, the introduction of robot vision may solve these problems. In this study, by using depth images, processed images, and deep learning models, we aim to achieve object color-independent high-accuracy grasp position estimation for thin objects. We sharpen the depth image mainly by applying grayscale transformation and modify the deep learning model. The experimental results show that our design can achieve good results.

DOI： 10.1117/12.2663251

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146708731&origin=inward

Grasp Position Estimation from Depth Image Using Stacked Hourglass Network Structure 査読有り

Hamamoto K., Lu H., Li Y., Kamiya T., Nakatoh Y.O., Serikawa S.

Proceedings - 2022 IEEE 46th Annual Computers, Software, and Applications Conference, COMPSAC 2022 1188 - 1192 2022年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, robots have been used not only in factories. However, most robots currently used in such places can only perform the actions programmed to perform in a predefined space. For robots to become widespread in the future, not only in factories, distribution warehouses, and other places but also in homes and other environments where robots receive complex commands and their surroundings are constantly being updated, it is necessary to make robots intelligent. Therefore, this study proposed a deep learning grasp position estimation model using depth images to achieve intelligence in pick-and-place. This study used only depth images as the training data to build the deep learning model. Some previous studies have used RGB images and depth images. However, in this study, we used only depth images as training data because we expect the inference to be based on the object's shape, independent of the color information of the object. By performing inference based on the target object's shape, the deep learning model is expected to minimize the need for re-training when the target object package changes in the production line since it is not dependent on the RGB image. In this study, we propose a deep learning model that focuses on the stacked encoder-decoder structure of the Stacked Hourglass Network. We compared the proposed method with the baseline method in the same evaluation metrics and a real robot, which shows higher accuracy than other methods in previous studies.

DOI： 10.1109/COMPSAC54236.2022.00187

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85137004247&origin=inward

3D object detection using improved PointRCNN 査読有り

Fukitani K., Shin I., Lu H., Yang S., Kamiya T., Nakatoh Y., Serikawa S.

Cognitive Robotics 2 242 - 254 2022年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

Recently, two-dimensional object detection (2D object detection) has been introduced in numerous applications such as building exterior diagnosis, crime prevention and surveillance, and medical fields. However, the distance (depth) information is not enough for indoor robot navigation, robot grasping, autonomous running, and so on, with conventional object detection. Therefore, in order to improve the accuracy of 3D object detection, this paper proposes an improvement of Point RCNN, which is a segmentation-based method using RPNs and has performed well in 3D detection benchmarks on the KITTI dataset commonly used in recognition tasks for automatic driving. The proposed improvement is to improve the network in the first stage of generating 3D box candidates in order to solve the problem of frequent false positives. Specifically, we added a Squeeze and Excitation (SE) Block to the network of pointnet++ that performs feature extraction in the first stage and changed the activation function from ReLU to Mish. Experiments were conducted on the KITTI dataset, which is commonly used in research aimed at automated driving, and an accurate comparison was conducted using AP. The proposed method outperforms the conventional method by several percent on all three difficulty levels.

DOI： 10.1016/j.cogr.2022.12.001

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85144888388&origin=inward

Recognition of Sidewalk Environment Based on WideSegPlus 査読有り国際誌

Sakai Y., Lu H., Li Y., Kamiya T.

Proceedings of SPIE - The International Society for Optical Engineering 12508 2022年01月

担当区分：最終著者,　責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, the elderly population in Japan has been increasing. Expectations for the utilization of welfare equipment are also increasing. Electric wheelchairs are one of equipment and are widely used as a convenient means of transportation. On the other hand, accidents have also occurred, and dangers have been pointed out when driving the electric wheelchair. Therefore, we believe that the development of an autonomous mobile electric wheelchair can improve the causes of accidents. In addition, it can be expected to reduce accidents and improve the convenience of electric wheelchairs. For the development of an autonomous electric wheelchair, environment recognition such as estimation of the current position, recognition of sidewalks and traffic lights, and prediction of movement of objects is indispensable. To solve these problems, we develop an algorithm to recognize the sidewalks, crosswalks, and traffic lights from video images. In recent years, deep learning has been widely applied in the field of image recognition. Therefore, we improve WideSeg, one of the semantic segmentation algorithms that apply CNN (Convolutional Neural Networks), and develop an object recognition method using a new CNN model. In our approach, we perform adding the sidewalk correction and noise removal processing after performing semantic segmentation with the proposed model.

DOI： 10.1117/12.2655680

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85146673279&origin=inward

scSE-CRNNと3種類の呼吸音変換画像による呼吸音の分類査読有り

浅谷, 陸, 神谷, 間普, 木戸

医用画像情報学会雑誌 38 ( 4 ) 152 - 159 2021年12月

U-survival for prognostic prediction of disease progression and mortality of patients with COVID-19 査読有り

Näppi J.J., Uemura T., Watari C., Hironaka T., Kamiya T., Yoshida H.

Scientific Reports 11 ( 1 ) 2021年12月

記述言語：英語掲載種別：研究論文（学術雑誌）

The rapid increase of patients with coronavirus disease 2019 (COVID-19) has introduced major challenges to healthcare services worldwide. Therefore, fast and accurate clinical assessment of COVID-19 progression and mortality is vital for the management of COVID-19 patients. We developed an automated image-based survival prediction model, called U-survival, which combines deep learning of chest CT images with the established survival analysis methodology of an elastic-net Cox survival model. In an evaluation of 383 COVID-19 positive patients from two hospitals, the prognostic bootstrap prediction performance of U-survival was significantly higher (P < 0.0001) than those of existing laboratory and image-based reference predictors both for COVID-19 progression (maximum concordance index: 91.6% [95% confidence interval 91.5, 91.7]) and for mortality (88.7% [88.6, 88.9]), and the separation between the Kaplan–Meier survival curves of patients stratified into low- and high-risk groups was largest for U-survival (P < 3 × 10–14). The results indicate that U-survival can be used to provide automated and objective prognostic predictions for the management of COVID-19 patients.

DOI： 10.1038/s41598-021-88591-z

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85105193010&origin=inward

Detection of abnormal opacity on chest CT images using a 3D-CNN 査読有り国際誌

Koizumi, Kamiya, Aoki, Kido

International Conference on Biomedical and Bioinformatics Engineering 26 - 29 2021年11月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, the rate of deaths from cancer tends to increase in Japan. Especially the number of deaths from lung cancer is increasing. Computed tomography (CT) device is effective screening tool for early detection of lung cancer. Simultaneously, there is concern that an increase in burden on doctors will be caused by high performance of CT improving. Therefore, by presenting the "second opinion"by the computer aided diagnosis (CAD) system, it reduces the burden on the doctor. In this paper, we developed a CAD system for automatic detection of abnormal opacities such as lung nodules or ground glass opacity (GGO) on CT images from temporal subtraction image by incorporating a sparse coding technique to convolutional neural network (CNN). We extend the sparse coding and CNN to 3D and applied them to 32 cases which is obtained different time series on same subject. As a result, we obtained true positive rate (TPR) of 80.556%, false positive rate (FPR) of 22.892% and AUC of 0.768.

DOI： 10.1145/3502871.3502876

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85126727510&origin=inward

Wavelet Scattering TransformとCRNNによる呼吸音の分類

浅谷、陸、神谷、間普、木戸

電子情報通信学会医用画像研究会、信学技報 62 - 65 2021年11月

深層距離学習に基づく歯科パノラマX線画像からの歯根吸収の識別

田村、神谷、小田、田中、森本

電子情報通信学会医用画像研究会、信学技報 68 - 71 2021年11月

A generative flow-based model for volumetric data augmentation in 3D deep learning for computed tomographic colonography 査読有り国際誌

Tomoki Uemura, Janne J. Näppi, Yasuji Ryu, Chinatsu Watari, Tohru Kamiya, Hiroyuki Yoshida

International Journal of Computer Assisted Radiology and Surgery 2021年11月

Compare Derived U-Nets Using for Retinal Vessels Segmentation 査読有り国際誌

Li G., Wang X., Wang Z., Wu J., Kamiya T.

ACM International Conference Proceeding Series 1 - 8 2021年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

U-Net frameworks reveal potential on image segmentation tasks of the complex morphologic objects, such as capillaries. To compare the performance of vessels segmentation in fundus images, in this paper, we review U-Net and its three derived architectures: Residual Spatial Attention Network, Generative Adversarial Networks and IterNet. The networks training and testing were completed using the same image datasets under the same configurations. We calculated the accuracy and precision of segmentation results. The results showed that the cascade U-Net architecture provided better results especially on the capillaries parts.

DOI： 10.1145/3502803.3502804

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124455424&origin=inward

Shape Restoration by Shadow Information and Photometric Stereo 査読有り国際誌

Kameda, Lu, Kamiya, Serikawa

International Symposium on Artificial Intelligence and Robotics ( SPIE ) 2021年10月

Image Quality Improvement Using Local Adaptive Neighborhood-based Dark Channel Prior 査読有り

Onoyama, Lu, Soomro, Mokhtar, Kamiya, Serikawa

International Symposium on Artificial Intelligence and Robotics 2021年10月

Underwater Image Super-resolution Using SRCNN 査読有り国際誌

Ooyama, Lu, Kamiya, Serikawa

International Symposium on Artificial Intelligence and Robotics ( SPIE ) 2021年10月

Automatic Segmentation of Finger Bone Regions from CR Images Using Improved DeepLabv3+ 査読有り

Ono, Murakami, Kamiya, Aoki

International Conference on Control, Automation and Systems ( International Conference on Control, Automation and Systems ) 1788 - 1791 2021年10月

Incorporating Ghost Module into RCAN for Super-Resolution of Satellite Images 査読有り国際誌

Ikeda, Li, Kamiya

International Conference on Control, Automation and Systems 2059 - 2063 2021年10月

Determination of Abnormality of IGBT Images Using VGG16 査読有り

Ogawa, Watanabe, Omura, Kamiya

International Conference on Control, Automation and Systems 2055 - 2058 2021年10月

Automatic Identification of CTC in Fluorescence Microscope Images Using Segmentation Algorithm of Cell Nucleus 査読有り

Hashimoto, Kamiya, Yoneda, Tanaka

International Conference on Control, Automation and Systems 2051 - 2054 2021年10月

A Classification Method for Magnetic Particle Testing Image Using U-Net 査読有り国際誌

Moritsuka, Kamiya

International Conference on Control, Automation and Systems 2047 - 2050 2021年10月

Environment Recognition from A Spherical Camera Image Based on DeepLab v3+ 査読有り

Nishida, Li, Kamiya

International Conference on Control, Automation and Systems 2043 - 2046 2021年10月

Classification of Respiratory Sounds Improved CRNN 査読有り国際誌

Asatani, Kamiya, Mabu, Kido

International Conference on Control, Automation and Systems 1804 - 1808 2021年10月

Detection of the root resorption from panoramic X-ray images using deep metric learning 査読有り国際誌

Tamura, Kamiya, Oda, Tanaka, Moritomo

International Conference on Control, Automation and Systems 1800 - 1803 2021年10月

A Method for Evaluating of Asymmetry on Cleft Lip Using Symmetry Plane 査読有り国際誌

Sawada, Kamiya, Kimura-Nomoto, Okawachi, Nozoe, Nakamura

International Conference on Control, Automation and Systems 1796 - 1799 2021年10月

Extraction of Cervical Lymph Nodes Based on Three-Dimensional Image Registration 査読有り国際誌

Shime, Kamiya, Ishida

International Conference on Control, Automation and Systems ( International Conference on Control, Automation and Systems ) 1792 - 1795 2021年10月

Weakly unsupervised conditional generative adversarial network for image-based prognostic prediction for COVID-19 patients based on chest CT 査読有り

Uemura T., Näppi J.J., Watari C., Hironaka T., Kamiya T., Yoshida H.

Medical Image Analysis 73 2021年10月

記述言語：英語掲載種別：研究論文（学術雑誌）

Because of the rapid spread and wide range of the clinical manifestations of the coronavirus disease 2019 (COVID-19), fast and accurate estimation of the disease progression and mortality is vital for the management of the patients. Currently available image-based prognostic predictors for patients with COVID-19 are largely limited to semi-automated schemes with manually designed features and supervised learning, and the survival analysis is largely limited to logistic regression. We developed a weakly unsupervised conditional generative adversarial network, called pix2surv, which can be trained to estimate the time-to-event information for survival analysis directly from the chest computed tomography (CT) images of a patient. We show that the performance of pix2surv based on CT images significantly outperforms those of existing laboratory tests and image-based visual and quantitative predictors in estimating the disease progression and mortality of COVID-19 patients. Thus, pix2surv is a promising approach for performing image-based prognostic predictions.

DOI： 10.1016/j.media.2021.102159

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85111029043&origin=inward

Classification of respiratory sounds using improved convolutional recurrent neural network 査読有り

Asatani N., Kamiya T., Mabu S., Kido S.

Computers and Electrical Engineering 94 2021年09月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

Currently, auscultation using a stethoscope is performed for the diagnosis of respiratory diseases. Auscultation is a simple and non-invasive diagnostic method; however, the diagnostic results depend on the experience of the doctor, thereby rendering quantitative diagnosis difficult. Therefore, we herein propose a new automatic classification method based on deep learning algorithms for respiratory sounds to support the diagnosis of respiratory diseases. The proposed method comprises two stages. First, a spectrogram is generated by applying a short-time Fourier transform to the respiratory sound data. Subsequently, the obtained spectrogram is classified into normal and abnormal (three classes: crackle, wheeze, and both) respiratory sounds using an improved convolutional recurrent neural network. By classifying the respiratory sounds using the proposed method, the following results are obtained: sensitivity, 0.63; specificity,0.83; average score, 0.73; harmonic score, 0.72. Furthermore, the proposed method yields better accuracy compared with other methods.

DOI： 10.1016/j.compeleceng.2021.107367

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85112012928&origin=inward

Construction of a Hierarchical Feature Enhancement Network and Its Application in Fault Recognition 査読有り

Chen Z., Lu H., Tian S., Qiu J., Kamiya T., Serikawa S., Xu L.

IEEE Transactions on Industrial Informatics 17 ( 7 ) 4827 - 4836 2021年07月

記述言語：英語掲載種別：研究論文（学術雑誌）

Industrial Internet of Things (IIoT) provide significant support for observing and controlling industrial machinery. In this article, a novel hierarchical feature enhancement network (HFEN) is proposed by combining signal processing and representation learning. The signal processing block extracts features with definite physical significance. Then, the representability of the physical features is improved by connecting stacked denoising autoencoders and squeeze-and-excitation networks. A novel two-stream architecture is designed for HFEN to fuse two types of features. Consequently, HFEN can extract features that can be analyzed for physical significance and that are also representative in terms of recognizable patterns. The experimental results prove that the performance of HFEN is satisfactory in terms of accuracy and efficiency when compared to other methods. Finally, this article also aims to demonstrate the potential of a new pairing that fuses the model-and data-driven strategies for IIoT.

DOI： 10.1109/TII.2020.3021688

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85104187899&origin=inward

Computer-aided prediction of COVID-19 progression using unsupervised deep learning of chest CT images 査読有り国際誌

Uemura T, Näppi JJ, Watari C, Hironaka T, Kamiya T, Yoshida H.

Computer Assisted Radiology and Surgery 2021年06月

HPSSを用いた呼吸音の自動分類査読有り

丸橋,　浅谷, 陸, 神谷, 間普, 木戸

医用画像情報学会雑誌 38 ( 2 ) 95 - 10 2021年06月

Visual information processing for deep-sea visual monitoring system 査読有り国際誌

Ma, Li, Li, Tian, Wang, Kim, Serikawa

Cognitive Robotics ( Cognitive Robotics ) 1 3 - 11 2021年06月

記述言語：英語掲載種別：研究論文（学術雑誌）

Due to the rising demand for minerals and metals, various deep-sea mining systems have been developed for the detection of mines and mine-like objects on the seabed. However, many of them contain some issues due to the diffusion of dangerous substances and radioactive substances in water. Therefore, efficient and accurate visual monitoring is expected by introducing artificial intelligence. Most recent deep-sea mining machines have little intelligence in visual monitoring systems. Intelligent robotics, e.g., deep learning-based edge computing for deep-sea visual monitoring systems, have not yet been established. In this paper, we propose the concept of a learning-based deep-sea visual monitoring system and use testbeds to show the efficiency of the proposed system. For example, to sense the underwater environment in real time, a large quantity of observation data, including captured images, must be transmitted from the seafloor to the ship, but large-capacity underwater communication is difficult. In this paper, we propose using deep compressed learning for real-time communication. In addition, we propose the gradient generation adversarial network (GGAN) to recover the heavily destroyed underwater images. In the application layer, wavelet-aware superresolution is used to show high-resolution images. Therefore, the development of an intelligent remote control deep-sea mining system with good convenience using deep learning applicable to deep-sea mining is expected.

DOI： 10.1016/j.cogr.2020.12.002

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85109094624&origin=inward

Characteristics based visual servo for 6DOF robot arm control 査読有り国際誌

Tsuchida, Lu, Kamiya, Serikawa

Cognitive Robotics ( Cognitive Robotics ) 1 76 - 82 2021年06月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

Visual servo is a method for robot arm motion control. It is controlled by the end effector velocity that is the result in calculating the internal Jacobi matrix and vector of feature error. In general, automatic robotic task require high quality sensor that can measure a 3-dimential distance, and do calibration in order to suit the sensor frame and robot frame in Euclidean space. In this paper, we only use RGB camera as the data collection, which not requiring the calibration in sensor frame. Thus, our method is simpler than any other automatic motion methods. Meanwhile, the proposed characteristics based visual servo method has varying the hyper parameter, and show the effectiveness for indicating the precision of pose error both simulation and actual environments.

DOI： 10.1016/j.cogr.2021.06.002

DOI： 10.1016/j.cogr.2021.06.002

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85125125810&origin=inward

CT temporal subtraction: Techniques and clinical applications 招待有り査読有り

Aoki T., Kamiya T., Lu H., Terasawa T., Ueno M., Hayashida Y., Murakami S., Korogi Y.

Quantitative Imaging in Medicine and Surgery 11 ( 6 ) 2021年06月

Optimizing early cancer diagnosis and detection using a temporal subtraction technique 査読有り

Miyake N., Lu H., Kamiya T., Aoki T., Kido S.

Technological Forecasting and Social Change 167 2021年06月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

To optimize the early diagnosis and detection of lung cancer, computer-aided diagnostic (CAD) systems have been a useful tool for analyzing medical images. The temporal subtraction technique, which is a CAD system, performs the subtraction operation between the current image and the previous image on the same patient, and supports observation by emphasizing the temporal changes. However, the temporal subtraction technique for 3D images, such as thoracic CT images, has not yet been established. There is a need to develop efficient and highly accurate 3D nonrigid registration techniques to reduce subtraction artifacts. This study aims to develop a 3D nonrigid registration technique to establish a 3D temporal subtraction technique. In particular, we focus on the Finite Element Method, which is versatile, applicable to a wide range of fields, and capable of handling any shape. Our new method was examined on 46 clinical cases with multidetector row computed tomography images. As a result, the proposed method improved by 6.93% (p = 3.0 × 10−6) compared to the conventional methods in terms of the rate of reduction of artifacts, and the effectiveness was verified. Therefore, this study contributes to the literature on early detection and treatment.

DOI： 10.1016/j.techfore.2021.120745

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85102645602&origin=inward

WideSegNeXt: Semantic Image Segmentation Using Wide Residual Network and NeXt Dilated Unit 査読有り

Nakayama Y., Lu H., Li Y., Kamiya T.

IEEE Sensors Journal 21 ( 10 ) 11427 - 11434 2021年05月

記述言語：英語掲載種別：研究論文（学術雑誌）

Semantic segmentation is widely applied in autonomous driving, in robotic picking, and for medical purposes. Due to the breakthrough of deep learning in recent years, the fully convolutional network (FCN)-based method has become the de facto standard in semantic segmentation. However, the simple FCN has difficulty in capturing global context information, since the local receptive field is small. Furthermore, there is a problem of low image resolution because of the existence of the pooling layer. In this paper, we address the shortcomings of the FCN by proposing a new architecture called WideSegNeXt, which captures the image context on various spatial scales and is effective in identifying small objects. In addition, there is little loss of position information, since there are no pooling layers in the structure. The proposed method achieves a mean intersection over union (MIoU) of 72.5% and a global accuracy (GA) of 92.4% on the CamVid dataset and achieves higher performance than previous methods without additional input datasets.

DOI： 10.1109/JSEN.2020.3008908

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85104614498&origin=inward

U survival for prognostic prediction of disease progression and mortality of patients with COVID 19 査読有り国際誌

Janne J. Näppi, Tomoki Uemura, Chinatsu Watari, Toru Hironaka, Tohru Kamiya, Hiroyuki Yoshida

Nature Scientific Reports 11 ( 1 ) 2021年04月

HPSSを用いた呼吸音の自動分類

丸橋優生、神谷、間普、木戸

信学技報MI2020-77 128 - 133 2021年03月

近未来の医療～コンピュータ画像支援診断未来医療招待有り

神谷

純真の翼 ( 純真学園大学機関紙 ) 7 - 7 2021年03月

A supervoxel classification based method for multi-organ segmentation from abdominal ct images 査読有り

Wu J., Li G., Lu H., Kamiya T.

Journal of Image and Graphics(United Kingdom) 9 ( 1 ) 9 - 14 2021年03月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

Multi-organ segmentation is a critical step in Computer-Aided Diagnosis (CAD) system. We proposed a novel method for automatic abdominal multi-organ segmentation by introducing spatial information in the process of supervoxel classification. Supervoxels with boundaries adjacent to anatomical edges are separated from the image by using the Simple Linear Iterative Clustering (SLIC) from the images. Then a random forest classifier is built to predict the labels of the supervoxels according to their spatial and intensity features. Thirty abdominal CT images are used in the experiment of segmentation task for spleen, right kidney, left kidney, and liver region. The experiment result shows that the proposed method achieves a higher accuracy of segmentation compares to our previous model-based method.

DOI： 10.18178/joig.9.1.9-14

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85109040224&origin=inward

Detection of Abnormal Shadows in Low-dose CT Images Using CNN 査読有り

Ikeda, Kamiya, Aoki

International Conference on Artificial Life and Robotics 148 - 151 2021年01月

Deep-Learning Based Segmentation Algorithm for Defect Detection in Magnetic Particle Testing Images 査読有り

Ueda, Lu, Kamiya

International Conference on Artificial Life and Robotics 235 - 238 2021年01月

Image Registration Method for Chest MDCT Images Based on 2-D Finite Element Method 査読有り

Ogimoto, Kamiya, Aoki

International Conference on Artificial Life and Robotics 144 - 147 2021年01月

A generative flow-based model for volumetric data augmentation in 3D deep learning for computed tomographic colonography 査読有り

Uemura T., Näppi J.J., Ryu Y., Watari C., Kamiya T., Yoshida H.

International Journal of Computer Assisted Radiology and Surgery 16 ( 1 ) 81 - 89 2021年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

Purpose: Deep learning can be used for improving the performance of computer-aided detection (CADe) in various medical imaging tasks. However, in computed tomographic (CT) colonography, the performance is limited by the relatively small size and the variety of the available training datasets. Our purpose in this study was to develop and evaluate a flow-based generative model for performing 3D data augmentation of colorectal polyps for effective training of deep learning in CADe for CT colonography. Methods: We developed a 3D-convolutional neural network (3D CNN) based on a flow-based generative model (3D Glow) for generating synthetic volumes of interest (VOIs) that has characteristics similar to those of the VOIs of its training dataset. The 3D Glow was trained to generate synthetic VOIs of polyps by use of our clinical CT colonography case collection. The evaluation was performed by use of a human observer study with three observers and by use of a CADe-based polyp classification study with a 3D DenseNet. Results: The area-under-the-curve values of the receiver operating characteristic analysis of the three observers were not statistically significantly different in distinguishing between real polyps and synthetic polyps. When trained with data augmentation by 3D Glow, the 3D DenseNet yielded a statistically significantly higher polyp classification performance than when it was trained with alternative augmentation methods. Conclusion: The 3D Glow-generated synthetic polyps are visually indistinguishable from real colorectal polyps. Their application to data augmentation can substantially improve the performance of 3D CNNs in CADe for CT colonography. Thus, 3D Glow is a promising method for improving the performance of deep learning in CADe for CT colonography.

DOI： 10.1007/s11548-020-02275-z

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85095125050&origin=inward

Unsupervised survival prediction model from CT images of patients with COVID-19 査読有り

Uemura T., Näppi J.J., Watari C., Kamiya T., Yoshida H.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE 11597 2021年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

We developed an image-based unsupervised survival prediction model, called pix2surv, based on a conditional generative adversarial network (cGAN), and evaluated its performance based on chest CT images of patients with the coronavirus disease 2019 (COVID-19). The architecture of the pix2surv model includes a time generator that consists of an encoding convolutional network and a fully connected prediction network, and a discriminator network. The time generator is trained to generate survival-time images from chest CT images of each patient. The discriminator is a patch-based convolutional network that is trained to differentiate between "fake pairs"of a chest CT image and a generated survival-time image from "true pairs"of the chest CT image and the corresponding observed survival-time image of the patient. For evaluation, we retrospectively collected high-resolution chest CT images of COVID-19 patients. The survival predictions of the pix2surv model on these patients were compared with those of existing clinical prognostic biomarkers by use of a two-sided t-test with bootstrapping. Concordance index (C-index) and relative absolute error (RAE) were used as measures of the prediction performance. The bootstrap evaluation yielded C-index and RAE values of 80.4% and 15.6% for the pix2surv model, whereas those for the extent of the well-aerated lung parenchyma were 49.8% and 33.6%, and for a combination of blood tests of lactic dehydrogenase, lymphocyte, and C-reactive protein were 69.8% and 25.5%, respectively. The increase in survival prediction by the pix2surv model was statistically significant (p < 0.0001), indicating high effectiveness of the pix2surv model as a prognostic biomarker for the survival of patients with COVID-19.

DOI： 10.1117/12.2581929

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85103686657&origin=inward

U-radiomics for predicting survival of patients with COVID-19 査読有り

Uemura T., Näppi J., Watari C., Kamiya T., Yoshida H.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE 11601 2021年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

We developed and evaluated the effect of deep radiomic features, called U-radiomics, on the prediction of the overall survival of patients with the coronavirus disease 2019 (COVID-19). A U-net was trained on chest CT images of patients with interstitial lung diseases to classify lung regions of interest into five characteristic lung tissue patterns. The trained Unet was applied to the chest CT images of patients with COVID-19, and a U-radiomics vector for each patient was identified from the bottleneck layer of the U-net across all the axial CT images of the patient. The U-radiomics vector was subjected to a Cox proportional hazards model with an elastic-net penalty for predicting the survival of the patient. The evaluation was performed by use of bootstrapping, where the concordance index (C-index) was used as the comparative performance metric. Our preliminary comparative evaluation of existing prognostic biomarkers and the proposed U-survival model yielded the C-index values of (a) extent of well-Aerated lung parenchyma: 51%, (b) combination of blood tests of lactic dehydrogenase, lymphocyte, and C-reactive protein: 63%, and (c) U-survival: 87%. Thus, the U-survival significantly outperformed clinical biomarkers in predicting the survival of COVID-19 patients, indicating that the U-radiomics vector of the U-survival model may provide a highly accurate prognostic biomarker for patients with COVID-19.

DOI： 10.1117/12.2581907

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85103199185&origin=inward

Image Registration Method for Temporal Subtraction Based on Salient Region Features 査読有り

Sato S., Lu H., Kim H., Murakami S., Ueno M., Terasawa T., Aoki T.

EAI/Springer Innovations in Communication and Computing 13 - 20 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

While diagnosis imaging is an indispensable technology in medical field, it causes an increase in the burden to radiologists. In recent years, computer-aided diagnosis (CAD) system for supporting a radiologist has been developed to solve this problem. Temporal subtraction, which is one of CAD, is a technique to generate images emphasizing temporal changes in lesions and facilitates diagnosis of radiologists. To make a temporal subtraction image, image registration technique is required. In this chapter, we propose an image registration method for image registration of current image and previous image to generate temporal subtraction images in a short time. The proposed method consists of three steps: (i) segmentation of the region of interest (ROI) using position information of the spine based on anatomical information, (ii) using global image matching to select pairs of previous image and current image in which the same portion is depicted, and iii) final image matching based on salient region features (SRF). We performed our registration technique to the synthetic data and confirmed usefulness of the proposed method. The rotated synthesis image gives TP 100.0% and FP 12.16%. The synthesis image obtained by applying a Gaussian filter gives TP 70.40% and FP 0.00%. The synthesis image obtained by adding artificial pseudo-lesion region gives TP 99.45% and FP 17.89%. The synthesis image obtained by adding random noise of 5% gives TP 83.05% and FP 16.95%. Furthermore, radiologist conducted comparative experiments without and with temporal subtraction images created by proposed method. As a result, radiologist showed high reading performance by using temporal subtraction images.

DOI： 10.1007/978-3-030-46032-7_2

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85091327306&origin=inward

Extreme ROS Reality: A Representation Framework for Robots Using Image Dehazing and VR 査読有り

Ueda A., Lu H., Kim H.

EAI/Springer Innovations in Communication and Computing 21 - 32 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

It is necessary to work in response to various situations when using robots at disaster areas. So, the complete automation remote control technique of the robot is essential. In recent years, the remote control technique of the robot is mainly focused on recognizing the working environment of the robot from the 2D monitor and performing the operation. In order to operate the robot correctly, a corresponding technique and sufficiently trained engineers are required. To this end, a new remote control technology that can operate more easily is required. Also, depending on the disaster environment that working at night or indoors is assumed, it becomes difficult to recognize the work environment correctly from the image by the sensor such as the camera due to insufficient light source. In this chapter, in order to solve these problems, as a development of new remote control technology, we propose the method of reproducing the working environment and the image improvement method under the low light source environment. In this chapter, we reproduce the working environment of remote areas by using virtual reality. In addition, we improve the image under low light source environment using haze removal algorithm. The work environment of the robot is acquired by RGBD sensor. Image improvement processing is performed on the obtained color image, and the image data is transmitted to a remote place where the operator is located via WebSocket. The experiments show that our method reproduces the working environment of the robot as point group to the virtual reality space and examine its usefulness.

DOI： 10.1007/978-3-030-46032-7_3

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85091329616&origin=inward

An Adaptable Feature Synthesis for Camouflage 査読有り

Li G., Yu Z., Chang J., Kim H.

EAI/Springer Innovations in Communication and Computing 81 - 91 2021年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Military camouflage should be various in order to quickly adapt environmental changes of battlefield. The elusiveness is depressed if the constant camouflage patterns of the clothes and vehicles are used, as applying in the most of current military. In this paper, we propose a texture synthesis method from an image using convolutional neural networks, which has an effect on the camouflage generation on a 3D surface. We use the latest advances in style transitions in 2D images and map characteristic textures onto 3D surfaces according to geometric feature. This allows us to make an adaptive texture map of the 3D objects, even in the case of complex topologies.

DOI： 10.1007/978-3-030-46032-7_8

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85091331451&origin=inward

A Classification Method for Magnetic Particle Testing Image Using U-Net 査読有り

Moritsuka S., Kamiya T.

International Conference on Control, Automation and Systems 2021-October 2047 - 2050 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Magnetic Particle Testing (MPT) is a method for determining the presence or absence of a defect by magnetizing the object to be inspected and sprinkling magnetic powder, which is absorbed by the defective part such as a crack and appears as a magnetic powder pattern, which is then evaluated by a specialist. By using the MTP, inspection can be performed without breaking the object to be inspected. However, there are some problems such as the possibility of overlooking defects. In this paper, to solve the problems we develop a classification method of defect images by deep learning for the automation of MPT. The proposed method is based on the structure of U-Net, which has excellent segmentation capability in image processing, and performs segmentation using an improved model that adds convolutional layers to U-Net. Then, an algorithm that combines the result with the last part of the encoder of U-Net is used to discriminate the presence or absence of defects. Using this method, defects were classified from the images obtained during MPT. The results showed that Accuracy of 85.8%, TPR of 65.2%, and FPR of 13.8%.

DOI： 10.23919/ICCAS52745.2021.9650052

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124232188&origin=inward

Underwater image super-resolution using SRCNN 査読有り国際誌

Ooyama S., Lu H., Kamiya T., Serikawa S.

Proceedings of SPIE - The International Society for Optical Engineering 11884 2021年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, energy minerals have become more important due to the rapid industrialization worldwide. Due to the rapid industrialization on a global scale, there is a shortage of mineral resources, and there are more opportunities to rely on alternative energy sources. Therefore, the exploration of marine resources, which are abundant in the ocean, is being promoted. However, it is dangerous and impractical for humans to dive and search for marine resources by hand. Therefore, it is possible to proceed with underwater exploration safely by having a robot do the work instead. Robots have been used as a mainstream search tool in the underwater environment due to the existence of various hazardous environmental conditions. However, there are several problems associated with robot control in underwater environments, one of which is poor visibility in the water. One of the problems is the poor visibility in the water. To improve the visibility in the water, we are trying to increase the resolution of underwater images by using super-resolution technology. In this paper, we conduct experiments using SRCNN, which is a basic super-resolution technique for underwater images. In addition, we investigate the effectiveness of "Mish", which has been attracting attention in recent years for its potential to surpass the performance of "ReLU", although "ReLU"is a typical activation function of neural networks, on SRCNN.

DOI： 10.1117/12.2603761

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85120500372&origin=inward

Shape restoration by shadow information and photometric stereo 査読有り国際誌

Kameda S., Lu H., Kamiya T., Serikawa S.

Proceedings of SPIE - The International Society for Optical Engineering 11884 2021年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Virtual Reality (VR) systems are become popular in recent years, and the capture 3D objects from the real world have been studied. 3D objects have large and complex data. In this paper, we propose a novel method that use the shadow information and Photometric stereo with their surface from one point of view to recover the 3D shapes. The experimental results show that the proposed method performs well accuracy.

DOI： 10.1117/12.2604193

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85120506573&origin=inward

Image quality improvement using local adaptive neighborhood-based dark channel prior 査読有り国際誌

Onoyama T., Lu H., Soomro A.A., Mokhtar A.A., Kamiya T., Serikawa S.

Proceedings of SPIE - The International Society for Optical Engineering 11884 2021年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In-vehicle cameras and surveillance cameras are used in many situations in our daily lives. Visibility degradation in foggy environments is caused by the scattering of reflected light from real objects by minute water droplets or fog in the medium through which light passes. The degree of degradation depends on the density of suspended microparticles existing between the observed object and the observation point in the medium. In general, the farther the object is from the camera, the more it is affected by the fog. The purpose of image de-fogging is to improve the clarity of an object by removing the effects of fog in the image.

DOI： 10.1117/12.2603771

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85120496362&origin=inward

Incorporating Ghost Module into RCAN for Super-Resolution of Satellite Images 査読有り国際誌

Ikeda H., Li G., Kamiya T.

International Conference on Control, Automation and Systems 2021-October 2059 - 2063 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

With the explosion of amount of low cost satellites, satellite images have been widely used for many non-military applications, such as agriculture, landscape, and recognition of environment. Improving the image resolution to mine useful information becomes one of the immediate problems. Therefore, it is expected to improve the recognition accuracy by increasing the resolution of satellite images. Recently, deep learning technique has been proposed to increase the resolution of images. However it requires a large number of learning parameters, which results in huge computational cost. To overcome this problem, we develop a new deep learning model based on ghost module to reduce the parameters while maintaining the quality of results. We utilized Google Earth Pro satellite imagery for the network training and testing. Comparing to the classical convolutional neural network module based methods, the number of parameters used in our model was reduced 49.31 % but keeping the same level of Peak Signal-to-Noise Ratio (24.1578) and Structural Similarity (0.7174).

DOI： 10.23919/ICCAS52745.2021.9649982

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124232834&origin=inward

Extraction of Cervical Lymph Nodes Based on Three-Dimensional Image Registration 査読有り国際誌

Shime N., Kamiya T., Ishida T.

International Conference on Control, Automation and Systems 2021-October 1792 - 1795 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

One of the difficulties of cancer is the metastasis of cancer through the lymph nodes. Therefore, early detection and treatment of cancerous is important. Visual screening is one of useful tool for diagnosing of the cancer. However, the number of images obtained at a time is large, and the burden on the reading physician is increasing. Furthermore, reading of image is based on the subjective judgment of the physician, which may lead to different diagnostic results. To solve these problems, computer aided diagnosis (CAD) systems have been attracting attention in recent years. One of the CAD systems is the temporal subtraction image technology. In this paper, we propose an image alignment method for generating temporal subtraction images from images taken before and after contrast agent was administered to the cervical lymph nodes of the same subject. In addition, we propose an image analysis method that suppresses overextraction of lymph node candidate regions on the temporal subtraction image based on the features of lymph nodes. We applied the proposed method to the CT images of three sets and compared the temporal subtraction images with the final lymph node extraction images, and confirmed that the proposed method can suppress the overextraction of lymph node regions.

DOI： 10.23919/ICCAS52745.2021.9649797

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124183732&origin=inward

Environment Recognition from A Spherical Camera Image Based on DeepLab v3+ 査読有り国際誌

Nishida Y., Li Y., Kamiya T.

International Conference on Control, Automation and Systems 2021-October 2043 - 2046 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

The number of users of electric wheelchairs has been increasing in recent years because it is easy to operate the electric wheelchair and do not require physical strength. However, the traffic accidents are also increasing because of the large number of wheelchairs. The development of autonomous electric wheelchairs is expected to reduce the risk of accidents and improve the convenience of electric wheelchairs. Environmental recognition is essential for the development of autonomous electric wheelchairs. In this paper, we propose a method for recognizing roads, sidewalks, buildings, electric wheelchair drivers, poles, electric wheelchairs, vegetation, curbs, sky, pedestrians, lanes, cars, steps, and bicycles. For recognizing those objects, we use a panoramic image acquired from a spherical camera. As the machine techniques, we use DeepLab v3+, a semantic segmentation algorithm based on Convolutional Neural Network (CNN). In the proposed method, a new CNN model is constructed by adding deformable convolution, SE-block, and MobileNet v2 to DeepLab v3+ into the original DeepLab v3+. In the experiment, IoU 38.8% and Dice of 46.7% were obtained.

DOI： 10.23919/ICCAS52745.2021.9649846

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124192140&origin=inward

Determination of Abnormality of IGBT Images Using VGG16 査読有り国際誌

Ogawa T., Watanabe A., Omura I., Kamiya T.

International Conference on Control, Automation and Systems 2021-October 2055 - 2058 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

A power device is a semiconductor device for power control used for power conversion such as converting direct current to alternating current and alternating current to direct current. It is widely used such as refrigerators, air conditioners which is implemented electronic components that are closely related to our daily lives. Therefore, high reliability and safety are required, and power cycle tests are conducted for the purpose of evaluating them. In the conventional test, there is a problem that it is difficult to perform analysis because sparks are generated during the test and the device is severely damaged after the test. To solve this problem, a new technology has been developed that adds ultrasonic that enable internal observation during the test. However, there are remains a problem that the method for analyzing the ultrasonic image obtained in the new technology has not been established. Also, few abnormal images are obtained in the test. In this paper, we propose a method for detection of abnormal devices based on CNN. Especially, we implement a Cycle-GAN to extend the abnormal data and classify the known image based on improved VGG16. As an experimental result, classification accuracy of Precision = 97.06%, Recall = 93.58%, F-measure = 95.17% were obtained.

DOI： 10.23919/ICCAS52745.2021.9650029

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124191152&origin=inward

Detection of the root resorption from panoramic X-ray images using deep metric learning 査読有り国際誌

Tamura K., Kamiya T., Oda M., Tanaka T., Morimoto Y.

International Conference on Control, Automation and Systems 2021-October 1800 - 1803 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Root resorption is a pathological process characterized by the loss of tooth roots because of inflammation induced by bacterial infection, trauma, physical or chemical irritation. As a result, the development of periodontal disease, increased susceptibility to infection and crooked teeth. In the worst case, it can lead to tooth extraction. Root resorption is often caused by pressure during orthodontic treatment. The presence of root resorption should be checked regularly during orthodontic treatment, as it often occurs. It is necessary to check for root resorption periodically during orthodontic treatment. However, it is difficult to detect the root resorption using a panoramic radiograph. As a result, root resorption is often latent and goes undetected. In this paper, we propose an image analysis method based on deep learning technique for detecting the root resorption on panoramic radiograph. We incorporate the EfficientNet for feature extraction in deep learning to the center loss and triplet loss as the loss function for metric learning. Our proposed method performed to 337 images which is obtained by panoramic radiograph. Accuracy of 71 %, true positive rate of 77%, false positive rate of 30% were obtained.

DOI： 10.23919/ICCAS52745.2021.9649745

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124211898&origin=inward

Classification of Respiratory Sounds by Generated Image and Improved CRNN 査読有り国際誌

Asatani N., Kamiya T., Mabu S., Kido S.

International Conference on Control, Automation and Systems 2021-October 1804 - 1808 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

The death toll from respiratory illness reached nearly 8 million in 2019. Auscultation is used to diagnose for respiratory illness. Highly accurate diagnosis is required to reduce the number of deaths. However, unlike diagnostic imaging, auscultation of respiratory sounds could not visualize the diagnostic results. In addition, since there is a problem that the experience of a doctor affects the diagnosis results, it is required to develop a diagnostic system for quantitative analysis. In recent years, the development of a diagnostic system using the ICBHI 2017 Challenge Respiratory Sound Database has been carried out in the field of respiratory sound analysis. However, the proposed system still has accuracy problems. Therefore, in this study, we improve the proposed method by classifying the improved CRNN (Convolutional Recurrent Neural Network) by inputting multiple respiratory sound images. As a result, Sensitivity: 0.64, Specificity: 0.83, Average Score: 0.74, Harmonic Score: 0.72 were obtained, and excellent results were achieved compared with other methods.

DOI： 10.23919/ICCAS52745.2021.9649906

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124181434&origin=inward

Automatic Segmentation of Finger Bone Regions from CR Images Using Improved DeepLabv3+ 査読有り国際誌

Ono H., Murakami S., Kamiya T., Aoki T.

International Conference on Control, Automation and Systems 2021-October 1788 - 1791 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

The number of hospitalized patients and the number of people requiring nursing care are serious social problems in Japan due to the increasing elderly population. The major causes of bedridden patients are bone and joint disorders caused by rheumatoid arthritis and osteoporosis. Early detection and treatment of these bone diseases are important because they significantly interfere with the quality of life (QOL) as the symptoms progress. Visual screening based on CR is used as a diagnosing tool for bone diseases. However, imaging diagnosis is subjective and lacks objectivity, and there is a possibility that lesions may be overlooked. In addition, it is difficult to find out subtle changes from images, increasing the workload for doctors. To solve these problems, there is a need to develop a computer aided diagnosis (CAD) system that can quantitatively diagnose bone diseases. We propose a method for automatic extraction of phalange regions for the CAD system to diagnose these diseases. The proposed method can extract the phalanges with high accuracy by using the improved DeepLabv3+. In this paper, we apply the proposed method to 101 cases of CR images and mIoU of 0.949 was obtained.

DOI： 10.23919/ICCAS52745.2021.9649864

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124185622&origin=inward

Automatic Identification of CTC in Fluorescence Microscope Images Using Segmentation Algorithm of Cell Nucleus 査読有り国際誌

Hashimoto K., Kamiya T., Yoneda K., Tanaka F.

International Conference on Control, Automation and Systems 2021-October 2051 - 2054 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Today, cancer is the number one cause of death in Japan, and cancer accounts for 27.3% of all deaths number. The development of cancer by repeating metastasis, hence it is important to operate early detection and early treatment. The diagnosis of cancer includes various treatments, but it is difficult to judge whether cancer is metastatic or not. Then, analysis of Circulating Tumor Cells (CTCs) in blood has been attracting attention as a new biomarker. However, because the ratio of CTCs in a billion blood cells is only a few, and there is a concern that the burden on doctors will increase. We propose a method for automatic identification of CTCs from fluorescence microscopy images to enable quantitative analysis by computer for the diagnosis of CTCs in blood. First, after detecting the cell candidate regions mainly by filtering, we set the region of interest in the cell candidate regions and reconstruct the region of interest by cutting out the cell nucleus region. In this paper, we applied the proposed method to 5,040 images of 6 samples and conducted experiments on the identification of CTCs. As a result, the number of detections was 148(TPR = 100%) and the number of over-detected non-CTCs was 988.

DOI： 10.23919/ICCAS52745.2021.9650022

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124240604&origin=inward

A Method for Evaluating of Asymmetry on Cleft Lip Using Symmetry Plane 査読有り国際誌

Sawada S., Kamiya T., Kimura-Nomoto N., Okawachi T., Nozoe E., Nakamura N.

International Conference on Control, Automation and Systems 2021-October 1796 - 1799 2021年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Cleft lip is one of the most common morphological abnormalities that occur when the fetal lips are torn. Cleft lip is treated with multiple surgical procedures. However, there is a problem that the evaluation of surgical results depends on the subjectivity of doctors and the degree of symmetry is difficult to evaluate quantitatively. Therefore, as a previous study, a method for detecting and evaluating the plane of symmetry of the face with an asymmetric region was proposed, however, there remained a problem with the accuracy of separation of the asymmetry between patients and healthy subjects. In this paper, we propose an asymmetry evaluation method with high separation accuracy using the plane of symmetry of the face. The plane of symmetry used as the evaluation standard is obtained by adjusting the plane of symmetry of the entire face to the plane of symmetry of the surgical region. The asymmetry is evaluated by the evaluation index using the depth information with the plane of symmetry as the reference. By comparing the proposed method with the conventional method, the degree of separation of asymmetry between patients and healthy subjects was improved, and more useful results were obtained as an evaluation index.

DOI： 10.23919/ICCAS52745.2021.9649934

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85124214360&origin=inward

Automatic Classification of Respiratory Sounds Based on LSTM with Affinity Loss

Minami, Kamiya, Mabu, Kido

8th International Symposium on Applied Engineering and Sciences (SAES2020) 2020年12月

A Detection Method for Liver Cancer Region Based on Faster R-CNN and Level-set Method

Furuzuki, Kamiya

8th International Symposium on Applied Engineering and Sciences (SAES2020) 2020年12月

多元計算解剖学における高度知能化診断招待有り査読有り

平野、神谷、木戸

Medical Imaging Technology 38 ( 5 ) 217 - 221 2020年11月

深層学習による胸部CT画像からの結節状陰影の検出

玉井、神谷、青木、木戸

第33回バイオメディカル・ファジィ・システム学会年次大会講演論文集 ( 第33回バイオメディカル・ファジィ・システム学会年次大会講演論文集 ) 26 - 29 2020年10月

YOLOｖ３を用いた全天球カメラ映像からの障害物認識

甲斐、陸、神谷

第33回バイオメディカル・ファジィ・システム学会年次大会講演論文集 ( 第33回バイオメディカル・ファジィ・システム学会年次大会講演論文集 ) 84 - 88 2020年10月

SVMによる胸部CT画像からのドライバー遺伝子情報変異の検出

吉福、寺澤、神谷、青木

第33回バイオメディカル・ファジィ・システム学会年次大会講演論文集 ( 第33回バイオメディカル・ファジィ・システム学会年次大会講演論文集 ) 80 - 83 2020年10月

２つの解像度スペクトルとTF-CRNNによる呼吸音の分類

浅谷、神谷、間普、木戸

第33回バイオメディカル・ファジィ・システム学会年次大会講演論文集 ( 第33回バイオメディカル・ファジィ・システム学会年次大会講演論文集 ) 64 - 67 2020年10月

Symmetric plane detection and symmetry analysis from a 3D point cloud data of face 査読有り国際誌

Hosoki D., Kamiya T., Kimura-Nomoto N., Okawachi T., Nozoe E., Nakamura N.

International Conference on Control, Automation and Systems 2020-October 402 - 406 2020年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 Institute of Control, Robotics, and Systems - ICROS. Cleft lip occurs in about 1 in 500 people and is a congenital anomaly that occurs when the lips are not completely formed in the process of fetal face fusion. The treatment of the cleft lip is performed for the purpose of forming bilaterally symmetrical lips and external nose. However, it is based on a criterion that depends on subjectivity, quantitatively determining the degree of symmetry of the surgical site is required. In this paper, we propose a method for detecting the face that serves as the symmetry basis of the face to analyze the symmetry of the surgical site, and an index that indicates the symmetry. In the proposed method, after detecting the facial organs as points in the 3D point cloud data of the subject's face, we align the mirror image inversion excluding the region from upper lip to the tip of the nose where the effect of shape change due to cleft lip is remarkable with the original point cloud. Next, the symmetric reference plane is set by finding a plane that bisects vertically between one point in the original point cloud and the corresponding point in the mirror image inversion. As a result, the target reference plane is detected with high accuracy. Furthermore, as a result of applying the symmetry analysis method based on the target reference plane, it was possible to visually represent the left-right asymmetric region.

DOI： 10.23919/ICCAS50221.2020.9268262

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85098071444&origin=inward

An image registration method for spine region in CT images considering sagittal plane 査読有り国際誌

Yamashita Y., Kamiya T., Aoki T.

International Conference on Control, Automation and Systems 2020-October 407 - 410 2020年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 Institute of Control, Robotics, and Systems - ICROS. In recent years, the computer-aided diagnosis (CAD) systems for supporting to the physician is attracting attention in medical research field. One of them is temporal subtraction technique. It is a technique to generate images emphasizing temporal changes in lesions by performing a differential operation between current and previous image of the same subject. However, there is a problem of mistakenly selecting current and previous slice of a spine region with similar geometry as the same slice because many of the spine regions in axial plane have similar geometries. In this paper, we propose an image registration method for the detection of bone metastases from the spine region by creating a temporal subtraction images from CT images. Especially, we develop an image registration system for spine region considering sagittal plane to accurately select the same slice for the current image and the previous image. In the experiment, we applied the proposed method to CT images of 27 cases with bone metastases and the results were compared with the markings of the lesions.

DOI： 10.23919/ICCAS50221.2020.9268246

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85098087691&origin=inward

Automatic Classification of Respiratory Sounds Considering Time Series Information Based on VGG16 with LSTM 査読有り国際誌

Asatani N., Kamiya T., Mabu S., Kido S.

International Conference on Control, Automation and Systems 2020-October 423 - 426 2020年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 Institute of Control, Robotics, and Systems - ICROS. According to the 2016 World Health Organization (WHO) survey, respiratory diseases are serious diseases that account for four of the top ten causes of death in the world, accounting for more than 8 million deaths worldwide. Currently, the diagnosis of respiratory disease is made by auscultation, but in order to make an accurate diagnosis, a number of abnormal patterns of respiratory sounds need to be memorized, and the results of the diagnosis are dependent on the proficiency of the physician. Therefore, a computer aided diagnosis (CAD) system is needed to quantitatively classify the respiratory sounds and output the results as a "second opinion". In this paper, a short-time Fourier transformed spectrogram, a Constant-Q transformed logarithmic frequency spectrogram, and a continuous wavelet transformed scalogram are simultaneously input to VGG16 which is one of the network models of CNN(Convolutional Neural Network) and classified by LSTM (Long short-term memory). The proposed method is applied to 26 respiratory sounds, and the 0.90 of accuracy, sensitivity of 0.97, and specificity of 0.90 is obtained.

DOI： 10.23919/ICCAS50221.2020.9268428

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85098088561&origin=inward

Identification of normal and abnormal from ultrasound images of power devices using VGG16 査読有り国際誌

Ogawa T., Lu H., Watanabe A., Omura I., Kamiya T.

International Conference on Control, Automation and Systems 2020-October 415 - 418 2020年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 Institute of Control, Robotics, and Systems - ICROS. Power devices are semiconductor devices that handle high voltages and large currents, which are used in electric vehicles, televisions, and trains. Therefore, high reliability and safety are required, and to ensure this, power cycle tests are performed to analyze the breakdown process. Conventional tests are often difficult to analyze due to the influence of sparks generated during the test. Therefore, new tests are being developed by adding ultrasound to conventional methods. The new technology is capable of continuously recording structural changes inside the device during testing, which is expected to make testing much easier than conventional testing. However, the new technology still has some challenges. The main problems are the lack of a method for analyzing large amounts of image data and the extraction of small changes in image features that are difficult to distinguish with the human eye, and the establishment of such a system is required. In this paper, we use deep learning for image classification of the obtained ultrasound images. We propose a new network model with the addition of Batch normalization and Global average pooling to VGG16, which is a pre-trained model. In the experiment, accuracy=98.29%, TPR=98.96% and FPR=7.43% classification accuracy was obtained.

DOI： 10.23919/ICCAS50221.2020.9268275

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85098091061&origin=inward

Image registration method from LDCT image using FFD Algorithm 査読有り国際誌

Tanaka C., Kamiya T., Aoki T.

International Conference on Control, Automation and Systems 2020-October 411 - 414 2020年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 Institute of Control, Robotics, and Systems - ICROS. In recent years, the number of lung cancer deaths has been increasing. In Japan, CT (Computed Tomography) equipment is used for its visual screening. However, there is a problem that seeing huge number of images taken by CT is a burden on the doctor. To overcome this problem, the CAD (Computer Aided Diagnosis) system is introduced on medical fields. In CT screening, LDCT (Low Dose Computed Tomography) screening is desirable considering radiation exposure. However, the image quality which is caused the lower the dose is another problem on the screening. A CAD system that enables accurate diagnosis even at low doses is needed. Therefore, in this paper, we propose a registration method for generating temporal subtraction images that can be applied to low-quality chest LDCT images. Our approach consists of two major components. Firstly, global matching based on the center of gravity is performed on the preprocessed images, and the region of interest (ROI) is set. Secondly, local matching by free-form deformation (FFD) based on B-Spline is performed on the ROI as final registration. In this paper, we apply our proposed method to LDCT images of 6 cases, and reduce 57.29% in the calculation time, 26.1% in the half value width, and 29.6% in the sum of histogram of temporal subtraction images comparing with the conventional method.

DOI： 10.23919/ICCAS50221.2020.9268267

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85098056092&origin=inward

Object recognition from spherical camera images based on YOLOv3 査読有り国際誌

Kai T., Lu H., Kamiya T.

International Conference on Control, Automation and Systems 2020-October 419 - 422 2020年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 Institute of Control, Robotics, and Systems - ICROS. The aging of Japan is remarkable, and attention has been focused on the use and utilization of assistive devices. One of them is electric wheelchair, which enables physical disability people to easily operate it using a handle or a joystick. However, accidents are occurring frequently with increasing demand by using electric wheelchair. Therefore, developing an autonomous electric wheelchair is required to reduce accidents such as maneuvering mistakes, reduce the accident rate, improve convenience, and reduce the burden on caregivers. In this paper, we focus on the recognition of obstacles and use panoramic images obtained from a spherical camera that can easily handle information from all directions at low cost. A spherical camera is attached to an electric wheelchair, and images are cut out from the sequential images obtained by running. For image analysis, YOLOv3, which has been successful in the field of image recognition in recent years, is used. In the proposed method, considering the distortion of the image caused by using the spherical camera, the improvement of the model of YOLOv3 is examined, and the validity with the actual data is verified.

DOI： 10.23919/ICCAS50221.2020.9268308

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85098088521&origin=inward

Automatic Classification of Respiratory Sounds Based on Convolutional Neural Network with Multi Images 査読有り

Minami K., Lu H., Kamiya T., Mabu S., Kido S.

ACM International Conference Proceeding Series 17 - 21 2020年09月

U-Net特徴量による特発性肺線維疾患者の予後予測査読有り

植村、渡利、Janne、弘中、神谷、吉田

第１９回情報科学技術フォーラム第2分冊 ( 第１９回情報科学技術フォーラム第2分冊 ) 75 - 78 2020年09月

Deep Learning for Visual Segmentation: A Review 査読有り国際誌

Sun J., Li Y., Lu H., Kamiya T., Serikawa S.

Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020 1256 - 1260 2020年07月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 IEEE. Big data-driven deep learning methods have been widely used in image or video segmentation. The main challenge is that a large amount of labeled data is required in training deep learning models, which is important in real-world applications. To the best of our knowledge, there exist few researches in the deep learning-based visual segmentation. To this end, this paper summarizes the algorithms and current situation of image or video segmentation technologies based on deep learning and point out the future trends. The characteristics of segmentation that based on semi-supervised or unsupervised learning, all of the recent novel methods are summarized in this paper. The principle, advantages and disadvantages of each algorithms are also compared and analyzed.

DOI： 10.1109/COMPSAC48688.2020.00-84

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85094163880&origin=inward

Automatic detection of lung nodules from temporal subtraction images based on residual 3D-CNN with linear multi-shortcut 査読有り国際誌

Yoshino, Lu, Kim, Aoki, Kido

International Journal of Computer Assisted Radiology and Surgery 15 ( 6 ) 2020年06月

U-radiomics combined with hyper-curvature features for predicting survival of patients with idiopathic pulmonary fibrosis 査読有り国際誌

Uemura, Watari, Nappi, Matsuhiro, Niki, Kim, Yoshida

International Journal of Computer Assisted Radiology and Surgery 15 ( 6 ) 2020年06月

Deep-Sea Organisms Tracking Using Dehazing and Deep Learning 査読有り国際誌

Lu H., Uemura T., Wang D., Zhu J., Huang Z., Kim H.

Mobile Networks and Applications 25 ( 3 ) 1008 - 1015 2020年06月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018, Springer Science+Business Media, LLC, part of Springer Nature. Deep-sea organism automatic tracking has rarely been studied because of a lack of training data. However, it is extremely important for underwater robots to recognize and to predict the behavior of organisms. In this paper, we first develop a method for underwater real-time recognition and tracking of multi-objects, which we call “You Only Look Once: YOLO”. This method provides us with a very fast and accurate tracker. At first, we remove the haze, which is caused by the turbidity of the water from a captured image. After that, we apply YOLO to allow recognition and tracking of marine organisms, which include shrimp, squid, crab and shark. The experiments demonstrate that our developed system shows satisfactory performance.

DOI： 10.1007/s11036-018-1117-9

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85055349873&origin=inward

Detection of Circulating Tumor Cells in Fluorescence Microscopy Images Based on ANN Classifier 査読有り国際誌

Tsuji K., Lu H., Tan J.K., Kim H., Yoneda K., Tanaka F.

Mobile Networks and Applications 25 ( 3 ) 1042 - 1051 2020年06月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018, Springer Science+Business Media, LLC, part of Springer Nature. Circulating tumor cells (CTCs) is a clinical biomarker for cancer metastasis. CTCs are cells circulating in the body of patients by being separated from primary cancer and entering into blood vessel. CTCs spread every positions in the body, and this is one of the cause of cancer metastasis. To analyze them, pathologists get information about metastasis without invasive test. CTCs test is conducted by analyzing the blood sample from patient. The fluorescence microscope generates a large number of images per each sample, and images contain a lot of cells. There are only a few CTCs in images and cells often have blurry boundaries. So CTCs identification is not an easy work for pathologists. In this paper, we develop an automatic CTCs identification method in fluorescence microscopy images. This proposed method has three section. In the first approach, we conduct the cell segmentation in images by using filtering methods. Next, we compute feature values from each CTC candidate region. Finally, we identify CTCs using artificial neural network algorithm. We apply the proposed method to 5895 microscopy images (7 samplesas), and evaluate the effectiveness of our proposed method by using leave-one-out cross validation. We achieve the result of performance tests, a true positive rate is 92.57% and false positive rate is 9.156%.

DOI： 10.1007/s11036-018-1121-0

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85053481889&origin=inward

High-quality-guided artificial bee colony algorithm for designing loudspeaker 査読有り国際誌

Gao H., Li H., Liu Y., Lu H., Kim H., Pun C.M.

Neural Computing and Applications 32 ( 9 ) 4473 - 4480 2020年05月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018, The Natural Computing Applications Forum. Designing the loudspeaker could be concerned as an optimization problem. Like most electromagnetic device design issues, it demonstrates multimodal, multidimensional and constrained. The traditional design method cannot achieve a satisfactory model within a certain period of time. In this paper, a high-quality-guided artificial bee colony algorithm is proposed to increase its convergence speed and search accuracy by gradually changing the number of updating dimensions and searching closer to better locations. The algorithm is first tested on some representative basic benchmark functions and then is applied to a loudspeaker design problem. By comparing with some classical algorithms, the performance of our algorithm is verified.

DOI： 10.1007/s00521-018-3568-0

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85048041015&origin=inward

An Automatic Design of Camouflage Patterns Based on CNNs 査読有り

Wei X., Wang K., Li G., Kim H.

ACM International Conference Proceeding Series 257 - 260 2020年04月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In order to get with the environmental changes of battlefield quickly, the military camouflage should be changeable. If the camouflage patterns of the clothes and vehicles like tanks are different from the environment, it's very easy for cameras of enemies to find. As we all know that the same patterns is used in the most of current military all over the world. In this paper, we propose a novel feature-extraction method from an image using convolutional neural networks. Then the pattern will be combined with the environmental style pattern. The composite image is mapped onto the surface of the actual 3D clothes and vehicles finally. In this paper, the Eye-Movement equipment is applied to evaluate the results for better comparison. We can produce the proper pattern even the different and complicated environment.

DOI： 10.1145/3404555.3404637

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85092212124&origin=inward

SqueezeNetによる蛍光顕微鏡画像上のCTC自動検出査読有り

中道、陸、金、米田、田中

画像電子学会誌 49 ( 2 ) 136 - 143 2020年04月

CT 像からの骨転移検出のための画像位置合わせ法

佐藤、陸、金、青木

電子情報通信学会総合大会 108 2020年03月

時系列情報を考慮したCNNによる呼吸音の自動分類

浅谷、陸、金、間普、木戸

信学技報IE2019-97 9 - 10 2020年03月

比較読影のための画像位置合わせ法招待有り

金、陸、青木、木戸

信学技報IE2019-97 69 - 72 2020年03月

深層学習による磁粉探傷検査における欠陥検出法

上田、陸、金

信学技報IE2019-97 1 - 2 2020年03月

顔の3次元点群からの対称面検出と対称性解析

細木、陸、金、木村、大河内、野添、中村

信学技報IE2019-97 263 - 266 2020年03月

Deep Radiomicsによる胸部CT画像からのドライバー遺伝子情報変異の検出

吉福、寺澤、陸、金、青木

信学技報IE2019-97 35 - 39 2020年03月

Clinical Applications of Multidisciplunary Computational Anatomy to Diagnosis Progress Overview FY2014-Fy2018

Kido, Hashimoto, Hirano, Mabu, Kim, Kumura, Noriki, Inai, Tachibana

多元計算解剖学」最終成果報告書 223 - 232 2020年03月

3D-CNN with residual sub-blocks for automatic detection of lung nodules from temporal subtraction images 査読有り

Yoshino Y., Lu H., Kim H., Aoki T., Kido S.

2020 International Conference on Artificial Intelligence in Information and Communication, ICAIIC 2020 647 - 652 2020年02月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 IEEE. Temporal subtraction (TS) technique is one of computer aided diagnosis (CAD) systems. A TS image is obtained by subtracting a previous image, which are warped to match between the structures of the previous image and one of a current image, from the current image. TS technique removes normal structures and enhances interval changes. However, many subtraction artifacts that can be detected as false positives still remain on a TS image. In this paper, we propose 3D-CNNs with residual sub-blocks based on 3D-VGG16-like architecture for detection of nodules accurately from TS images.

DOI： 10.1109/ICAIIC48513.2020.9065197

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85084033837&origin=inward

Automatic extraction of abnormalities on temporal subtraction images using sparse coding and 3D-CNN 査読有り国際誌

Koizumi, Miyake, Lu, Kim, Aoki, Kido

The 2020 International Conference on Artificial Lefe and Robotics 783 - 786 2020年01月

Object tracking method considering time series information using Re3 with stochastic depth 査読有り国際誌

Kitayama, Kim

The 2020 International Conference on Artificial Lefe and Robotics 2020年01月

Hyperspectral Images Segmentation Using Active Contour Model for Underwater Mineral Detection 査読有り

Lu H., Zheng Y., Hatano K., Li Y., Nakashima S., Kim H.

Studies in Computational Intelligence 810 513 - 522 2020年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2020, Springer Nature Switzerland AG. In this paper, we design a novel underwater hyperspectral imaging technique for deep-sea mining detection. The spectral sensitivity peaks are in the region of the visible spectrum, 580, 650, 720, 800 nm. In addition, to the underwater objects recognition, because of the physical properties of the medium, the captured images are distorted seriously by scattering, absorption and noise effect. Scattering is caused by large suspended particles, such as in turbid water, which contains abundant particles, algae, and dissolved organic compounds. In order to resolve these problems of recognizing mineral accurately, fast and effectively, an identifying and classifying algorithm is proposed in this paper. We take the following steps: firstly, through image preprocessing, hyperspectral images are gained by using denoising, smoothness, image erosion. After that, we segment the cells by the method of the modified active contour method. These methods are designed for real-time execution on limited-memory platforms, and are suitable for detecting underwater objects in practice. The Initial results are presented and experiments demonstrate the effectiveness of the proposed imaging system.

DOI： 10.1007/978-3-030-04946-1_50

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85062907061&origin=inward

DeepEye: A Dedicated Camera for Deep-Sea Tripod Observation Systems 査読有り

Lu H., Li Y., Kim H., Serikawa S.

Studies in Computational Intelligence 810 507 - 511 2020年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2020, Springer Nature Switzerland AG. The deep-sea tripod systems are designed and built at the U.S. Geological Survey (USGS) Pacific Coastal and Marine Science Center (PCMSC) in Santa Cruz, California. They are recovered in late September 2014 after spending about half a year collecting data on the floor of the South China Sea. The deep-sea tripod systems are named as Free-Ascending Tripod (FAT), are deployed at 2,100 m water depth—roughly 10 times as deep as most tripods dedicated to measuring currents and sediment movement at the seafloor. Deployment at this unusual depth was made possible by the tripod’s ability to rise by itself to the surface rather than being pulled up by a line. Instruments mounted on the tripod took bottom photographs and measured such variables as water temperature, current velocity, and suspended-sediment concentration. FAT is used to better understand how and where deep-seafloor sediment moves and accumulates. Besides of this, we also use them to study the deep-sea biology. The obtained the images from the camera, the biology animals are hardly to be distinguished. In this project, we are concerned to use novel underwater imaging technologies for recovering the deep-sea scene.

DOI： 10.1007/978-3-030-04946-1_49

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85062904328&origin=inward

Supervoxel Graph Cuts: An Effective Method for GGO Candidate Regions Extraction on CT Images 査読有り国際誌

Lu H., Kondo M., Li Y., Tan J., Kim H., Murakami S., Aoki T., Kido S.

IEEE Consumer Electronics Magazine 9 ( 1 ) 61 - 66 2020年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2012 IEEE. In this article, a method to reduce artifacts on temporal difference images is introduced. The proposed method uses a nonrigid registration method for ground glass opacification (GGO), which is light in concentration and difficult to detect early. In this method, global matching, local matching, and three-dimensional (3D) elastic matching are performed on the current and previous images, and an initial temporal subtraction image is generated. After that, we use an Iris filter, which is the gradient vector concentration degree filter, to determine the initial GGO candidate regions and use supervoxel and graph cuts to segment region of interest in the 3D images. For each extracted region, a support vector machine is used to reduce the oversegmentation. The voxel matching is applied to generate the final temporal difference image, emphasizing the GGO regions while reducing the artifact.

DOI： 10.1109/MCE.2019.2941468

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85076374234&origin=inward

時間-周波数解析と畳み込みニューラルネットワークを用いた呼吸音の自動分類査読有り

南弘毅, 陸慧敏, 金亨燮, 平野靖, 間普真吾, 木戸尚治

Ｍｅｄｉｃａｌ　Ｉｍａｇｉｎｇ　Ｔｅｃｈｎｏｌｏｇｙ ( 日本医用画像工学会 ) 38 ( 1 ) 40 - 47 2020年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

<p>呼吸器疾患の診断方法としては，聴診器を用いた呼吸音の聴診が長年用いられてきた．これは簡便で安全な診断方法である一方，聴診音の診断には定量的な評価基準がないため，医師の診断支援を行うシステムの開発が必要である．そこで本論文では，畳み込みニューラルネットワーク（CNN： convolutional neural network）を用いた呼吸音の自動分類手法の提案を行う．おもな手法の流れとしては，呼吸音データに対して短時間フーリエ変換と連続ウェーブレット変換を適用し，スペクトログラム画像およびスカログラム画像を生成する．その後，生成した画像を用いてCNN による正常呼吸音，連続性ラ音，断続性ラ音の識別を行う．提案手法を呼吸音データ22 症例に適用した結果，分類性能として，全体正解率は 79.44［％］，ROC（receiver operating characteristic）曲線に基づくAUC（area under the curve）は0.942 を得た．</p>

DOI： 10.11409/mit.38.40

その他リンク： https://ci.nii.ac.jp/naid/130007796400

U-radiomics for predicting survival of patients with idiopathic pulmonary fibrosis 査読有り国際誌

Uemura T., Watari C., Näppi J.J., Hironaka T., Kim H., Yoshida H.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE 11314 2020年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 SPIE. We developed and evaluated the effect of U-Net-based radiomic features, called U-radiomics, on the prediction of the overall survival of patients with idiopathic pulmonary fibrosis (IPF). To generate the U-radiomics, we retrospectively collected lung CT images of 72 patients with interstitial lung diseases. An experienced observer delineated regions of interest (ROIs) from the lung regions on the CT images, and labeled them into one of four interstitial lung disease patterns (ground-glass opacity, reticulation, consolidation, and honeycombing) or a normal pattern. A U-Net was trained on these images for classifying the ROIs into one of the above five lung tissue patterns. The trained U-Net was applied to the lung CT images of an independent test set of 75 patients with IPF, and a U-radiomics vector for each patient was identified as the average of the bottleneck layer of the U-Net across all the CT images of the patient. The U-radiomics vector was subjected to a Cox proportional hazards model with elastic-net penalty for predicting the survival of the patient. The evaluation was performed by using bootstrapping with 500 replications, where concordance index (C-index) was used as the comparative performance metric. The preliminary results showed the following C-index values for two clinical biomarkers and the U-radiomics: (a) composite physiologic index (CPI): 64.6%, (b) gender, age, and physiology (GAP) index: 65.5%, and (c) U-radiomics: 86.0%. The U-radiomics significantly outperformed the clinical biomarkers in predicting the survival of IPF patients, indicating that the U-radiomics provides a highly accurate prognostic biomarker for patients with IPF.

DOI： 10.1117/12.2551273

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85085531092&origin=inward

GAN-based survival prediction model from CT images of patients with idiopathic pulmonary fibrosis 査読有り国際誌

Uemura T., Watari C., Näppi J.J., Hironaka T., Kim H., Yoshida H.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE 11318 2020年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only. We developed a novel survival prediction model for images, called pix2surv, based on a conditional generative adversarial network (cGAN), and evaluated its performance based on chest CT images of patients with idiopathic pulmonary fibrosis (IPF). The architecture of the pix2surv model has a time-generator network that consists of an encoding convolutional network, a fully connected prediction network, and a discriminator network. The fully connected prediction network is trained to generate survival-time images from the chest CT images of each patient. The discriminator network is a patchbased convolutional network that is trained to differentiate the "fake pair" of a chest CT image and a generated survivaltime image from the "true pair" of an input CT image and the observed survival-time image of a patient. For evaluation, we retrospectively collected 75 IPF patients with high-resolution chest CT and pulmonary function tests. The survival predictions of the pix2surv model on these patients were compared with those of an established clinical prognostic biomarker known as the gender, age, and physiology (GAP) index by use of a two-sided t-test with bootstrapping. Concordance index (C-index) and relative absolute error (RAE) were used as measures of the prediction performance. Preliminary results showed that the survival prediction by the pix2surv model yielded more than 15% higher C-index value and more than 10% lower RAE values than those of the GAP index. The improvement in survival prediction by the pix2surv model was statistically significant (P < 0.0001). Also, the separation between the survival curves for the low- and high-risk groups was larger with pix2surv than that of the GAP index. These results show that the pix2surv model outperforms the GAP index in the prediction of the survival time and risk stratification of patients with IPF, indicating that the pix2surv model can be an effective predictor of the overall survival of patients with IPF.

DOI： 10.1117/12.2551369

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85082610220&origin=inward

Comparative performance of 3D-DenseNet, 3D-ResNet, and 3D-VGG models in polyp detection for CT colonography 査読有り国際誌

Uemura T., Näppi J.J., Hironaka T., Kim H., Yoshida H.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE 11314 2020年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 SPIE. Three-dimensional (3D) convolutional neural networks (CNNs) can process volumetric medical imaging data in their native volumetric input form. However, there is little information about the comparative performance of such models in medical imaging in general and in CT colonography (CTC) in particular. We compared the performance of a 3D densely connected CNN (3D-DenseNet) with those of the popular 3D residual CNN (3D-ResNet) and 3D Visual Geometry Group CNN (3D-VGG) in the reduction of false-positive detections (FPs) in computer-aided detection (CADe) of polyps in CTC. VGG is the earliest CNN design of these three models. ResNet has been used widely as a de-facto standard model for constructing deep CNNs for image classification in medical imaging. DenseNet is the most recent of these models and improves the flow of information and reduces the number of network parameters as compared to those of ResNet and VGG. For the evaluation, we used 403 CTC datasets from 203 patients. The classification performance of the CNNs was evaluated by use of 5-fold cross-validation, where the area under the receiver operating characteristic curve (AUC) was used as the figure of merit. Each training fold was balanced by use of data augmentation of the samples of real polyps. Our preliminary results showed that the AUC value of the 3D-DenseNet (0.951) was statistically significantly higher than those of the reference models (P < 0.005), indicating that the 3D-DenseNet has the potential of substantially outperforming the other models in reducing FPs in CADe for CTC. This improvement was highest for the smallest polyps.

DOI： 10.1117/12.2549103

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85085513046&origin=inward

Comparative performance of 3D machine-learning and deep-learning models in the detection of small polyps in dual-energy CT colonography 査読有り国際誌

Näppi J.J., Uemura T., Kim S.H., Kim H., Yoshida H.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE 11314 2020年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2020 SPIE. Colorectal cancer is the second leading cause of cancer deaths worldwide. Computed tomographic colonography (CTC) can detect large colorectal polyps and cancers at a high sensitivity, whereas it can miss some of the smaller but still clinically significant 6 - 9 mm polyps. Dual-energy CTC (DE-CTC) can be used to provide more detailed information about scanned materials than does conventional single-energy CTC. We compared the classification performance of a 3D convolutional neural network (DenseNet) with those of four traditional 3D machine-learning models (AdaBoost, support vector machine, random forest, Bayesian neural network) and their cascade and ensemble classifier variants in the detection of small polyps in DE-CTC. Twenty patients with colonoscopy-confirmed polyps were examined by DE-CTC with a reduced one-day bowel preparation. The traditional machine-learning models were designed to identify polyps based on native radiomic dual-energy features of the DE-CTC image volumes. The performance of the machine-learning models was evaluated by use of the leave-one-patient-out method. The DenseNet was trained with a large independent external dataset of single-energy CTC cases and tested on blended image volumes of the DE-CTC cases. Although the DenseNet yielded the highest detection accuracy for typical polyps, AdaBoost and its cascade classifier variant yielded the highest overall polyp detection performance.

DOI： 10.1117/12.2549793

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85085516392&origin=inward

Correction to: Deep-Sea Organisms Tracking Using Dehazing and Deep Learning (Mobile Networks and Applications, (2020), 25, 3, (1008-1015), 10.1007/s11036-018-1117-9) 査読有り国際誌

Lu H., Uemura T., Wang D., Zhu J., Huang Z., Kim H.

Mobile Networks and Applications 25 ( 6 ) 2020年01月

Automatic extraction of abnormalities on temporal ct subtraction images using sparse coding and 3d-cnn 査読有り

Koizumi Y., Miyake N., Lu H., Kim H., Aoki T., Kido S.

Proceedings of International Conference on Artificial Life and Robotics 2020 783 - 786 2020年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, the rate of death from cancer has tended to increase in Japan. Also, there is a concern that increasing the performance of CT will increase the burden on doctors. Therefore, by presenting a "second opinion" in the CAD system, the burden on doctors can be reduced. We develop a CAD system for automatic detection of lung cancer. In this paper we propose a method to detect abnormalities based on temporal subtraction technique, sparse coding and 3D-CNN. We obtain the result that sparse level contributed most to the score.

DOI： 10.5954/ICAROB.2020.GS3-3

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85108848194&origin=inward

Object tracking method considering time series information using re3 with stochastic depth 査読有り

Kitayama T., Kim H.

Proceedings of International Conference on Artificial Life and Robotics 2020 477 - 480 2020年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In recent years, there has been an increasing demand for automation of cargo handling operations in harbors. However, its automation has not been realized in Japan since there are many fundamental technologies to be solved until now. In this study, we propose a tracking method of the container gripping area based on Re3 for the purpose of automation of cargo handling work. Re3, an object tracking method used in the conventional method, has a problem that global features cannot be extracted well. In order to solve this problem, our method incorporates a model called Stochastic Depth.

DOI： 10.5954/ICAROB.2020.GS5-4

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85108782511&origin=inward

Marine Organisms Tracking and Recognizing Using YOLO 査読有り国際誌

Uemura T., Lu H., Kim H.

EAI/Springer Innovations in Communication and Computing 53 - 58 2020年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

A system that investigates deep sea automatically has never developed. A purpose of this study is developing such a system. We employed a technique of recognition and tracking of multi-objects, called “You Only Look Once: YOLO.” This method provides us very fast and accurate tracker. In our system, we remove the haze, which is caused by turbidity of water, from image. After its process, we apply “YOLO” to tracking and recognizing the marine organisms, which includes shrimp, squid, crab, and shark. Our developed system shows generally satisfactory performance.

DOI： 10.1007/978-3-030-17763-8_6

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85097568303&origin=inward

pix2surv: A Generative Adversarial Network Model for Prediction of Survival in Patients with Interstitial Lung Diseases 査読有り国際誌

Uemura T, Watari C, Näppi JJ, Hironaka T, Kim H, Yoshida H

Radiological Society of North America 2019 Scientific Assembly and Annual Meeting 2019年12月

YOLOｖ２を用いた海中生物のロバスト検出招待有り査読有り

陸、酒井、金

画像ラボ 30 ( 12 ) 21 - 25 2019年12月

Artificial intelligence (AI) for CT colonography: the new horizons of colorectal screening 査読有り国際誌

Näppi JJ, Uemura T, Pickhardt PJ, Kim H, Yoshida H.

Radiological Society of North America 2019 Scientific Assembly and Annual Meeting 2019年12月

Generative adversarial network models for prediction of survival in patients with interstitial lung diseases 査読有り国際誌

Uemura T, Watari C, Näppi JJ, Hironaka T, Kim H, Yoshida H

Radiological Society of North America 2019 Scientific Assembly and Annual Meeting 2019年12月

Touch switch sensor for cognitive body sensor networks 査読有り国際誌

Li Y., Lu H., Kim H., Serikawa S.

Computer Communications 146 32 - 38 2019年10月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2019 Elsevier B.V. With the global popularity of Internet of Things (IoT) technology, increasingly numbers of digital mobile products have been developed, and they have increased the productivity of people's daily lives. These electronic products are used in all aspects of life, such as medical care, office life, home services, and sports. However, most of these products are designed for healthy people with high literacy rates. For disabled people, these products cannot be widely used. In this paper, new, differently shaped touch sensors are proposed for body sensor network-based devices. This touch sensor can be formed into any shape because of the use of conductive fabric adhesive tape as a switch. That property is why the sensor can change positions in the body sensor network in which the human body is used as a trigger to safely activate the touch switch. The number of switch sensors can easily be increased or decreased without changing the wiring of the central controller. The number of sensors in a switch sensor system is greater than that in other touch switch systems, and the accuracy is higher.

DOI： 10.1016/j.comcom.2019.07.019

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85069818135&origin=inward

A Detection Method for Liver Cancer Region Based on Faster R-CNN 査読有り

Furuzuki M., Lu H., Kim H., Hirano Y., Mabu S., Tanabe M., Kido S.

International Conference on Control, Automation and Systems 2019-October 808 - 811 2019年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2019 Institute of Control, Robotics and Systems - ICROS. In recent years, liver cancer has become the fourth-largest number of deaths in the world. Surgery is a typical treatment for liver cancer. Therefore, advance information about the number and size of cancer is important for surgery. Multi-phase CT images are well known diagnostic method. By extracting the region of the liver and the region of cancer from the obtained CT image, the shape can be finally restored in 3D. In this paper, as a preliminary step to construct an image analysis method for efficiently extracting cancerous regions in multi-phase CT, we propose a method of obtaining a rectangular region as a rough cancerous region of interest. As a method, after preprocessing the input image, using Faster R-CNN, the region of interest including the cancer region is extracted as a rectangle. As a result of applying this method to 11 cases of arterial phase of multi-phase CT, the detection performance was different depending on the network model adopted for backbone part.

DOI： 10.23919/ICCAS47443.2019.8971627

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85079102007&origin=inward

Segmentation of Bone Metastasis in CT Images Based on Modified HED 査読有り国際誌

Song Y., Lu H., Kim H., Murakami S., Ueno M., Terasawa T., Aoki T.

International Conference on Control, Automation and Systems 2019-October 812 - 815 2019年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2019 Institute of Control, Robotics and Systems - ICROS. Segmentation of the bone metastasis area in medical images can reduce the workload for diagnosis and treatment. However, there are various shapes and sizes of bone metastasis also affected by noise. As a result, it is difficult to segment using classical segmentation methods. In this paper, we propose a convolutional neural network model-based segmentation method. The proposed method easily predicts the contour and location of the lesion area using side connection and modified network. In this study, we modified again the modified HED network to match the characteristics of bone metastasis. The experimental results using the proposed method for segmenting bone metastasis in the lesion area has 79.8[%] of TP rate and 69.2[%] of IOU rate.

DOI： 10.23919/ICCAS47443.2019.8971539

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85079084833&origin=inward

Recognition of Surrounding Environment for Electric Wheelchair Based on WideSeg 査読有り国際誌

Sakai Y., Nakayama Y., Lu H., Li Y., Kim H.

International Conference on Control, Automation and Systems 2019-October 816 - 820 2019年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2019 Institute of Control, Robotics and Systems - ICROS. At present, the aging population is growing in Japan. Along with that, the expectation for the utilization of welfare equipment is increasing. Electric wheelchair, a convenient transportation tool, is popularized rapidly. On the other hand, accidents have occurred, and the dangers for driving are pointed out. Therefore, it needs to improve accident factors, reduce accidents and improve the convenience of electric wheelchair by automation. Environmental recognition is necessary for the development of autonomous electric wheelchair. Environmental recognition includes self-position estimation, recognition of sidewalks, crosswalks and traffic lights, moving object prediction, etc. In order to solve these various problems, this paper examines the segmentation of sidewalks, crosswalks and traffic lights. We develop the WideSeg that is one of semantic segmentation algorithms applying convolutional neural networks (CNN).

DOI： 10.23919/ICCAS47443.2019.8971608

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85079103064&origin=inward

Detection of Lung Nodules from Temporal Subtraction Image Using Deep Learning 査読有り国際誌

Tamai K., Miyake N., Lu H., Kim H., Murakami S., Aoki T., Kido S.

International Conference on Control, Automation and Systems 2019-October 1033 - 1036 2019年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2019 Institute of Control, Robotics and Systems - ICROS. In recent years, the number of death due to lung cancer is increasing year by year worldwide. Early detection and early treatment of lung cancer are important. Especially, early detection of the abnormalities on thoracic MDCT images detection of small nodules is required in visual screening. Although a CT apparatus is used for the examination, the burden on the image interpretation doctor is large due to the high performance of the CT, so the diagnostic accuracy may be reduced. In this paper, we propose an image analysis method to detect abnormal shadows from chest CT images automatically. The initial lesion candidate areas are extracted by using temporal subtraction technique that emphasizes temporal change by subtracting from a current image to previous one which is obtained same subject. The image of the area is given as input and classification is performed by CNN (Convolutional Neural Network). In the discrimination experiment based on our proposed method, 90.26 [%] of true positive rates and 13.58 [%] of false positive rates are obtained from the 49 clinical cases.

DOI： 10.23919/ICCAS47443.2019.8971533

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85079099157&origin=inward

Detection of Facial Symmetric Plane Based on Registration of 3D Point Cloud 査読有り国際誌

Hosoki D., Lu H., Kim H., Kimura N., Okawachi T., Nozoe E., Nakamura N.

International Conference on Control, Automation and Systems 2019-October 1037 - 1041 2019年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2019 Institute of Control, Robotics and Systems - ICROS. Cleft lip is a birth defect that occurs when the lips are not completely formed during healing of the face of the fetus. In Japan, it occurs in about 1 in 500 people. Although treatment is performed to form a symmetrical outer nose, it is necessary to evaluate the degree of symmetry of the surgical site quantitatively because it is based on the judgment criteria that depend on the doctor's subjectivity. In this paper, we propose a method to detect the plane which is the symmetry basis of the face to analyze the symmetry of the operation site. In the proposed method, the face organ is detected as points from 3D point cloud of the face. Then, the mirror image inversion excluding the area affected by shape change due to cleft lip is aligned with the original point cloud. Next, a symmetric plane is set by finding a plane that bisects vertically between one point in the original point cloud and the corresponding point in the mirror image. As a result of applying the proposed method to real 3D point cloud, we could detect the symmetric plane with good accuracy.

DOI： 10.23919/ICCAS47443.2019.8971537

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85079096298&origin=inward

Classification of Circulating Tumor Cells in Fluorescence Microscopy Images Based on SqueezeNet 査読有り国際誌

Nakamichi K., Lu H., Kim H., Yoneda K., Tanaka F.

International Conference on Control, Automation and Systems 2019-October 1042 - 1045 2019年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2019 Institute of Control, Robotics and Systems - ICROS. Circulating Tumor Cells (CTC) is expected as a useful biomarker test that can evaluate cancer metastasis. CTC exists in the blood of cancer patients and is considered to be an incentive of cancer metastasis. Pathologists analyze the blood to find these metastasis cancers from three colors of fluorescence microscopy images, but the manual analysis is time-consuming. In this paper, we develop an automatic CTC classification method in fluorescence microscopy images to reduce the burden of pathologists. In the proposed method, we detect cell regions by the bacterial foraging-based edge detection (BFED) algorithm and classify CTC by SqueezeNet, which is the kind of convolutional neural network (CNN). We apply the proposed method to 5040 microscopy images (6 samples) and evaluate the effectiveness. The experimental results demonstrate that the proposed method has a true positive rate is 89.86% and a false positive rate is 3.27%.

DOI： 10.23919/ICCAS47443.2019.8971646

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85079092431&origin=inward

Automatic Segmentation Method of Phalange Regions Based on Residual U-Net and MSGVF Snakes 査読有り国際誌

Kawagoe K., Hatano K., Murakami S., Lu H., Kim H., Aoki T.

International Conference on Control, Automation and Systems 2019-October 1046 - 1049 2019年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2019 Institute of Control, Robotics and Systems - ICROS. Bone diseases include rheumatoid arthritis and osteoporosis. Although visual screening using computed radiography (CR) images is an effective method for diagnosing osteoporosis, there are some similar diseases that exhibit low bone mass status. To this end, we aim to develop a computer-aided diagnostic (CAD) system to support the automatic diagnosis of osteoporosis from CR images. In this paper, we use convolutional neural network (CNN) and multiscale gradient vector flow snakes (MSGVF Snakes) algorithms to segment each finger bone regions from the CR image. The proposed method is applied to 15 cases, 92.95 [%] of the true positive rates, 2.21 [%] of the false positive rates, 7.05 [%] of the false negative rates are obtained respectively.

DOI： 10.23919/ICCAS47443.2019.8971740

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85079102296&origin=inward

Automatic Classification of Large-Scale Respiratory Sound Dataset Based on Convolutional Neural Network 査読有り国際誌

Minami K., Lu H., Kim H., Mabu S., Hirano Y., Kido S.

International Conference on Control, Automation and Systems 2019-October 804 - 807 2019年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2019 Institute of Control, Robotics and Systems - ICROS. Auscultation of respiratory sounds is very important for discovering the respiratory disease. However, there is no quantitative evaluation method for the diagnosis of respiratory sounds until now. It is necessary to develop a system to support the diagnosis of respiratory sounds. In addition, there are few studies using dataset suitable for generating realistic classification models that can be used in clinical sites in algorithm development for automatic analysis of respiratory sounds. We describe the development of an algorithm for the automatic classification of the large-scale respiratory sound dataset used in ICBHI 2017 Challenge as containing crackles, containing wheeze, containing both, and normal. Our approach consists of two major components. Firstly, transformation of one-dimensional signals into two-dimensional time-frequency representation images using short-time Fourier transform and continuous wavelet transform. Secondly, classification of transferred images using convolutional neural networks. In this paper, we apply our proposed method to 920 respiratory sound data, and achieve score of 28[%], harmonic score of 81[%], sensitivity of 54[%] and specificity of 42[%].

DOI： 10.23919/ICCAS47443.2019.8971689

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85079096950&origin=inward

3D-DenseNetによるCT Colonographyにおける擬陽性陰影の低減査読有り

植村、Janne、陸、橘、弘中、金、吉田

第18回情報科学技術フォーラム第2分冊 387 - 388 2019年09月

SqueezeNetを用いた顕微鏡画像からの血中循環がん細胞の自動識別査読有り

中道、陸、金、米田、田中

第18回情報科学技術フォーラム第2分冊 391 - 394 2019年09月

敵対的生成ネットワークによる間質性肺疾患患者の予後予測査読有り

植村、渡利、Janne、弘中、金、吉田

第18回情報科学技術フォーラム第2分冊 389 - 390 2019年09月

Convolutional neural network (CNNs) based image diagnosis for failure analysis of power devices 査読有り国際誌

Watanabe A., Hirose N., Kim H., Omura I.

Microelectronics Reliability 100-101 2019年09月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2019 Elsevier Ltd An image diagnosis by deep learning was applied to failure analysis of power devices. A series of images during a process to failure by power cycling test was used for this method. The images were obtained by a scanning acoustic microscopy of our real-time monitoring system. An image classifier was designed based on a convolutional neural network (CNNs). A developed classifier successfully diagnosed input image into a normal device and an abnormal device. The accuracy of classification was improved by introducing a pre-training and an overlapping pooling into the system. A technique to extract a feature related a failure is essential for the failure analysis based on the real-time monitoring and the deep learning is one likely candidate for it.

DOI： 10.1016/j.microrel.2019.113399

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85074484253&origin=inward

Multi-organ segmentation from abdominal CT with random forest based statistical shape model 査読有り国際誌

Wu J., Li G., Lu H., Kim H.

ACM International Conference Proceeding Series 67 - 70 2019年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2019 Association for Computing Machinery. An automatic multi-organ segmentation method from upper abdominal CT image is proposed in this paper. A group of statistical shape models for multiple organs are generated by learning the statistical distribution of organs' shapes and intensity profiles. Then, a random forest regression model is trained to find the candidate position to initialize the statistical shape model. The proposed method is evaluated at segmentation of four abdomen organs (spleen, right kidney, left kidney and liver) from training set of 26 cases of upper abdominal CT images. The accuracy shows that the initialization improves the accuracy for statistical shape model-based segmentation.

DOI： 10.1145/3354031.3354042

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85073475971&origin=inward

Preface 査読有り

Kim H.

ACM International Conference Proceeding Series 2019年08月

Detection of Abnormal Regions on Temporal Subtraction Images Based on CNN 査読有り国際誌

Nagao, Lu, Kim, Aoki, Kido

International Conference on Biomedical Signal and Image Processing 83 - 86 2019年08月

Pilot Study of Visual and Quantitative Image Analysis of Facial Surface Asymmetry in Unilateral Complete Cleft Lip and Palate 査読有り国際誌

Kimura N., Kim H., Okawachi T., Fuchigami T., Tezuka M., Kibe T., Amir M., Inada E., Ishihata K., Nozoe E., Nakamura N.

The Cleft palate-craniofacial journal : official publication of the American Cleft Palate-Craniofacial Association 56 ( 7 ) 960 - 969 2019年08月

記述言語：英語掲載種別：研究論文（学術雑誌）

OBJECTIVE: To visualize and quantitatively analyze facial surface asymmetry following primary cleft lip repair in patients with unilateral cleft lip and palate (UCLP) and to compare this with noncleft controls. DESIGN: Retrospective comparative study. PATIENTS: Twenty-two patients with complete UCLP who underwent primary lip repair from 2009 to 2013 were enrolled in this study. The preserved 3-dimensional (3D) data of 23 healthy Japanese participants with the same age were used as controls. INTERVENTIONS: All patients had received primary labioplasty in accordance with Cronin triangular flap method with orbicular oris muscle reconstruction. MAIN OUTCOME MEASURES: Shadow and zebra images established from moiré images, which were reconstructed from 3D facial data using stereophotogrammetry, were bisected and reversed by the symmetry axes (the middle line of the face). The discrepancies of the gravity and density between cleft and noncleft sides in 2 regions of interest, facial and lip areas, were then calculated and compared with those of healthy participants. RESULTS: In the UCLP group, the mean discrepancies of gravity on shadow and zebra images were 1.76 ± 0.70 and 2.63 ± 1.72 pixels, respectively, in the facial area and 1.31 ± 0.36 and 3.83 ± 2.08 pixels, respectively, in the lip area. There was a significant difference in the mean discrepancies of gravity and density on zebra images in the lip area between the UCLP and control groups. CONCLUSIONS: Our image analysis of digital facial surface asymmetry in patients with UCLP provides visual and quantitative information, and it may contribute to improvements in muscle reconstruction on cleft lip repair.

DOI： 10.1177/1055665618819645

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85069287671&origin=inward

Extreme ROS reality: A representation framework for robot using image dehazing and VR 査読有り国際誌

Ueda, Lu, Kim

3rd EAI International Conference on Robotic Sensor Networks 21 - 32 2019年08月

時間‐周波数解析とCNNを用いた呼吸音の自動分類

南　弘毅, 陸, 金, 平野, 間普, 木戸

第38回日本医用画像工学会大会予稿集 518 - 532 2019年07月

胸部CT像中の肺結節の良悪性鑑別における自動抽出された画像特徴の可視化

平島翔, 平野靖, 木戸尚治, 岩野信吾, 本田健, 関順彦, 金亨燮

第38回日本医用画像工学会大会予稿集 433 - 439 2019年07月

Faster R-CNNによる肝臓がん候補領域の抽出法

古月　夢奇, 陸, 金, 平野, 間普, 田辺, 木戸

第38回日本医用画像工学会大会予稿集 638 - 644 2019年07月

3次元点群の位置合わせによる顔の対称面検出

細木　大祐, 陸, 金, 木村, 大河内, 野添, 中村

第38回日本医用画像工学会大会予稿集 599 - 605 2019年07月

3D-CNN による経時的差分像上の結節状陰影検出査読有り

芳野由利子, 陸慧敏, 金亨燮, 村上誠一, 青木隆敏, 木戸尚治

医用画像情報学会雑誌 ( 医用画像情報学会 ) 36 ( 2 ) 77 - 82 2019年06月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

<p>A temporal subtraction image is obtained by subtracting a previous image, which are warped to match between the structures of the previous image and one of a current image, from the current image. The temporal subtraction technique removes normal structures and enhances interval changes such as new lesions and changes of existing abnormalities from a medical image. However, many artifacts remain on a temporal subtraction image and these can be detected as false positives on the subtraction images. In this paper, we propose a 3D-CNN after initial nodule candidates are detected using temporal subtraction technique. To compare the proposed 3D-CNN, we used 7 model architectures, which are 3D ShallowNet, 3D-AlexNet, 3D-VGG11, 3D-VGG13, 3D-ResNet8, 3D-ResNet20, 3D-ResNet32, with these performance on 28 thoracic MDCT cases including 28 small-sized lung nodules. The higher performance is showed on 3D-AlexNet.</p>

DOI： 10.11318/mii.36.77

その他リンク： https://search.jamas.or.jp/link/ui/2019298826

CONet: A Cognitive Ocean Network 査読有り国際誌

Lu H., Wang D., Li Y., Li J., Li X., Kim H., Serikawa S., Humar I.

IEEE Wireless Communications 26 ( 3 ) 90 - 96 2019年06月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2019 IEEE. The scientific and technological revolution of the Internet of Things has begun in the area of oceanography. Historically, humans have observed the ocean from an external viewpoint in order to study it. In recent years, however, changes have occurred in the ocean, and laboratories have been built on the sea floor. Approximately 70.8 percent of the Earth's surface is covered by oceans and rivers. The Ocean of Things is expected to be important for disaster prevention, ocean resource exploration, and underwater environmental monitoring. Unlike traditional wireless sensor networks, the Ocean Network has its own unique features, such as low reliability and narrow bandwidth. These features will be great challenges for the Ocean Network. Furthermore, the integration of the ocean network with artificial intelligence has become a topic of increasing interest for oceanology researchers. The cognitive ocean network (CONet) will become the mainstream of future ocean science and engineering developments. In this article, we define the CONet. The contributions of the article are as follows: a CONet architecture is proposed and described in detail; important and useful demonstration applications of the CONet are proposed; and future trends in CONet research are presented.

DOI： 10.1109/MWC.2019.1800325

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85061026040&origin=inward

3D registration method based on intensity gradient information for temporal subtraction technique in thoracic MDCT image 査読有り

Miyake, Lu, Kim, Murakami, Aoki, Kido

nternational Journal of Computer Assited Radiology and Surgery S22 - S23 2019年06月

3D convolutional neural networks for automatic detection of lung nodules from temporal subtraction images 査読有り

Yoshino, Lu, Kim, Murakami, Aoki, Kido

International Journal of Computer Assited Radiology and Surgery S71 - S72 2019年06月

Reduction of false-positive polyp detections in CT colonography using multiscale 3D residual networks 査読有り

Uemura, Janne J. Näppi, Lu, Kim, Tachibana, Hironaka, Yoshida

International Journal of Computer Assited Radiology and Surgery S67 - S68 2019年06月

畳み込みニューラルネットワークを用いた指骨CR画像からの骨粗しょう症の識別査読有り

畠野、村上、植村、陸、金、青木

医用画像情報学会雑誌 ( 医用画像情報学会雑誌 ) 36 ( 2 ) 72 - 76 2019年06月

３D-CNNによる経時的差分像上の結節状陰影検出査読有り

芳野、陸、金、村上、青木、木戸

医用画像情報学会雑誌 36 ( 2 ) 77 - 82 2019年06月

The Cognitive Internet of Vehicles for Autonomous Driving 査読有り国際誌

Lu H., Liu Q., Tian D., Li Y., Kim H., Serikawa S.

IEEE Network 33 ( 3 ) 65 - 73 2019年05月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 1986-2012 IEEE. As it combines AI and IoT, autonomous driving has attracted a great deal of attention from both academia and industry because of its benefits to the economy and society. However, ultra-low delay and ultra-high reliability cannot be guaranteed by individual autonomous vehicles with limited intelligence and the existing architectures of the Internet of Vehicles. In this article, based on a cloud/fog-computing pattern and the IoT AI service framework, we propose a cross-domain solution for auto-driving. In contrast to existing studies, which mainly focus on communication technologies, our solution achieves intelligent and flexible autonomous driving task processing and enhances transportation performance with the help of the Cognitive Internet of Vehicles. We first present an overview of the enabling technology and the architecture of the Cognitive Internet of Vehicles for autonomous driving. Then we discuss the autonomous driving Cognitive Internet of Vehicles specifically from the perspectives of what to compute, where to compute, and how to compute. Simulations are then conducted to prove the effect of the Cognitive Internet of Vehicles for autonomous driving. Our study explores the research value and opportunities of the Cognitive Internet of Vehicles in autonomous driving.

DOI： 10.1109/MNET.2019.1800339

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85068858755&origin=inward

Low-light underwater image enhancement for deep-sea tripod 査読有り国際誌

Yujie Li, Jianru Li, Yun Li, H KIM, S SERIKAWA

IEEE Access 7 44080 - 44086 2019年04月

Ensemble 3D residual network(E3D-ResNet) for reduction of false-positive polyp detections in CTcolonography 査読有り

Tomoki Uemura, Janne J. Näppi, Huimin Lu, Hyoungseop Kim, Rie Tachibana, Toru Hironaka, Hiroyuki Yoshida

Proc. SPIE 10950, Medical Imaging 2019 1095013-1 - 1095013-7 2019年03月

対称性解析に基づく３次元データから顔の対称面の検出

細木、陸、金、木村、大河内、野添、中村

信学技報、IE2018-158 285 - 286 2019年03月

対称性解析に基づく口唇裂３次元点群からの正中面抽出法

山田、陸、金、木村、大河内、野添、中村

信学技報、IE2018-158 287 - 290 2019年03月

経時的差分像上の関心領域内の統計的特徴量に基づく結節状陰影の自動検出

金，田中，陸，村上，青木，木戸

電子情報通信学会総合大会 108 - 108 2019年03月

YOROを用いた海中生物のロバスト自律追跡

陸、金

動的画像処理実用化ワークショップ 65 - 66 2019年03月

Clinical Applications of Multidisciplunary Computational Anatomy to Diagnosis

Kido, Hashimoto, Hirano, Mabu, Kim, Kumura, Noriki, Inai, Tachibana

The 5th International Symposium on Mutidisciplinary Computational Anatomy 202 - 212 2019年03月

Recognition of surrounding environment from electric wheelchair videos based on modified YOLOv2 査読有り国際誌

Sakai Y., Lu H., Tan J., Kim H.

Future Generation Computer Systems 92 157 - 161 2019年03月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018 Elsevier B.V. Currently, the aging population is growing in Japan, and the needs for the utilization of welfare equipment are consequently increasing. The electric wheelchair, which is a convenient transportation tool, has rapidly become popular. However, many accidents have occurred when using electric wheelchairs, and the dangers of driving have been noted. Therefore, there is a need to improve accident factors, reduce accidents and improve the convenience of electric wheelchairs by using automation. Environmental recognition is the key technology for developing autonomous electric wheelchairs. Environmental recognition includes self-position estimation, the recognition of sidewalks, crosswalks and traffic lights, and moving object predictions. To solve these problems, this paper develops a system for detecting sidewalks, crosswalks and traffic lights. We develop the object recognition methods using a modified YOLOv2, which is an object detection algorithm that applies convolutional neural networks (CNNs). We detect the object through YOLOv2 and perform processing steps, such as unnecessary bounding box deletion and interpolation. The experimental results demonstrate that the average AUC of the detection rate is 0.587.

DOI： 10.1016/j.future.2018.09.068

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85054855462&origin=inward

CNNを用いた指骨CR画像からの骨粗しょう症の自動識別査読有り

畠野、村上、植村、陸、金、青木

Medical Imaging Technology 37 ( 2 ) 107 - 115 2019年03月

Dust removal from high turbid underwater images using convolutional neural networks 査読有り国際誌

Li Y., Zhang Y., Xu X., He L., Serikawa S., Kim H.

Optics and Laser Technology 110 2 - 6 2019年02月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017 Elsevier Ltd In recent years, underwater image processing has been a focus of many studies. Most underwater image processing technologies are focused on descattering, absorption correction and reflection correction. For a deep sea mining machine, dust seriously affects visual acuity. To correct for this, a two-part dust removal approach is proposed. Underwater red-green minimum channel prior descattering is used to remove fine dust in the first stage. However, the impact of dust streaks on images is always undesirable. Consequently, we propose a further deep convolutional neural-network-based dust removal method. The experimental results show that the proposed method performs better in removing haze-like scatters and dust-like scatters.

DOI： 10.1016/j.optlastec.2017.09.017

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85029658908&origin=inward

Statistical shape model building method using surface registration and model prototype 査読有り国際誌

Li G., Wu J., Xiao Z., Kim H., Ogunbona P.

Optics and Laser Technology 110 234 - 238 2019年02月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017 Elsevier Ltd Automatic segmentation of organs from medical images is indispensable for the computer-assisted medical applications. Statistical Shape Models (SSMs) based scheme has been developed as an accurate and robust approach for extraction of anatomical structures, in which a crucial step is the need to place the sampled points (landmarks) with well corresponding across the whole training set. On the one hand, the correspondence of landmarks is related the quality of shape model. On the other hand, in clinical application some key positions of landmarks should be specified by physicians referring to the anatomic structure. In this paper, we develop an interactive method to build SSM that the landmark distribution can be modified manually without influencing the model quality. We extend an existing remeshing method to produce a model prototype in advance and surface features driven registration to insure the universal optimization of correspondence. The key landmarks are fixed during the prototype generation. We experimented and evaluated the proposed SSM method for lung regions, the deformations of which are considerable large.

DOI： 10.1016/j.optlastec.2017.09.018

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85030029527&origin=inward

Retinal vessel grading in fundoscopic images using GANs 査読有り

Guangxu Li, Ruan, Wan, Yan, Xiao, Kim, Deheng Li:

International Symposium on Artificial Life and Robotics 418 - 422 2019年01月

Segmentation of Spine Region in CT Images Using 3D Edge Detection and 3D Region Growing Technique 査読有り

G. Fu, H. Lu, G. Li, H. Kim, S. Murakami, M. Ueno, T. Terasawa, X. Zhu, T. Aoki

International Symposium on Artificial Life and Robotics 95 - 98 2019年01月

Ensemble 3D residual network (E3D-ResNet) for reduction of false-positive polyp detections in CT colonography 査読有り国際誌

Uemura T., Näppi J., Lu H., Kim H., Tachibana R., Hironaka T., Yoshida H.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE 10950 2019年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2019 SPIE. We developed a novel ensemble three-dimensional residual network (E3D-ResNet) for the reduction of false positives (FPs) in computer-aided detection (CADe) of polyps on CT colonography (CTC). To capture the volumetric multiscale information of CTC images, each polyp candidate was represented with three different sizes of volumes of interest (VOIs), which were enlarged to a common size and were individually subjected to three 3D-ResNets. These 3D-ResNets were trained to calculate three polyp-likelihood probabilities, p1, p2 and p3, corresponding to each input VOI. The final polyp likelihood, p, was obtained as the maximum of p1, p2 and p3. We compared the classification performance of the E3D-ResNet with that of a non-ensemble 3D-ResNet, ensemble 2D-ResNet, and ensemble of 2D-and 3D-convolutional neural network (CNN) models. All models were trained and evaluated with 21,021 VOIs of polyps and 19,557 VOIs of FPs that were sampled with data augmentation from the CADe detections on the CTC data of 20 patients. We evaluated the classification performance of the models with receiver operating characteristics (ROC) analysis using cross-validation, where the area under the ROC curve (AUC) was used as the figure of merit. Preliminary results showed that AUC value (0.98) of the E3D-ResNet was significantly higher than that of the reference models (P < 0.001), indicating that the E3D-ResNet has the potential of substantially reducing the FPs in CADe of polyps on CTC.

DOI： 10.1117/12.2512173

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85068173052&origin=inward

Low-Light Underwater Image Enhancement for Deep-Sea Tripod 査読有り国際誌

Li Y., Li J., Li Y., Kim H., Serikawa S.

IEEE Access 7 44080 - 44086 2019年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2013 IEEE. To monitor the sedimentary process and morphological evolution in the South China Sea, free-ascending deep-sea tripod (FDT) has been developed. This FDT was equipped with a deep-sea camera and landed on the sea floor at a depth of 2100 m. Although the FDT was equipped with an artificial light, the battery capacity limited the duration and intensity of light. Therefore, enhancing such low-illumination images to obtain clear visual effects is an important advancement for analyzing the geological evolution process. In this paper, an adaptive bright-color channel-based low-light underwater image-enhancement method and a denoising method are proposed to enhance such images and remove noise and artifacts. The experimental results demonstrated that the proposed method outperformed state-of-the-art methods.

DOI： 10.1109/ACCESS.2019.2897691

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85064689140&origin=inward

CNNを用いた指骨CR画像からの骨粗しょう症の自動識別査読有り

畠野和裕, 村上誠一, 植村知規, 陸慧敏, 金亨燮, 青木隆敏

Ｍｅｄｉｃａｌ　Ｉｍａｇｉｎｇ　Ｔｅｃｈｎｏｌｏｇｙ ( 日本医用画像工学会 ) 37 ( 2 ) 107 - 115 2019年01月

担当区分：最終著者記述言語：日本語掲載種別：研究論文（学術雑誌）

<p>骨のおもな疾患として，骨粗しょう症が挙げられる．骨粗しょう症に対する画像診断は有効であるが，医師の負担増加や経験差による診断結果のばらつき，病変部の未検出等が懸念されている．そこで本稿では，指骨computed radiography（CR）画像から骨粗しょう症の識別を行い，医師に提示するための診断支援手法を提案する．提案手法では，畳み込みニューラルネットワークの一種である，Residual Network （ResNet）を用いた識別器を構築し，骨粗しょう症有無の識別を行う．ResNetへの入力画像には，CR画像から生成した画像を用いる．本稿では，3種類の入力画像を提案し，各画像で学習および，識別の評価を行う．実験では，101症例に対し提案手法を適用し，receiver operating characteristics（ROC）曲線上のarea under the curve（AUC）値を用いて評価したところ，最大で0.931という結果を得た．</p>

DOI： 10.11409/mit.37.107

その他リンク： https://ci.nii.ac.jp/naid/130007633096

畳み込みニューラルネットワークを用いた指骨CR画像からの骨粗しょう症の識別査読有り

畠野和裕, 村上誠一, 植村知規, 陸慧敏, 金亨燮, 青木隆敏

医用画像情報学会雑誌 ( 医用画像情報学会 ) 36 ( 2 ) 72 - 76 2019年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

<p>Osteoporosis is known as one of the main diseases of bone. Although image diagnosis for osteoporosis is effective, there are concerns about increased burden of radiologists associated with diagnostic imaging, uneven diagnostic results due to experience difference, and undetected lesions. Therefore, in this study, we propose a diagnosis supporting method for classifying osteoporosis from phalanges computed radiography images and presenting classification results to physicians. In the proposed method, we construct classifiers using convolution neural network and classify normal cases and abnormal cases about osteoporosis. In our experiments, two kinds of CNN models were constructed using input images generated from 101 cases of CR images and evaluated using Area Under the Curve（AUC）value on Receiver Operating Characteristics（ROC）curve. Finaly, AUC of 0.995 was obtained.</p>

DOI： 10.11318/mii.36.72

その他リンク： https://ci.nii.ac.jp/naid/130007668726

Automatic extraction of abnormalities on temporal CT subtraction images using sparse coding and 3D-CNn 査読有り国際誌

Koizumi Y., Miyake N., Lu H., Kim H., Murakami S., Aoki T., Kido S.

International Conference on Control, Automation and Systems 2018-October 1468 - 1471 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. In recent years, the proportion of deaths from cancer tends to increase in Japan, especially the number of deaths from lung cancer is increasing. CT device is effective for early detection of lung cancer. However, there is concern that an increase in burden on doctors will be caused by high performance of CT improving. Therefore, by presenting the “second opinion” by the CAD system, it reduces the burden on the doctor. In this paper, we develop a CAD system for automatic detection of lesion candidate regions such as lung nodules or ground glass opacity (GGO) from 3D CT images. Our proposed method consists of three steps. In the first step, lesion candidate regions are extracted using temporal subtraction technique. In the second step, the image is reconstructed by sparse coding for the extracted region. In the final step, 3D Convolutional Neural Network (3D-CNN) identification using reconstructed images is performed. We applied our method to 51 cases and True Positive rate (TP) of 79.81 % and False Positive rate (FP) of 37.65 % are obtained.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060466308&origin=inward

ROI-based fully automated liver registration in multi-phase CT images 査読有り国際誌

Saito K., Lu H., Kim H., Kido S., Tanabe M.

International Conference on Control, Automation and Systems 2018-October 645 - 649 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. In this paper, we propose a registration method for fully automated liver tumor detection. Multiple phases CT is used for the detection of the liver tumor because multiple phase CT can give different characteristic features of lesions for each time phases. Registration accuracy is important when obtaining image features from multiple time phases. However, since each time phases have different image density characteristics, therefore registration of multi-phase CT is a challenging task. In this paper, we propose a robust initial alignment method independent of changing image density features in each time phase, and deformable registration method with region of interests (ROI) as liver region extracted by U-Net. Our proposed method is evaluated on 15 patient image sets. This method is applied to the early arterial phase and the equilibrium phase to registries. Experimental results show that segmentation of early arterial phase is 83% and registration is 93% accuracy.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060464031&origin=inward

Registration of phalange region from CR images based on genetic algorithm 査読有り国際誌

Kawagoe K., Murakami S., Lu H., Tan J., Kim H., Aoki T.

International Conference on Control, Automation and Systems 2018-October 1464 - 1467 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. In Japan, the number of patients with osteoporosis and rheumatoid arthritis is increasing. Image diagnosis using CR images is effective for osteoporosis and rheumatoid arthritis. Development of a CAD system is important for reducing burdens on doctors. In this paper, we propose an automatic registration algorithm in the CAD system. In the proposed method, the genetic algorithm is used to register bone regions between identical parts of the same subject with different time series. In the experiment, the proposed method is applied to 176 bone area, and 98.14 % of TPR, 1.85 % of FPR are obtained respectively. Even when the area difference is used as the fitness of the genetic algorithm, it has cross-correlation and positioning accuracy equivalent to mutual information.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060451078&origin=inward

Extraction of median plane from facial 3D point cloud based on symmetry analysis using ICP algorithm 査読有り国際誌

Yamada S., Lu H., Tan J., Kim H., Kimura N., Okawachi T., Nozoe E., Nakamura N.

International Conference on Control, Automation and Systems 2018-October 1347 - 1350 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. Cleft lip is a kind of congenital facial morphological abnormality. In the clinical field of cleft lip, it is necessary to analyze symmetric shape. However, there is no method to analyze the cleft lip technique based on symmetrical viewpoints. On the other hand, in our previous method to find a symmetric axis using a 2D image, since the middle line is extracted only from the front view of the face moire image. There was a problem that low accuracy was obtained by slight rotation of the face and it was not possible to consider 3D information. In this paper, we propose a method to extract the median plane of the face by analyzing based on bilateral symmetry by using 3D point cloud on the face of front. By extracting the median plane, we believe that not only surgical assistance of doctor be possible but also become a clue to development of simulation software which is the end goal.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060498535&origin=inward

Enhancement of bone metastasis from CT images based on salient region feature registration 査読有り国際誌

Sato S., Lu H., Kim H., Murakami S., Ueno M., Terasawa T., Aoki T.

International Conference on Control, Automation and Systems 2018-October 1329 - 1332 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. In recent years, the development of the computer-aided diagnosis (CAD) systems to support radiologist is attracting attention in medical research field. One of them is temporal subtraction technique. It is a technique to generate images emphasizing temporal changes in lesions by performing a differential operation between current and previous image of the same subject. In this paper, we propose an image registration method for image registration of current and previous image, to generate temporal subtraction images from CT images and enhanced bone metastasis region. The proposed registration method is composed into three main steps: i) automatic segmentation of the region of interest (ROI) using position information of the spine based on biology, ii) use global image matching to select pairs from previous and current image, and iii) final image matching based on salient region feature. We perform registration technique on synthetic data and confirm usefulness of the proposed method. Furthermore, radiologist conduct comparative experiments without and with temporal subtraction images created by proposed method. As a result, they show high reading performance by using temporal subtraction images.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060475609&origin=inward

Detection of phalange region based on U-Net 査読有り国際誌

Hatano K., Murakami S., Lu H., Tan J., Kim H., Aoki T.

International Conference on Control, Automation and Systems 2018-October 1338 - 1342 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. Osteoporosis is one of the famous bone diseases. It is a major cause of deteriorating the quality of life, and early detection and early treatment are becoming socially important. Visual screening using Computed Radiography (CR) images is effective for diagnosis of osteoporosis, but there are problems of increasing the burden on doctors, variation in diagnostic results due to differences in experiences of doctors, and undetected lesions. In order to solve this problem, we are working on a computer-aided diagnosis (CAD) system for osteoporosis. In this paper, we propose segmentation methods of the phalange region from the phalangeal CR images as a preprocessing of classification of osteoporosis. In the proposed method, we construct a segmentation model using U-Net, which is a type of deep convolution neural network (DCNN). The proposed method was applied to input images generated from CR images of 101 patients with both hands, and evaluated using the Intersection over Union (IoU) values. The result was 0.914 in IoU.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060468143&origin=inward

Detection of grasping position from video images based on SSD 査読有り国際誌

Kitayama T., Lu H., Li Y., Kim H.

International Conference on Control, Automation and Systems 2018-October 1472 - 1475 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. Recently, consistent container transportation of roads and ships is mainstream of international freight transport. Because of various factors, automation of cargo handling work is required at the container terminal. Various causes are decrement of future labor force population by an increasing trend of container moving amount and declining birthrate and aging population. Therefore, this study presents the relative position of hanger and container measurement technology using Single Shot Multibox Detector (SSD) for the purpose of improvement of cargo handling work efficiency and unmanned container terminal. In the case of undetected by SSD, it will be detected using AKAZE feature. The proposed method is applied to 368 images of container gripping taken by a camera installed in a container crane. As a result, Interaction of Union (IoU) targeted for container gripping is 87.79%, and a detection rate is 94.57%.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060451358&origin=inward

Detection of abnormal shadows on temporal subtraction images based on multi-phase CNN 査読有り国際誌

Nagao M., Miyake N., Yoshino Y., Lu H., Kim H., Murakami S., Aoki T., Kido S.

International Conference on Control, Automation and Systems 2018-October 1333 - 1337 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. Recently, visual screening based on CT images become useful tools in the medical fields. However, due to the large number of images and the complexity of the image processing algorithms, image processing technique for the high screening quality is still required. To overcome this problem, some computer aided diagnosis (CAD) algorithms are proposed. Cancer is a leading cause of death both in Japan and worldwide. Detection of cancer region in CT images is the most important task to early detection and early treatment. We have designed and developed a framework combining machine learning based on multi-phase convolutional neural networks (CNN) and temporal subtraction techniques based on non-rigid image registration algorithm. Our main classification method can be built into three main steps; i) preprocessing for image segmentation, ii) image matching for registration, and iii) classification of abnormal regions based on machine learning algorithms. We performed our proposed technique to 25 thoracic MDCT sets and obtained true positive rates of 93.55%, false positive rates of 10.93 /case.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060434756&origin=inward

Swallowing motion analyzing from dental MR imaging based on AKAZE and particle filter algorithm 査読有り国際誌

Suetani K., Lu H., Tan J., Kim H., Tanaka T., Kitou S., Morimoto Y.

International Conference on Control, Automation and Systems 2018-October 1343 - 1346 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. In recent years, dysphagia is problem among elderly people. Therefore, it is necessary to accurately evaluate swallowing function in order to prevent swallowing disorder beforehand or to detect it early. And it is considered that evaluation of swallowing function using Magnetic Resonance Imaging (MRI) is useful. In order to accurately analyzing of the swallowing motion using a computer aided diagnosis (CAD) system on MR imaging, automatic extraction of the esophagus region, which is a region of interest by the image analysis method, is required. Extraction of the spinal region is required as a preliminary step of the esophagus region extraction. Therefore, in this paper, we develop an analysis method of swallowing movement by three steps of extraction of spinal region, extraction of esophageal region, and analysis of swallowing movement. As an analytical method of swallowing movement, we emphasize the liquid part at the time of swallowing movement using an emphasis map, then follow the liquid tracing by using the AKAZE feature quantity and the particle filter algorithm, and analyze the swallowing motion.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060441189&origin=inward

Wide residual networks for semantic segmentation 査読有り国際誌

Nakayama Y., Lu H., Li Y., Kim H.

International Conference on Control, Automation and Systems 2018-October 1476 - 1480 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© ICROS. In the task of object recognition, convolutional neural networks (CNNs) have achieved high performance. In addition, these CNNs are also applied to the field of semantic image segmentation. However, applying the classification models to semantic segmentation tasks has a problem, lack of global context and reduction in resolution. In this work, we propose global context module and high resolution path in order to solve above problems. By simply combining them with an existing classification model (wide residual networks), our methods yield high-accuracy segmentation models. Our proposed approaches produce competitive results, the mean intersection over union (IoU) 67.6% and global accuracy 91.1%, on CamVid test set.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060445374&origin=inward

Extraction of GGO Candidate Regions on Thoracic CT Images using SuperVoxel-Based Graph Cuts for Healthcare Systems 査読有り国際誌

Lu H., Kondo M., Li Y., Tan J., Kim H., Murakami S., Aoki T., Kido S.

Mobile Networks and Applications 23 ( 6 ) 1669 - 1679 2018年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018, Springer Science+Business Media, LLC, part of Springer Nature. In this paper, we propose a method to reduce artifacts on temporal difference images by improving the conventional method using a non-rigid registration method for ground glass opacification (GGO), which is light in concentration and difficult to detect early. In this method, global matching, local matching, and 3D elastic matching are performed on the current image and past image, and an initial temporal difference image is generated. After that, we use an Iris filter, which is the gradient vector concentration degree filter, to determine the initial GGO candidate regions and perform segmentation using SuperVoxel and Graph Cuts in which a superpixel is extended to three dimensions for each region of interest. For each extracted region, a support vector machine (SVM) is used to reduce the over-segmentation. Finally, in the method that greatly reduces artifacts other than the remaining GGO candidate regions, Voxel Matching is applied to generate the final temporal difference image, emphasizing the GGO regions while reducing the artifact. The resulting ratio of artifacts to lung volume is 0.101 with an FWHM of 28.3, which is an improvement over the conventional method and shows the proposed method’s effectiveness.

DOI： 10.1007/s11036-018-1111-2

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85053445997&origin=inward

細胞領域の論理積を用いた蛍光顕微鏡画像からの血中循環がん細胞の自動検出査読有り

辻、陸、タン、金、米田、田中

医用画像情報学会誌 34 ( 4 ) 151 - 155 2018年12月

Segmentation of Spinal Canal Region in CT Images using 3D Region Growing Technique 査読有り国際誌

Fu G., Lu H., Tan J., Kim H., Zhu X., Lu J.

2018 International Conference on Information and Communication Technology Robotics, ICT-ROBOT 2018 2018年11月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2018 IEEE. Tumor of spinal cord (Spinal Cord Neoplasms (SCN)) is considered as one of the life threatening diseases that causes death. Early detection of the SCN plays an important role in the management of the lesions. To analyze the treatment, it is necessary to segment the spinal canal based on accurate three-dimensional image processing technique. This paper presents a segmentation algorithm based on 3D region growing for extracting spinal canal from CT images with high accuracy. Intersection over union (IoU) is used to compare the results of segmentation with the manual segmentation results. In the experiment, the proposed method was tested on 3373 CT slices of 10 patients. The proposed method has an average accuracy of 0.7732 and a variance of 0.0061. Satisfactory results have been achieved rapidly, which demonstrates the effectiveness and superiority of the proposed method.

DOI： 10.1109/ICT-ROBOT.2018.8549913

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060055563&origin=inward

Object Detection on Video Images Based on R-FCN and GrowCut Algorithm 査読有り国際誌

Mouri K., Lu H., Tan J.K., Kim H.

2018 International Conference on Information and Communication Technology Robotics, ICT-ROBOT 2018 2018年11月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Since the declining birthrate and the aging of society, there is concern about the labor shortage in Japan. There is a movement to compensate for the labor shortage by automation of factories by robots. Automation technique is wildly promoted in logistics industry, while there is few studies in objects picking. To solve this issue, we develop an image detection scheme for robotics picking from a video image. It is difficult to recognize and grasp different types of objects in robot vision field. Therefore, in the proposed method, object detection and object recognition method are proposed using Region-based Fully Convolutional Networks that is a type of object detection using deep learning. After detecting the object individually, final target object can select by applying the GrowCut algorithm. As a result, we achieve 0.6773 of the average precision and 0.6395 of Intersection over Union as the segmentation result respecively.

DOI： 10.1109/ICT-ROBOT.2018.8549879

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060010049&origin=inward

実数値GAに基づく指骨CR画像の位置合わせ

川越、村上、陸、タン、金、青木

第38回医療情報学連合大会抄録集 388 - 391 2018年11月

CT temporal subtraction method for detection of sclerotic bone metastasis in the thoracolumbar spine 査読有り国際誌

Ueno M., Aoki T., Murakami S., Kim H., Terasawa T., Fujisaki A., Hayashida Y., Korogi Y.

European Journal of Radiology 107 54 - 59 2018年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018 Elsevier B.V. Purpose: To assess the effectiveness of a CT temporal subtraction (TS) method on radiologists’ performance in sclerotic metastasis detection in the thoracolumbar spine. Materials and methods: 20 pairs (current and previous CTs) of standard-dose CT and their TS images in patients with sclerotic bone metastasis and 20 pairs (current and previous CTs) of those in patients without bone metastasis were used for an observer performance study. A total of 135 lesions were identified as the reference standard of actionable lesions (sclerotic metastasis newly appeared or increased in size or in attenuation). 4 attending radiologists and 4 radiology residents participated in this observer study. Ratings and locations of “lesions” determined by the observers were utilized for assessing the statistical significance of differences between radiologists’ performances without and with the CT-TS images in JAFROC analysis. The statistical significance of differences in the reviewing time was determined by a two-tailed paired t-test. Results: The average figure-of-merit (FOM) values for all but one radiologist increased to a statistically significant degree, from 0.856 without the CT-TS images to 0.884 with the images (P =.037). The average sensitivity for detecting the actionable lesions was improved from 60.7 % to 72.5% at a false-positive rate of 0.15 per case by use of the CT-TS images. The average reading time with CT-TS images was significantly shorter than that without (150.6 s vs. 166.5 s, P =.004). Conclusion: The use of CT-TS would improve the observer performance for the detection of the sclerotic bone metastasis in the thoracolumbar spine.

DOI： 10.1016/j.ejrad.2018.07.017

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85051670298&origin=inward

Statistical shape model generation using K-means clustering 査読有り国際誌

Wu J., Li G., Lu H., Kim H.

ACM International Conference Proceeding Series 207 - 211 2018年09月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2018 Association for Computing Machinery. Statistical shape models (SSMs) is a robust and efficient method in medical image segmentation. In this paper, a novel landmark corresponding method based on k-means clustering and demons registration is proposed to train a 3-D statistical shape model with higher quality. The k-means clustering method is performed on the original geometric surface to obtain a simplified surface as standard set of landmarks to find correspondent landmarks on each mapped spherical surface obtained from demon registration in the training set. Twenty cases of left lung and right lung regions in thoracic MDCT images are used in the experiment to build two SSMs. Performance evaluation results show that SSMs generated by the proposed method achieve better generalization ability and specificity while maintaining the same compactness and accuracy of segmentation as those reported by state-of-the-art methods.

DOI： 10.1145/3277453.3277467

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85060643606&origin=inward

R-FCNとGrowCutを用いたボールペンの検出

毛利、陸、金

産業応用工学会全国大会2018講演論文集 85 - 86 2018年09月

Object Detection on Video Images Based on GrowCut Algorithm 査読有り国際誌

Mouri, Lu, Tan, Kim

ICT-Robot 2018年09月

Activity representation by SURF-based templates 査読有り国際誌

Ahad M., Tan J., Kim H., Ishikawa S.

Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization 6 ( 5 ) 573 - 583 2018年09月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018, © 2018 Informa UK Limited, trading as Taylor & Francis Group. This paper presents a method to recognise actions, which are overlapping and multi-dimensionalities. A spatio-temporal representation is illustrated on local interest points to compute global features. Motion history image (MHI) is computed and motion overwriting the motion overwriting problem of the MHI. The main contribution of this paper is that it demonstrates a higher discriminative ability of various complex actions when compared to the other MHI-based approaches. It selects local interest feature points to capture motion information using Speeded-Up Robust Features (SURF). These key interest points are exploited to compute gradient-based optical flow into four channels. RANSAC is exploited to remove outliers. It incorporates frame-subtracted accumulated image so that we can mask out points that are not required. Afterwards, feature vectors are computed based on moments. Actions are recognised by employing a nearest neighbour classifier and leave-one-out cross-validation partitioning scheme. The proposed method provides satisfactory recognition rates over several other approaches for some challenging actions in outdoor scenes.

DOI： 10.1080/21681163.2017.1298472

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85016247850&origin=inward

FDCNet: filtering deep convolutional network for marine organism classification 査読有り国際誌

Lu H., Li Y., Uemura T., Ge Z., Xu X., He L., Serikawa S., Kim H.

Multimedia Tools and Applications 77 ( 17 ) 21847 - 21860 2018年09月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017, Springer Science+Business Media New York. Convolutional networks are currently the most popular computer vision methods for a wide variety of applications in multimedia research fields. Most recent methods have focused on solving problems with natural images and usually use a training database, such as Imagenet or Openimage, to detect the characteristics of the objects. However, in practical applications, training samples are difficult to acquire. In this study, we develop a powerful approach that can accurately learn marine organisms. The proposed filtering deep convolutional network (FDCNet) classifies deep-sea objects better than state-of-the-art classification methods, such as AlexNet, GoogLeNet, ResNet50, and ResNet101. The classification accuracy of the proposed FDCNet method is 1.8%, 2.9%, 2.0%, and 1.0% better than AlexNet, GooLeNet, ResNet50, and ResNet101, respectively. In addition, we have built the first marine organism database, Kyutech10K, with seven categories (i.e., shrimp, squid, crab, shark, sea urchin, manganese, and sand).

DOI： 10.1007/s11042-017-4585-1

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85015699675&origin=inward

BrainNets: Human Emotion Recognition Using an Internet of Brian Things Platform 査読有り国際誌

Lu H., Kim H., Li Y., Zhang Y.

2018 14th International Wireless Communications and Mobile Computing Conference, IWCMC 2018 1313 - 1316 2018年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2018 IEEE. Human wearable helmet is a useful tool for monitoring the status of miners in the mining industry. However, there is little research regarding human emotion recognition in an extreme environment. In this paper, an emotional state evoked paradigm is designed to identify the brain area where the emotion feature is most evident. Next, the correct electrode position is determined for the collection of the negative emotion by the electroencephalograph (EEG) based on the international 10-20 system of electrode placement. And then, a fusion algorithm of the anxiety level is proposed to evaluate the person's mental state using the θ, α, and β rhythms of an EEG. Experiments demonstrate that the position Fp2 is the best electrode position for obtaining the anxiety level parameter. The most visible EEG changes appear within the first two seconds following stimulation. The amplitudes of the θ rhythm increase most significantly in the negative emotional state.

DOI： 10.1109/IWCMC.2018.8450382

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85053937610&origin=inward

Automatic detection of cell regions in microscope images based on BFED algorithm 査読有り国際誌

Nakamichi K., Lu H., Kim H., Yoneda K., Tanaka F.

ACM International Conference Proceeding Series 38 - 42 2018年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2018 Association for Computing Machinery. Circulating tumor cells (CTC) attract attention as a biomarker that can evaluate cancer metastasis and therapeutic effects. The CTC exists in the blood of cancer patients, so pathologists analyze blood by using a fluorescence microscope. However, manual analysis by pathologists is hard-work since the number of CTC to substances contained in the blood is very few and the cell regions are often unclear depending on shooting environments. In addition, there are few studies on automatic identification of CTC. In this paper, we develop an automatic detection method of cell regions in microscope images based on bacterial foraging-based edge detection (BFED) algorithm to analyze CTC. In the first step, we detect the initial cell regions by BFED algorithm. Second, we identify whether the region is a single cell or multiple cells come in connect with other cell(s) by SVM. Third, when a cell is connected with other one, we separate the connecting cells by branch and bound algorithm and obtain the final cell regions. We applied our proposed method to 1680 microscopy images (6 cases). The experimental results demonstrate that the proposed method has a true positive rate of 93.9% and a false positive 1.29 /case.

DOI： 10.1145/3278229.3278232

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85058561021&origin=inward

Environment recognition for electric wheelchair based on YOLOv2 査読有り国際誌

Sakai Y., Lu H., Tan J., Kim H.

ACM International Conference Proceeding Series 112 - 117 2018年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2018 Association for Computing Machinery. At present, the aging population is growing in Japan. Along with that, the need for the utilization of welfare equipment is increasing. Electric wheelchair, a convenient transportation tool, is popularized rapidly. However, many accidents have occurred by using electric wheelchair, and the dangers for driving are pointed out. Therefore, it needs to improve accident factors, reduce accidents and improve the convenience of electric wheelchair by automation. Environmental recognition is the key technology for developing autonomous electric wheelchair. Environmental recognition includes self-position estimation, recognition of sidewalks, crosswalks, traffic lights, and moving object prediction, etc. In order to solve these problems, this paper describes a system for the detection of sidewalks, crosswalks and traffic lights. We develop the object recognition methods using a modified YOLOv2 that is one of object detection algorithms applying convolutional neural networks (CNN). We detect the object through YOLOv2 and perform processing such as unnecessary bounding box deletion and interpolation. The experimental results demonstrate that the area under the curve (AUC) of the detection rate is 0.620.

DOI： 10.1145/3278229.3278231

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85058546828&origin=inward

Preface 招待有り査読有り国際誌

Kim H.

ACM International Conference Proceeding Series 2018年08月

Motor anomaly detection for unmanned aerial vehicles using reinforcement learning 査読有り国際誌

Lu H., Li Y., Mu S., Wang D., Kim H., Serikawa S.

IEEE Internet of Things Journal 5 ( 4 ) 2315 - 2322 2018年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017 IEEE. Unmanned aerial vehicles (UAVs) are used in many fields including weather observation, farming, infrastructure inspection, and monitoring of disaster areas. However, the currently available UAVs are prone to crashing. The goal of this paper is the development of an anomaly detection system to prevent the motor of the drone from operating at abnormal temperatures. In this anomaly detection system, the temperature of the motor is recorded using DS18B20 sensors. Then, using reinforcement learning, the motor is judged to be operating abnormally by a Raspberry Pi processing unit. A specially built user interface allows the activity of the Raspberry Pi to be tracked on a Tablet for observation purposes. The proposed system provides the ability to land a drone when the motor temperature exceeds an automatically generated threshold. The experimental results confirm that the proposed system can safely control the drone using information obtained from temperature sensors attached to the motor.

DOI： 10.1109/JIOT.2017.2737479

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85028968216&origin=inward

Automatic road detection system for an airland amphibious car 査読有り

Li, Lu, Nakayama, Kim, Serikawa

Future Generation Computer Systems 85 51 - 59 2018年08月

Automatic road detection system for an air–land amphibious car drone 査読有り国際誌

Li Y., Lu H., Nakayama Y., Kim H., Serikawa S.

Future Generation Computer Systems 85 51 - 59 2018年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018 Elsevier B.V. In recent years, unmanned aerial vehicle (UAV) technologies have rapidly developed. Drones, which are one type of UAV, are used in many industrial fields, such as photography, delivery and agriculture. However, a commercial drone can fly for only approximately 20 min on one charge. Furthermore, drones are prohibited from flying in some areas, and cannot be operated in bad weather. Due to the development of drone technologies, we must reduce energy consumption and achieve long-range movement. To overcome these limitations, we develop a new air–land amphibious car drone that can fly and requires less power consumption in land mode; this extends the range of mobility of the drone. Moreover, land mode can be used to pass through restricted areas or bad weather conditions by sliding. Furthermore, we develop a Convolutional Neural Network (CNN)-based algorithm for detecting the road in a captured scene. To more accurately segment the road region based on images from the equipped camera of the drone, we propose atrous spatial pyramid pooling (ASPP) ResNet blocks, instead of Resblocks, which were proposed by DeepLab. The experimental results demonstrate that the proposed method improves the pixel accuracy (PA) to 85.6% and achieves a mean Intersection over Union (mIoU) of 55.8%.

DOI： 10.1016/j.future.2018.02.036

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85043254725&origin=inward

ResNetを用いた指骨CR画像からの骨粗しょう症の自動識別

畠野、村上、植村、陸、タン、金、青木

第37回日本医用画像工学会大会予稿集 238 - 244 2018年07月

3次元Residual networks(ResNets)を用いた大腸CADにおける偽陽性陰影の識別

植村、陸、金、橘、弘中、Nappi、吉田

第37回日本医用画像工学会大会予稿集 436 - 442 2018年07月

CNNによる胸部CT画像からの経時的差分画像上の異常陰影の検出

長尾、植村、三宅、陸、タン、金、村上、青木、平野、木戸

第37回日本医用画像工学会大会予稿集 404 - 410 2018年07月

Image Registration for Detection of Sclerotic Bone Metastasis in CT Images 査読有り国際誌

Kim, Lu, Tan, Murakami, Ueno, Terasawa, TAoki

40th International Conference of the IEEE Engineering in Medicine and Biology Society 2018年07月

3D deelp residual convolutional networks for computer-aided detection of polyos in CT colonography 査読有り

Uemura, Lu, Kim, Tachibana, Hironaka, Nappi, Yoshida

International Journal of Computer Assited Radiology and Surgery 94 - 95 2018年06月

Low illumination underwater light field images reconstruction using deep convolutional neural networks 査読有り国際誌

Lu H., Li Y., Uemura T., Kim H., Serikawa S.

Future Generation Computer Systems 82 142 - 148 2018年05月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2018 Elsevier B.V. Underwater optical images are usually influenced by low lighting, high turbidity scattering and wavelength absorption. To solve these issues, a great deal of work has been performed to improve the quality of underwater images. Most of them use the high-intensity LEDs for lighting to obtain the high contrast images. However, in high turbidity water, high-intensity LEDs cause strong scattering and absorption. In this paper, we propose a light field imaging approach for solving underwater imaging problems in a low-intensity light environment. As a solution, we tackle the problem of de-scattering from light field images by using deep convolutional neural networks with depth estimation. Furthermore, a spectral characteristic-based color correction method is used for recovering the color reduction. Experimental results show the effectiveness of the proposed method by challenging real-world underwater imaging.

DOI： 10.1016/j.future.2018.01.001

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85040510117&origin=inward

Automatic identification of bone erosions in rheumatoid arthritis from hand radiographs based on deep convolutional neural network 査読有り

Murakami, Hatano, Tan, Kim, Aoki

International Journal of Multimedia Tools and Applications 77 ( 9 ) 10921 - 10937 2018年05月

Active contour model-based segmentation algorithm for medical robots recognition 査読有り

Yujie Li, Yun Li, Kim, Serikawa

Journal of Multimedia Tools and Applications 77 ( 9 ) 10485 - 10500 2018年05月

Detection of lung carcinoma with predominant ground-glass opacity on CT using temporal subtraction method 査読有り

Terasawa, Aoki, Murakami, Kim, Fujii, Kobayashi, Chihara, Hayashida, Korogi

European Radiology 28 ( 4 ) 1594 - 1599 2018年04月

Brain Intelligence: Go beyond Artificial Intelligence 査読有り国際誌

Lu H., Li Y., Chen M., Kim H., Serikawa S.

Mobile Networks and Applications 23 ( 2 ) 368 - 375 2018年04月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017, Springer Science+Business Media, LLC. Artificial intelligence (AI) is an important technology that supports daily social life and economic activities. It contributes greatly to the sustainable growth of Japan’s economy and solves various social problems. In recent years, AI has attracted attention as a key for growth in developed countries such as Europe and the United States and developing countries such as China and India. The attention has been focused mainly on developing new artificial intelligence information communication technology (ICT) and robot technology (RT). Although recently developed AI technology certainly excels in extracting certain patterns, there are many limitations. Most ICT models are overly dependent on big data, lack a self-idea function, and are complicated. In this paper, rather than merely developing next-generation artificial intelligence technology, we aim to develop a new concept of general-purpose intelligence cognition technology called “Beyond AI”. Specifically, we plan to develop an intelligent learning model called “Brain Intelligence (BI)” that generates new ideas about events without having experienced them by using artificial life with an imagine function. We will also conduct demonstrations of the developed BI intelligence learning model on automatic driving, precision medical care, and industrial robots.

DOI： 10.1007/s11036-017-0932-8

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85029688332&origin=inward

Non-uniform de-Scattering and de-Blurring of Underwater Images 査読有り国際誌

Li Y., Lu H., Li K., Kim H., Serikawa S.

Mobile Networks and Applications 23 ( 2 ) 352 - 362 2018年04月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017, Springer Science+Business Media, LLC. Optical underwater images often demonstrate low contrast, heavy scatter, and color distortion. Contrast enhancement methods have been proposed to solve these issues. However, such methods typically do not consider high-level inhomogeneous scatter removal and do not focus on real-scene color restoration. We proposed a hierarchical transmission fusion method and a color-line ambient light estimation method for image de-scattering from a single input image. Our proposed method can be summarized into three steps. Firstly, we take the dark channel as prior information to estimating the preliminary transmission and ambient light. In the second step, we then use color lines to estimate the refined ambient light in selected patches. The refined transmission is obtained by hierarchical transmission maps using maximum local energy-based fusion at different turbidity levels. We then use a joint normalized filter to obtain the final transmission. Finally, a chromatic color correction method and de-blurring algorithm are used to recover the scene color. Experimental results demonstrate that the accurate estimation of the depth map and ambient light by the proposed method can recover visually appealing images with sharp details.

DOI： 10.1007/s11036-017-0933-7

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85029692278&origin=inward

DCNNによる指骨CR画像からの骨粗しょう症の自動識別査読有り

畠野、村上、植村、陸、タン、金、青木

Medical Imaging Technology 36 ( 2 ) 90 - 95 2018年03月

経時的差分像技術を用いたGGO陰影の強調表示法

金、近藤、陸、村上、寺澤、青木、平野、木戸

第10回呼吸機能イメージング研究会学術集会抄録集 45 - 45 2018年02月

Detection of a specific moving object from head-mounted camera images 査読有り国際誌

Ishitobi K., Tan J., Kim H., Ishikawa S.

SII 2017 - 2017 IEEE/SICE International Symposium on System Integration 2018-January 817 - 822 2018年02月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 IEEE. In this paper, a method is proposed for detecting and tracking a specific moving object (e.g., a bus) on the road from images of a camera attached to the head of a user, aiming at developing a system to support daily lives of visually impaired people. The proposed method traces feature points on the images, extracts a moving object region, and detects a bus by applying Haar-like feature and random trees to the region. The effectiveness of the proposed method is shown experimentally.

DOI： 10.1109/SII.2017.8279323

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85050871522&origin=inward

A unified action recognition framework 査読有り

Amine, Tan, Kim, Ishikawa

he 23rd International Syposium on Artificial Lefe and Robotics 57 - 62 2018年01月

Bone erosions detection on hand CR images based on DCNN 査読有り

Murakami, Hatano, Lu, Tan, Kim, Aoki

The 23rd International Syposium on Artificial Lefe and Robotics 357 - 360 2018年01月

A Diffeomorphic Demons Approach to Statistical Shape Modeling 査読有り国際誌

Li G., Wu J., Xiao Z., Lu H., Kim H., Ogunbona P.

Studies in Computational Intelligence 752 123 - 131 2018年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© Springer International Publishing AG 2018. Automatic segmentation of organs from medical images is indispensable for the applications of computer-aided diagnosis (CAD) and computer-assisted surgery (CAS). Statistical Shape Models (SSMs) based scheme have been proved as the accurate and robust methods for extraction of anatomical structures. A key step of this approach is the need to place the sampled points(landmarks) with correspondence across the training set. On the one hand, the correspondence of landmarks is related the quality of SSMs. On the other hand, in many cases the location of key landmarks should be manipulated by physicians, since an unattended system is hard to use in most clinical applications. In this paper, we establish a dense correspondence across the whole training set automatically by surface features, which are registered using diffeomorphic demons approach. And the optimization is executed on spherical domain. We establish the SSM for lung regions, the deformation of where is greatly. Finally, we derive quantitative measures of model quality and comparison of segmentation results using the model with non-optimized correspondence.

DOI： 10.1007/978-3-319-69877-9_14

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85033680194&origin=inward

Underwater Light Field Depth Map Restoration Using Deep Convolutional Neural Fields 査読有り国際誌

Lu H., Li Y., Kim H., Serikawa S.

Studies in Computational Intelligence 752 305 - 312 2018年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© Springer International Publishing AG 2018. Underwater optical images are usually influenced by low lighting, high turbidity scattering and wavelength absorption. In order to solve these issues, a great deal of work has bee n used to improve the quality of underwater images. Most of them used the high-intensity LED for lighting to obtain the high contrast images. However, in high turbidity water, high-intensity LED causes strong scattering and absorption. In this paper, we firstly propose a light field imaging approach for solving underwater depth map estimation problems in low-intensity lighting environment. As a solution, we tackle the problem of de-scattering from light field images by using deep convolutional neural fields in depth estimation. Experimental results show the effectiveness of the proposed method through challenging real world underwater imaging.

DOI： 10.1007/978-3-319-69877-9_33

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85033692294&origin=inward

Automatic identification of bone erosions in rheumatoid arthritis from hand radiographs based on deep convolutional neural network 査読有り国際誌

Murakami S., Hatano K., Tan J., Kim H., Aoki T.

Multimedia Tools and Applications 77 ( 9 ) 10921 - 10937 2018年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© Springer Science+Business Media, LLC 2017. Although radiographic assessment of joint damage is essential in characterizing disease progression and prognosis in patients with rheumatoid arthritis (RA), it is often difficult even for trained radiologists to find radiographic changes on hand and foot radiographs because lesion changes are often subtle. This paper proposes a novel quantitative method for automatically detecting bone erosion on hand radiographs to assist radiologists. First, the proposed method performs with the crude segmentation of phalanges regions from hand radiograph and extracts the detailed phalanges regions by the multiscale gradient vector flow (MSGVF) Snakes method. Subsequently, the region of interest (ROI; 40 × 40 pixels) is automatically set on the contour line of the segmented phalanges by the MSVGF algorithm. Finally, these selected ROIs are identified by the presence or absence of bone erosion using a deep convolutional neural network classifier. This proposed method is applied to the hand radiographs of 30 cases with RA. The true-positive rate and the false-positive rate of the proposed method are 80.5% and 0.84%, respectively. The number of false-positive ROIs is 3.3 per case. We believe that the proposed method is useful for supporting radiologists in imaging diagnosis of RA.

DOI： 10.1007/s11042-017-5449-4

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85056120147&origin=inward

DCNNによる指骨CR画像からの骨粗しょう症の自動識別査読有り

畠野和裕, 村上誠一, 植村知規, 陸慧敏, タンジュークイ, 金亨燮, 青木隆敏

Ｍｅｄｉｃａｌ　Ｉｍａｇｉｎｇ　Ｔｅｃｈｎｏｌｏｇｙ ( 日本医用画像工学会 ) 36 ( 2 ) 90 - 95 2018年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

骨のおもな疾患として，骨粗しょう症が挙げられる．骨粗しょう症に対する画像診断は有効であるが，類似した低骨量を呈する画像も多く，画像診断における客観性や再現性の問題がある．そこで本稿では，指骨computed radiography（CR）画像から骨粗しょう症の自動識別手法を提案する．提案手法では，深層畳み込みニューラルネットワーク（DCNN）を用いた識別器を構築し，骨粗しょう症有無の識別を行う．DCNNの学習および識別には，CR画像から3種類の画像を作成し，各指骨領域内部からROIを抽出後，この3種類のROIをR，G，Bチャンネルに割り当て生成した疑似カラー画像を用いる．実験では，101症例に対し提案手法を適用し，真陽性率（TPR）：75.5［％］，偽陽性率（FPR）：13.9［％］という結果を得た．

DOI： 10.11409/mit.36.90

その他リンク： https://ci.nii.ac.jp/naid/130006588793

Classification of osteoporosis from phalanges CR images based on DCNN 査読有り国際誌

Hatano K., Murakami S., Lu H., Kooi Tan J., Kim H., Aoki T.

International Conference on Control, Automation and Systems 2017-October 1593 - 1596 2017年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Institute of Control, Robotics and Systems - ICROS. Osteoporosis is known as a disease of bone. Visual screening using Computed Radiography (CR) images is an effective method for osteoporosis, however, there are many similar diseases that exhibit state of low bone mass. In this paper, we propose an automatic identification method of osteoporosis from phalanges CR images. In the proposed method, we implement a classifier based on Deep Convolutional Neural Network (DCNN), and identify unknown CR images as normal or abnormal. For training and evaluating of CNN, we use pseudo color images. In the experiment, we apply our proposal method to 101 cases and TPR of 64.7 [%] and FPR of 6.51 [%] were obtained.

DOI： 10.23919/ICCAS.2017.8204241

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85044473373&origin=inward

Image registration of vertebral region from CT images based on salient region feature 査読有り国際誌

Sato S., Lu H., Tan J., Kim H., Murakami S., Ueno M., Terasawa T., Aoki T.

International Conference on Control, Automation and Systems 2017-October 1597 - 1600 2017年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Institute of Control, Robotics and Systems - ICROS. In recent years, the development of the computer-aided diagnosis (CAD) systems to support radiologist is attracting attention in medical research field. Temporal subtraction technique, which is one of CAD, is a technique to generate images emphasizing temporal changes in lesions by performing a differential operation between current and previous image of the same subject. In this paper, we propose an image registration method for image registration of current and previous image, to generate temporal subtraction images from CT images and enhanced bone metastasis region. The proposed registration method is composed into three main steps: i) segmentation of the region of interest (ROI) using graph cut, ii) use global image matching to select pairs from previous and current image, and iii) final image matching based on salient region feature. We perform our proposed method to synthesis and satisfactory registration experiments. The rotated synthesis image give TP 100.0[%] and FP 12.16[%] . The synthesis image obtained by applying a Gaussian filter give TP 70.40[%] and FP 0.00[%] . The synthesis image obtained by adding artificial pseudo lesion region give TP 99.45[%] and FP 17.89[%] . The synthesis image obtained by adding random noise of 5[%], which gave TP 83.05[%] and FP 16.95[%].

DOI： 10.23919/ICCAS.2017.8204242

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85044465807&origin=inward

Image analysis of cleft lip from moire image based on symmetry analysis 査読有り国際誌

Yamada S., Lu H., Tan J., Kim H., Kimura N., Okawachi T., Nozoe E., Nakamura N.

International Conference on Control, Automation and Systems 2017-October 1586 - 1589 2017年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Institute of Control, Robotics and Systems - ICROS. About 4% of babies in Japan suffer from congenital anomalies. Cleft lip is the most common disorder among external malformations occurring at a rate of 1 in 500 people. The goal of treatment is to recover symmetric and functional lips and nose forms. However, in the case of unilateral cleft lip where the lips and nose themselves are shifted from the midline of the face, it is difficult to set the midline which is the symmetry axis for evaluating the degree of symmetry of the face. In this paper, we propose an image processing method for extracting the midline using zebra image and shadow image which are two dimensional image created based on three dimensional analysis. Also, we evaluate the degree of asymmetry of the face by using the midline as the axis. As a result of applying the proposed method to 25 cases, which indicated that the proposed method is useful.

DOI： 10.23919/ICCAS.2017.8204239

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85044463365&origin=inward

Extraction of spinal candidate region from a dental MR imaging 査読有り国際誌

Suetani K., Lu H., Tan J., Kim H., Tanaka T., Kitou S., Morimoto Y.

International Conference on Control, Automation and Systems 2017-October 1601 - 1604 2017年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Institute of Control, Robotics and Systems - ICROS. In recent years, dysphagia is problem among elderly people. Therefore, it is necessary to accurately evaluate swallowing function in order to prevent swallowing disorder beforehand or to detect it early, and it is considered that evaluation of swallowing function using MRI is useful, and the demand for development of CAD system using MRI is increasing. In order to accurately analyze the swallowing motion, automatic extraction of the region of interest by the image analysis method is necessary. In this paper, as a pretreatment of a method for automatic extraction of esophageal region on dental MR image, we develop automatic extraction method of spinal candidate region. We first extract the posterior region of the spine and then extract the intervertebral disc region. Then we perform the detection of the anterior region of the spine and finally we extract the region between the anterior region of the spine and the posterior region of the spine as a spinal region. The proposed method was applied to 5 cases of dental MR images, and the results of TP 91.2 [%] and FP 8.8 [%] are obtained.

DOI： 10.23919/ICCAS.2017.8204243

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85044477204&origin=inward

Detection of lung nodules on temporal subtraction images Using 3D sparse coding 査読有り国際誌

Tanaka T., Miyake N., Lu H., Tan J., Kim H., Murakami S., Aoki T., Hirano Y., Kido S.

International Conference on Control, Automation and Systems 2017-October 1455 - 1457 2017年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Institute of Control, Robotics and Systems - ICROS. In recent years, the death rate caused by lung cancer is increasing. To detect the lung cancer, multi detector-row computed tomography (MDCT) images are used in visual screening. Lung cancer can be easily detected by using the chest MDCT images, however, it has enormous images and burden to radiologists. Research and development of the computer aided diagnosis (CAD) system have been assisted the diagnosis. As one of the CAD technologies, temporal subtraction technique is possible to emphasize the changing interval on the CT images. It uses subtraction operation between previous and current CT images of the same patient. On the other hand, pattern recognition using image reconstruction by sparse coding method has attracted attention. This technique is mathematically modeling the information processing by the primary visual cortex of human. It is the technique for representing images by the linear combination of a small number of basis. In this paper, candidate nodules under 20[mm] were segmented from temporal subtraction images based on the 3D sparse coding technique.3D sparse coding is three dimensional expansion of the sparse coding. Also, we classified the final candidate nodules using support vector machine (SVM) method based on coefficient matrix which are obtained by the 3D sparse coding. We applied proposed method to 31 cases of chest MDCT images in which the number of nodules was more than one. We achieved experimental result with true positive rates (TPR) of 70.2[[%] , and false positive rates (FP) of 34.7[/scan], respectively.

DOI： 10.23919/ICCAS.2017.8204220

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85044482344&origin=inward

Detection of abnormal candidate regions on temporal subtraction images based on DCNN 査読有り国際誌

Nagao M., Miyake N., Yoshino Y., Lu H., Kooi Tan J., Kim H., Murakami S., Aoki T., Hirano Y., Kido S.

International Conference on Control, Automation and Systems 2017-October 1444 - 1448 2017年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Institute of Control, Robotics and Systems - ICROS. Cancer is a leading cause of death both in Japan and worldwide. Detection of cancer region in CT images is the most important task to early detection. Recently, visual screening based on CT images become useful tools for cancer detection. However, due to the large number of images and the complexity of the image processing algorithms, image processing technique is still required a high screening quality. To overcome this problem, some computer aided diagnosis (CAD) algorithms are proposed. In this paper, we have designed and developed a framework combining machine learning based on deep convolutional neural networks (DCNN) and temporal subtraction techniques based on non-rigid image registration algorithm. Our main classification method can be built into three main steps; i) pre-processing for image segmentation, ii) image matching for registration, and iii) classification of abnormal regions based on machine learning algorithms. We performed our proposed technique to 25 thoracic MDCT sets and obtained true positive rates of 92.31 [%], false positive rates of 6.32 [/case] were obtained.

DOI： 10.23919/ICCAS.2017.8204218

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85044482182&origin=inward

Extraction of GGO regions from chest CT images using deep learning 査読有り国際誌

Hirayama K., Miyake N., Lu H., Tan J., Kim H., Tachibana R., Hirano Y., Kido S.

International Conference on Control, Automation and Systems 2017-October 351 - 355 2017年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Institute of Control, Robotics and Systems - ICROS. Lung cancer is the leading cause of death which accounts for the number of deaths in cancer in the world. Early detection and early treatment are regarded as an important. Especially, the ground glass opacity (GGO) is a shadow called pre-cancerous lesion, but it is a shadow which is difficult to detect by a radiologist because of haze and complicated shape. Therefore, in recent years, a computer aided diagnosis (CAD) system has been developed for the purpose of improving the detection accuracy for early detection and reducing the burden to radiologists. In this paper, we extract the GGO using Deep Convolutional Neural Network (DCNN) based on emphasized images. Before detect a GGO region, we apply preprocessing such as isotropic voxel to the original images, and extraction of the lung area. Next, we remove the vessel and bronchial region by 3D line filter based on Hessian matrix, and extract the initial candidate regions using density gradient, volume and sphericity. Subsequently, we segment the candidate regions, extraction of features, and reducing false positive shadows. Finally we create emphasize images and identify with DCNN using those images. As a result of applying the proposed method to 31 cases on Lung Image Database Consortium (LIDC), we obtained a true positive rate (TP) of 86.05 [%] and false positive number (FP) of 4.81 [/case].

DOI： 10.23919/ICCAS.2017.8204464

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85044442253&origin=inward

Environment recognition for navigation of autonomous wheelchair from a video image 査読有り国際誌

Nakayama Y., Lu H., Tan J., Kim H.

International Conference on Control, Automation and Systems 2017-October 1439 - 1443 2017年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Institute of Control, Robotics and Systems - ICROS. The Japanese population has rapidly aged and the number of aged persons who have lower physical ability has increased recently. Thus the development of medical and healthcare devices is expected. Wheelchair requires care support in most cases. Therefore the development of autonomous wheelchair is meaningful since we can expect to improve convenience and to reduce burden of caregivers. The autonomous wheelchair requires several techniques. Our research is to develop a navigation system based on image processing techniques. However, we assume that the system instructs an appropriate direction to head towards the destination when a wheelchair user comes to a crossing. Incidentally, deep learning, a kind of artificial neural network, has attracted attention in the field of machine learning in recent years. This paper proposes methodology for supporting autonomous driving by use of a classifier trained on a video images with deep learning. Also, we apply visual odometry to generate training data.

DOI： 10.23919/ICCAS.2017.8204217

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85044473264&origin=inward

Automatic identification of circulating tumor cells in fluorescence microscopy images based on AdaBoost 査読有り国際誌

Tsuji K., Lu H., Tan J.K., Kim H., Yoneda K., Tanaka F.

International Conference on Control, Automation and Systems 2017-October 1449 - 1454 2017年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Circulating tumor cells (CTCs) is a useful biomarker for cancer metastasis. The blood from a cancer patient is analyzed by a fluorescence microscope. Each case takes a large number of images, which usually have a lot of cell regions. Thus, analyzing the images is hard work for pathologists, and misdiagnosis may happen. In this paper, we develop an automatic CTCs identification method for fluorescence microscopy images. The proposed method consists of three steps. First, we extract cell regions in images using filtering methods. Second, we compute features of each CTC candidate regions. Finally, we identify the CTCs using AdaBoost algorithm. And we analyze the features to know which ones are effective for characterizing CTCs and normal cells. We apply the proposed method to 5040 microscopy images, and evaluate the effectiveness of our method by using leave-one-out cross validation. We achieve a true positive rate of 97.30 [%] and a false positive rate of 12.82 [%].

DOI： 10.23919/ICCAS.2017.8204219

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85044436784&origin=inward

Automatic liver segmentation from multiphase CT images by using level set method 査読有り国際誌

Saito K., Lu H., Tan J.K., Kim H., Yamamoto A., Kido S., Tanabe M.

International Conference on Control, Automation and Systems 2017-October 1590 - 1592 2017年12月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Segmentation of liver from Multi-phase CT images is one of the essential technology for computer aided diagnosis. Contrast medium gives multi-phase CT images different intensity feature which enables to detect tumor. It is a challenging problem to segment liver region from multi-phase CT images. There are many approaches for solving this problem, however, these methods depend on other phases or registration. In order to solve this problem, we propose anatomy feature-based method which is mostly independent for each phase in this paper. This method uses level set method for final segmentation. The accuracy of segmentation result by level set methods relay on initial contour, so we preprocess initial region of liver by anatomical feature. Then we introduced contour constrain by using ribs information to improve segmentaion accuracy. Our segmentation was evaluated on 5 multi-phase CT images which have 4 phases. Experimental results show that the proposed method is good accuracy for each phase.

DOI： 10.23919/ICCAS.2017.8204240

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85044443256&origin=inward

Underwater Optical Image Processing: a Comprehensive Review 査読有り国際誌

Lu H., Li Y., Zhang Y., Chen M., Serikawa S., Kim H.

Mobile Networks and Applications 22 ( 6 ) 1204 - 1211 2017年12月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017, Springer Science+Business Media New York. Underwater cameras are widely used to observe the sea floor. They are usually included in autonomous underwater vehicles (AUVs), unmanned underwater vehicles (UUVs), and in situ ocean sensor networks. Despite being an important sensor for monitoring underwater scenes, there exist many issues with recent underwater camera sensors. Because of light’s transportation characteristics in water and the biological activity at the sea floor, the acquired underwater images often suffer from scatters and large amounts of noise. Over the last five years, many methods have been proposed to overcome traditional underwater imaging problems. This paper aims to review the state-of-the-art techniques in underwater image processing by highlighting the contributions and challenges presented in over 40 papers. We present an overview of various underwater image-processing approaches, such as underwater image de-scattering, underwater image color restoration, and underwater image quality assessments. Finally, we summarize the future trends and challenges in designing and processing underwater imaging sensors.

DOI： 10.1007/s11036-017-0863-4

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85018366927&origin=inward

Weighted Local Laplacian Filtering for Underwater Image Enhancement 査読有り

Lu, Li, Serikawa, Kim

Proceeding of The 2nd International Symposium on Artificial Intelligence and Robotics 36 - 47 2017年11月

DeepEye: A dedicated Camera for Deep-sea Tripod Observation Systems in the South China Sea 査読有り

Lu, Li, Uemura, Tadoh, Kihara, Serikawa, Kim

Proceeding of The 2nd International Symposium on Artificial Intelligence and Robotics 198 - 201 2017年11月

A Diffeomorphic Demons Approch to Statistical Shape Modeling 査読有り

Li, Wu, Xiao, Lu, Kim, Ogubona

Proceeding of The 2nd International Symposium on Artificial Intelligence and Robotics 175 - 182 2017年11月

AKEZE局所特徴量に基づく歯科MR画像からの嚥下動作の解析法

末谷、陸、タン、金、田中、鬼頭、森本

第35回計測自動制御学会九州支部学術講演会論文集 25 - 26 2017年11月

Non-Rigid Image Registration Techniques and Its Application to Medical Imaging 招待有り査読有り

Kim

International Symposium on Artificial Intelligence and Robotics 2017年11月

経時的差分像技術を用いた胸部CT画像上のGGO候補領域の検出

近藤、三宅、陸、タン、金、村上、寺澤、青木、平野、木戸

日本医療情報学会講演会 884 - 887 2017年11月

Automatic segmentation of cell candidate regions in microscopy images based on an optimization algorithm 査読有り

Tsuji K., Tan J., Kim H., Yoneda K., Tanaka F.

International Conference on Control, Automation and Systems 720 - 723 2017年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2016 Institute of Control, Robotics and Systems - ICROS.Circulating tumor cells (CTCs) is an informative biomarker which assists pathologists in early diagnosis and evaluating therapeutic effects of patients with malignant tumors. The blood from a cancer patient is analyzed by a microscope and a large number of pictures including many cells are generated for each case. Thus, analyzing them is time-consuming work for pathologists, and misdiagnosis may happen since the diagnosis of CTCs tends to depend on the individual skill of pathologist. In this paper, we propose a method which detects cell candidate regions in microscopy images automatically to make quantitative analysis possible by computer. Our proposed method consists of three steps. In the first step, we extract initial cell candidate regions in microscopy images based on the saliency map. In the second step, we choose non-single cell regions from the initial candidates based on the SVM algorithm. In the third step, we separate connected regions into single cell regions based on the branch and bound algorithm. We demonstrated the effectiveness of our proposed method using 540 microscopy images and we achieved a true positive rate of 99.04[%] and a false positive rate of 3.95[%].

DOI： 10.1109/ICCAS.2016.7832397

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85014007441&origin=inward

Extraction of GGO candidate regions from the LIDC database using deep learning 査読有り

Hirayama K., Tan J., Kim H.

International Conference on Control, Automation and Systems 724 - 727 2017年10月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2016 Institute of Control, Robotics and Systems - ICROS.In recent years, development of the computer-aided diagnosis (CAD) systems for the purpose of reducing the false positive on visual screening and improving accuracy of lesion detection has been advanced. Lung cancer is the leading cause of cancer death in the world. Among them, GGO (Ground Glass Opacity) that exhibited early in the before cancer lesion and carcinoma in situ shows a pale concentration, have been concerned about the possibility of undetected on the screening. In this paper, we propose an automatic extraction method of GGO candidate regions from the chest CT image. Our proposed image processing algorithms is consist of four main steps; 1) segmentation of volume of interest from the chest CT image and removing the blood vessel regions, bronchus regions based on 3D line filter, 2) first detection of GGO regions based on density and gradient which is selected the initial GGO candidate regions, 3) identification of the final GGO candidate regions based on DCNN (Deep Convolutional Neural Network) algorithms. Finally, we calculates the statistical features for reducing the false-positive (FP) shadow by the rule-based method, performs identification of the final GGO candidate regions by SVM (Support Vector Machine). Our proposed method performed on to the 31 cases of the LIDC (Lung Image Database Consortium) database, and final identification performance of TP: 93.02[%], FP: 128.52[/case] are obtained respectively.

DOI： 10.1109/ICCAS.2016.7832398

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85014043688&origin=inward

Environment Recognition for Navigation of Autonomous Wheelchair from a Video Image 査読有り

Nakayama, Lu, Tan, Kim

Proceedings of the 17th International Conference on Control, Automation and Systems 1439 - 1443 2017年10月

Extraction of Spinal Candidate Region from A Dental MR Imaging 査読有り

Suetani, Lu, Tan, Kim, Tanaka, Kitou, Morimoto

Proceedings of the 17th International Conference on Control, Automation and Systems 1601 - 1604 2017年10月

Image Registration of Vertebral Region from CT Images Based on Salient Region Feature 査読有り

Sato, Lu, Tan, Kim, Murakami, Ueno, Terasawa, Aoki

Proceedings of the 17th International Conference on Control, Automation and Systems 1597 - 1600 2017年10月

Classification of Osteoporosis from Phalanges CR Images Based on DCNN 査読有り

Hatano, Murakami, Lu, Tan, Kim, Aoki

Proceedings of the 17th International Conference on Control, Automation and Systems 1593 - 1596 2017年10月

Automatic Liver Segmentation from Multiphase CT images by Using Level Set Method 査読有り

Saito, Lu, Tan, Kim, Yamamoto, Kido, Tanabe

Proceedings of the 17th International Conference on Control, Automation and Systems 1590 - 1592 2017年10月

Image Analysis of Cleft Lip from Moire Image Based on Symmetry Analysis 査読有り

Yamada, Lu, Tan, Kim, Kimura, Okawachi, Nozoe, Nakamura

Proceedings of the 17th International Conference on Control, Automation and Systems 1586 - 1589 2017年10月

Detection of Lung Nodules on Temporal Subtraction Images Using 3D Sparse Coding 査読有り

Tanaka, Miyake, Lu, Tan, Kim, Murakami, Aoki, Hirano, Kido

Proceedings of the 17th International Conference on Control, Automation and Systems 1455 - 1457 2017年10月

Automatic Identification of Circulating Tumor Cells in Fluorescence Microscopy Images Based on AdaBoost 査読有り

Tsuji, Lu, Tan, Kim, Yoneda, Tanaka

Proceedings of the 17th International Conference on Control, Automation and Systems 1449 - 1454 2017年10月

Detection of Abnormal Candidate Regions on Temporal Subtraction Images Based on DCNN 査読有り

Nagao, Miyake, Yoshino, Lu, Tan, Kim, Murakami, Aoki, Hirano, Kido

Proceedings of the 17th International Conference on Control, Automation and Systems 1444 - 1448 2017年10月

Automatic classification of lung nodules on MDCT images with the temporal subtraction technique 査読有り

Yoshino Y., Miyajima T., Lu H., Tan J., Kim H., Murakami S., Aoki T., Tachibana R., Hirano Y., Kido S.

International Journal of Computer Assisted Radiology and Surgery 12 ( 10 ) 1789 - 1798 2017年10月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017, CARS. Purpose: A temporal subtraction (TS) image is obtained by subtracting a previous image, which is warped to match the structures of the previous image and the related current image. The TS technique removes normal structures and enhances interval changes such as new lesions and substitutes in existing abnormalities from a medical image. However, many artifacts remaining on the TS image can be detected as false positives. Method: This paper presents a novel automatic segmentation of lung nodules using the Watershed method, multiscale gradient vector flow snakes and a detection method using the extracted features and classifiers for small lung nodules (20 mm or less). Result: Using the proposed method, we conduct an experiment on 30 thoracic multiple-detector computed tomography cases including 31 small lung nodules. Conclusion: The experimental results indicate the efficiency of our segmentation method.

DOI： 10.1007/s11548-017-1598-1

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85019139416&origin=inward

An Interactive Technique of Fast Vertebral Segmentation for Computed Tomography Images with Bone Metastasis 査読有り

Dong, Lu, Kim, Aoki, Yihong Zhao, Zhao

Proceedings of the 2nd International Conference on Biomedical Signal and Image Processing 29 - 36 2017年08月

Statistical Shape Model Generation Using Diffeomorphic Surface Registration 査読有り

J. Wu, G. Li, H. Lu, Kim, P.O. Ogubona

Proceedings of the 2nd International Conference on Biomedical Signal and Image Processing 37 - 41 2017年08月

Automatic Identification of Circulating Tumor Cells in Fluorescence Microscopy Images Based on ANN 査読有り

Tsuji, Lu, Tan, Kim, Yoneda, Tanaka

Proceedings of the 2nd International Conference on Biomedical Signal and Image Processing 1 - 6 2017年08月

Image Registration Techniques and Its Application for Computer Aided Dianosis in Medical Field 招待有り査読有り

Kim

2nd Ingernational Conference on Biomedical Signal and Image Processing 2017年08月

An interactive technique of fast vertebral segmentation for computed tomography images with bone metastasis 査読有り国際誌

Dong R., Lu H., Kim H., Aoki T., Zhao Y., Zhao Y.

ACM International Conference Proceeding Series 29 - 36 2017年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Association for Computing Machinery. Computer-aided diagnosis (CAD) system can assistant radiologists to diagnose bone metastasis, which not only reduces burden on workload but also improves diagnostic accuracy. As key step in CAD system, vertebral segmentation can directly affect diagnostic results. In order to obtain high accurate segmentation results, we propose a connected component Labeled Graph Cuts (LGC) algorithm. The proposed method is tested on 100 computed tomography (CT) slices. The assessed quantitatively of experimental results is compared with those by radiologist. The proposed method has a 96.72[%] of True Positive Rate (TPR), and 1.84[%] of False Positive Rate (FPR), which have better performance than conventional Graph Cuts algorithm, 90.07[%] of TPR and 2.32[%] of FPR.

DOI： 10.1145/3133793.3133795

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85052716817&origin=inward

Statistical shape model generation using diffeomorphic surface registration 査読有り国際誌

Wu J., Li G., Lu H., Kim H., Ogunbona P.

ACM International Conference Proceeding Series 37 - 41 2017年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Association for Computing Machinery. Statistical shape modelling is an efficient and robust method for medical image segmentation in computer-aided diagnosis. The key step in building a statistical shape model is to find corresponding landmarks in each instance of a training set. In this paper, a novel landmark correspondence estimation method that uses edge collapse surface simplification and the sphere registration is proposed. All the landmarks are selected and transformed by spherical conformal mapping from the instances of the training set and the associated correspondence are automatically found on the spheres. We applied our method on 21 cases of 3-D right lung shapes. The results of image segmentation experiment indicate that our method has a positive influence on the accuracy of segmentation result.

DOI： 10.1145/3133793.3133796

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85052679703&origin=inward

Automatic identification of circulating tumor cells in fluorescence microscopy images based on ANN 査読有り国際誌

Tsuji K., Lu H., Tan J., Kim H., Yoneda K., Tanaka F.

ACM International Conference Proceeding Series 1 - 6 2017年08月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2017 Association for Computing Machinery. Circulating tumor cells (CTCs) are a useful biomarker since they may have some information about cancer metastasis. The blood from cancer patient is analyzed by a fluorescence microscope. It takes a large number of photos for each case, and many cells are contained in the microscopy images. Thus, analyzing them is hard work for pathologists. This work tends to depend on the individual skill of pathologist so misdiagnosis may be happen. In this paper, we develop an automatic CTCs identification method in fluorescence microscopy images based on artificial neural network. We applied our proposed method to 5040 microscopy images (6 cases), and evaluated the effectiveness of our method by using leave-one-out cross validation. We achieve a true positive rate of 98.65 [%] and a false positive rate of 18.24 [%].

DOI： 10.1145/3133793.3133798

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85052677091&origin=inward

画像処理技術による診断支援への応用

金

全国歯科大学・歯学部付属病院診療放射線技師連絡協議会会誌 2017年07月

Moving objects detection employing iterative update of the background 査読有り国際誌

Setyawan F., Tan J., Kim H., Ishikawa S.

Artificial Life and Robotics 22 ( 2 ) 168 - 174 2017年06月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017, ISAROB.Detection of objects from a video is one of the basic issues in computer vision study. It is obvious that moving objects detection is particularly important, since they are those to which one should pay attention in walking, running, or driving a car. This paper proposes a method of detecting moving objects from a video as foreground objects by inferring backgrounds frame by frame. The proposed method can cope with various changes of a scene including large dynamical change of a scene in a video taken by a stationary/moving camera. Experimental results show satisfactory performance of the proposed method.

DOI： 10.1007/s10015-016-0347-9

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85009833761&origin=inward

Activity representation by SURF-based templates 査読有り

Md. Atiqur Rahman Ahad, JK Tan, H Kim, S Ishikawa

Computer Methods in Biamechanics and Biomedical Engineering: Image & Visualization 1 - 11 2017年04月

A supervised correspondence method for statistical shape model building 査読有り

Li G., Honda H., Yoshino Y., Kim H., Xiao Z.

2016 IEEE International Conference on Signal and Image Processing, ICSIP 2016 37 - 40 2017年03月

担当区分：責任著者記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2016 IEEE.The construction of statistical shape model (SSM) is an important research topic in medical imaging benefited from its robust and nature represent of anatomical structures. Place-march of corresponding landmarks is one of the major factors influencing 3D SSM quality. In this paper, we present a supervised correspondence method for fast building SSM, which includes two main steps, i.e., surface data alignment and landmarks specified based on surface parameterization. The framework is validated with statistical models of the liver constructed from contrast CT images. The experiment results demonstrate that the generated model is statistical and anatomically meaningful.

DOI： 10.1109/SIPROCESS.2016.7888219

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85018716962&origin=inward

An ego-camera based finger-spelling recognition system 査読有り国際誌

Tan J., Hamada S., Hirakawa M., Kim H., Ishikawa S.

IEEE Region 10 Annual International Conference, Proceedings/TENCON 358 - 363 2017年02月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2016 IEEE.This paper describes a portable system for finger-spelling recognition, employing a PC and a video taken by an ego-camera mounted on the body of a person who does finger-spelling. The system is intended to be a useful tool for an orally impaired person to communicate to anyone at any place by carrying it with him/her. The images of the finger-spelling hand of a user who is carrying the system is captured by an ego-camera. The hand is extracted from arbitrary backgrounds by use of a Gaussian mixture model and skin color evaluation: The trimmed and normalized image of the extracted hand is recognized employing the feature space defined by applying the principal component analysis to the learning data containing 45 finger-spelled Japanese Hiragana letters each with 50 samples and the nearest neighbor method. The on-line performance of the proposed system is experimentally shown.

DOI： 10.1109/TENCON.2016.7848021

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85015432805&origin=inward

経時的差分像技術を用いた結節状候補陰影の検出

金，村上，寺澤，青木，木戸

呼吸機能イメージング研究会学術集会講演論文集 58 - 58 2017年01月

Underwater image super-resolution by descattering and fusion 査読有り国際誌

Lu H., Li Y., Nakashima S., Kim H., Serikawa S.

IEEE Access 5 670 - 679 2017年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2013 IEEE. Underwater images are degraded due to scatters and absorption, resulting in low contrast and color distortion. In this paper, a novel self-similarity-based method for descattering and super resolution (SR) of underwater images is proposed. The traditional approach of preprocessing the image using a descattering algorithm, followed by application of an SR method, has the limitation that most of the high-frequency information is lost during descattering. Consequently, we propose a novel high turbidity underwater image SR algorithm. We first obtain a high resolution (HR) image of scattered and descattered images by using a self-similarity-based SR algorithm. Next, we apply a convex fusion rule for recovering the final HR image. The super-resolved images have a reasonable noise level after descattering and demonstrate visually more pleasing results than conventional approaches. Furthermore, numerical metrics demonstrate that the proposed algorithm shows a consistent improvement and that edges are significantly enhanced.

DOI： 10.1109/ACCESS.2017.2648845

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85018527963&origin=inward

Highly accurate energy-conserving flexible touch sensors 査読有り国際誌

Lu H., Li Y., Li Y., Serikawa S., Kim H.

Sensors and Materials 29 ( 6 ) 611 - 617 2017年01月

担当区分：責任著者記述言語：英語掲載種別：研究論文（学術雑誌）

© MYU K.K. Smart touch sensors are used in many applications, such as the iPhone and some smart home systems. Recent touch sensors perform well in consumer electrical devices; however, there are some drawbacks. For example, most touch sensors have low accuracy for detecting human movements. Many require significant power and have a fixed shape. We propose a flexible, highly accurate, and energy-conserving touch sensor. Our primary contributions are as follows: (1) an energy-conserving touch sensor is developed and tested experimentally; (2) a flexible and arbitrarily shaped touch sensor is designed; and (3) the manufacturing cost is very low and up to 200 touch sensors can be connected to the system. As a result, this energy-conserving touch sensor can be fabricated using common manufacturing processes for consumer electronic devices.

DOI： 10.18494/SAM.2017.1458

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85020809645&origin=inward

Depth Map Reconstruction for Underwater Kinect Camera Using Inpainting and Local Image Mode Filtering 査読有り国際誌

Lu H., Zhang Y., Li Y., Zhou Q., Tadoh R., Uemura T., Kim H., Serikawa S.

IEEE Access 5 7115 - 7122 2017年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2017 IEEE. Underwater optical cameras are widely used for security monitoring in ocean, such as earthquake prediction and tsunami alarming. Optical cameras recognize objects for autonomous underwater vehicles and provide security protection for sea-floor networks. However, there are many issues for underwater optical imaging, such as forward and backward scattering, light absorption, and sea snow. Many underwater image processing techniques have been proposed to overcome these issues. Among these techniques, the depth map gives important information for many applications of the post-processing. In this paper, we propose a Kinect-based underwater depth map estimation method that uses a captured coarse depth map by Kinect with the loss of depth information. To overcome the drawbacks of low accuracy of coarse depth maps, we propose a corresponding reconstruction architecture that uses the underwater dual channels prior dehazing model, weighted enhanced image mode filtering, and inpainting. Our proposed method considers the influence of mud sediments in water and performs better than the traditional methods. The experimental results demonstrated that, after inpainting, dehazing, and interpolation, our proposed method can create high-accuracy depth maps.

DOI： 10.1109/ACCESS.2017.2690455

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85028360504&origin=inward

細胞領域の論理積を用いた蛍光顕微鏡画像からの血中循環がん細胞の自動検出査読有り

辻幸喜, 陸慧敏, タンジュークイ, 金亨燮, 米田和恵, 田中文啓

医用画像情報学会雑誌 ( 医用画像情報学会 ) 34 ( 4 ) 151 - 155 2017年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

<p>Circulating tumor cells（CTCs）can be a useful biomarker. They may have some information about the malignant disease, since they are one of causes of the cancer metastasis. The blood sample from cancer patient is analyzed by fluorescence microscope. This microscope takes enlarged images with three types of lights（red, green and blue）,and specific materials are reacted respectively. The blood contains a lot of cells, but there are few CTCs. Therefore analyzing them is not easy work for pathologists. In this study, we develop a method which detects circulating tumor cells in fluorescence microscopy images automatically. Our proposed method has three steps. First, we extract cell regions in microscopy images by using filtering processing. Second, we separate the connecting cell regions into single cell regions,based on the branch and bound algorithm. Finally, we identify CTCs by using logical conjunction method. We demonstrated the effectiveness of our proposed method using 6 cases（5040 microscopy images）, and we evaluated the performance of CTCs identification. Our proposed method achieved, a true positive rate of 95.27 [%] and a false positive rate of 6.172 [%] respectively. And we confirmed the effectiveness of the logical conjinction for CTCs identicication.</p>

DOI： 10.11318/mii.34.151

その他リンク： https://ci.nii.ac.jp/naid/130006267703

転移深層学習畳み込みニューラルネットワークを用いたCTコロノグラフィ候補陰影からのポリープ分類法査読有り

植村知規, 陸慧敏, 金亨燮, 橘理恵, 弘中亨, Janne J. Näppi, 吉田広行

医用画像情報学会雑誌 ( 医用画像情報学会 ) 34 ( 2 ) 80 - 86 2017年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

<p>Computed tomographic colonography（CTC）, also known as virtual colonoscopy, provides a minimally invasive screening method for early detection of colorectal lesions. It can be used to solve the problems of accuracy, capacity, cost,and safety that have been associated with conventional colorectal screening methods. Computer-aided detection（CADe）has been shown to increase radiologists' sensitivity and to reduce inter-observer variance in detecting colonic polyps in CTC. However, although CADe systems can prompt locations of abnormalities at a higher sensitivity than that of radiologists,they also prompt relatively large numbers of false positives（FPs）. In this study, we developed and evaluated the effect of a transfer-learning deep convolutional neural network（TL-DCNN）on the classification of polyp candidates detected by a CADe system from dual-energy CTC images. A deep convolution neural network（DCNN）that had been pre-trained with millions of natural non-medical images was fine-tuned to identify polyps by use of pseudo-colored images that were generated by assigning axial, coronal, and sagittal images of the polyp candidates to the red, green, and blue channels of the images, respectively. The classification performances of the TL-DCNN and the corresponding non-transfer-learning DCNN were evaluated by use of 5-fold cross validation on 20 clinical CTC cases. The TL-DCNN yielded true- and falsepositive rates of 73.6［％］and 1.79［％］, respectively, which were significantly higher than those of the non-transferlearning DCNN. This preliminary result demonstrates the effectiveness of the TL-DCNN in the classification of polyp candidates from CTC images.</p>

DOI： 10.11318/mii.34.80

その他リンク： https://ci.nii.ac.jp/naid/130006846731

DCNNによるLIDCデータからのすりガラス状陰影の検出査読有り

平山一希, 陸慧敏, タンジュークイ, 金亨燮, 橘理恵, 平野靖, 木戸尚治

医用画像情報学会雑誌 ( 医用画像情報学会 ) 34 ( 2 ) 70 - 74 2017年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

<p>Lung cancer is one of the most important cancer in the world. Among them, Ground Glass Opacity（GGO）has a hazy area of increased attenuation in the lung image. In recent years, development of a Computer Aided Diagnosis （CAD）system for reducing the burden on work load and improving the detection rate of lesions has been advanced. In this paper, we propose a CAD system to extract GGO from CT images. Firstly, we extract the lung region from the input CT images and remove the vessel, and bronchial region based on 3 D line filter algorithm. After that, we extract initial GGO regions using concentration and gradient information. Next, we calculate the statistical features on the segmented regions. After that, we classify GGO regions using support vector machine（SVM）. Finally, we detect the final GGO regions using deep convolutional neural network（DCNN）. The proposed method is tested on 31 cases of CT images from the Lung Image Database Consortium（LIDC）. The results demonstrate that the proposed method has 86.05[%] of true positive rate and 39.03[/case] of false positive number.</p>

DOI： 10.11318/mii.34.70

その他リンク： https://ci.nii.ac.jp/naid/130006846732

Current distribution based power module screening by new normal/abnormal classification method with image processing 査読有り国際誌

Tsukuda M., Yuki D., Tomonaga H., Kim H., Omura I.

Proceedings of the International Symposium on Power Semiconductor Devices and ICs 407 - 410 2017年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

We developed a screening equipment for ceramic substrate level power module of IGBT. The equipment realizes a new screening test with current distribution. The equipment acquires magnetic field signals over bonding wires and finally classifies to normal/abnormal module automatically. We established statistics based normal/abnormal classification with image processing. It is expected to be applied for screening in a production line and analysis to prevent the failure of power modules.

DOI： 10.23919/ISPSD.2017.7988970

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85028510316&origin=inward

Automatic segmentation of cell candidate region in Microscopy images based on an optimization algorithm 査読有り

Tsuji, Tan, Kim, Yoneda, Tanaka

International Conference on Control, Automation and Systems 720 - 723 2016年10月

Extraction of GGO candidat regions from the LIDC database using deep learning 査読有り

Hirayama, Tan, Kim

International Conference on Control, Automation and Systems 724 - 727 2016年10月

Identify rheumatoid arthritis and osteoporosis from phalange CR images based on image registration and ANN 査読有り

Kajihara S., Murakami S., Murakami S., Tan J., Kim H., Aoki T.

ICIC Express Letters 10 ( 10 ) 2435 - 2440 2016年10月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2016 ICIC International.Rheumatoid arthritis and osteoporosis are two of the major diseases related to the phalanges. Diagnostic imaging is often used to diagnosis them. Especially, observing the temporal changes of the shape or internal structure of the phalanges in phalange CR images is an effective way to detect these diseases. However, there are various problems in image diagnosis, such as the evaluation of diagnosis is generally empiric, and the burden of reading images is heavy. In order to solve these problems, in this paper, we develop a computer aided diagnosis (CAD) system for automatically diagnosing, which includes segmentation of knuckles, registration of temporal images and features analysis of phalange regions. In the segmentation part, we perform the multi scale gradient vector flow (MSGVF) snakes to improve the precision of extracted results. Also we have developed an image registration technique based on salient region features (SRF) method. Two image features are used to train the artificial neural network (ANN) to identify the abnormal knuckle regions.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84991355197&origin=inward

Identify Rheumatoid Arthritis and Osteoporosis from Phalange CR Images Image　Based on Image Registration and ANN 査読有り国際誌

Kajihara, Murakami, Tan, Kim, Aoki

ICIC Express Letters 10 ( 10 ) 2435 - 2440 2016年10月

A Computer Aided Diagnosis System to Identify Rheumatoid Arthritis and Osteoporosis from Phalange CR Image 査読有り国際誌

Kajihara, Murakami, Tan, Kim, Aoki

11th International Conference on Innovative Computing, Information and Control 2016年08月

A method of detecting salient regions employing global and local saliency 査読有り

Kuwata I., Kooi Tan J., Kim H., Ishikawa S.

ICIC Express Letters, Part B: Applications 7 ( 3 ) 555 - 561 2016年03月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2016.This paper proposes a method of detecting prominent regions from an image using a global and local saliency measure. Detection of prominent regions from an image is important, because it is expected to improve the precision and the processing speed of object recognition through image processing. The proposed method employs two spatial redundancies, global saliency and local saliency. Global saliency is calculated by comparing the value of an interested pixel with the mode of the values of all pixels. On the other hand, local saliency is calculated using the relation among local pixels. In calculating the local saliency, the proposed method employs human visual characteristics, i.e., complementary color harmony to detect salient regions. Saliency map is made by integrating the global saliency and the local saliency considering some weights. Experimental results show the effectiveness of the proposed method.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84975702036&origin=inward

Detecting pedestrian and extracting their attributes from self-mounted camera views 査読有り

Sakai R., Kooi Tan J., Kim H., Ishikawa S.

ICIC Express Letters, Part B: Applications 7 ( 2 ) 279 - 286 2016年02月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2016 ICIC International.There are not a few visually impaired people in our society. When they want to go for a walk, it is difficult for them to perceive moving objects around them. When they walk on the sidewalk, they must pay strong attention to those coming closer to them or to moving objects. Therefore, it is indispensable for them to obtain the information on moving objects around them. This paper proposes a method of extracting moving objects, in particular, pedestrians, from self-wearable camera images. In the proposed system, a user is supposed to wear a camera and a PC and, by performing image analysis of the video taken from the camera, it acquires surrounding pedestrians’ information including their peculiar characteristics. The information is fed back to the user to realize his/her safe walk. Experimental results are shown and the performance of the proposed method is evaluated. The proposed system is primarily intended to be used by visually impaired people, but it may also be used by any pedestrian who is not paying much attention to his/her surroundings.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84959361584&origin=inward

Traffic signs and signals detection employing the my vision system for a visually impaired person 査読有り

Kumano T., Kooi Tan J., Kim H., Ishikawa S.

ICIC Express Letters, Part B: Applications 7 ( 2 ) 385 - 391 2016年02月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2016 ICIC International. In recent years, the equipment on walk support for visually impaired persons has spread to a certain extent. A studded paving block and a sound-type signal are set on the sidewalk or on the road in many places. However, there are still problems such that the studded paving block is still installed at limited places, or the sound of the signal is obscure because of the environmental noise such as roaring traffic or heavy rain. Therefore, a system that can support visually impaired persons in more effective way is necessary, such as a system which finds a signal or a traffic sign automatically and provides its information to the visually impaired. In order to realize such a system, this paper proposes a method of detecting pedestrian signals and crosswalk signs by a camera and a computer which we call MY VISION system. In the proposed method, color information is used in the first step to restrict the search for traffic signal and sign candidates, and then the HOG feature is introduced to describe the candidates by feature vectors. Recognition of the signals and the signs is performed by applying randomized trees to the candidates. The color sign of the pedestrian signal is also discriminated by using color information. Experimental results are shown and the method is evaluated.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84959357672&origin=inward

Detection method for candidate regions of ground glass opacity on LIDC database using image features 査読有り

Yokota K., Kim H., Tan J., Ishikawa S., Tachibana R., Hirano Y., Kido S., Aoki T.

Kyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers 70 ( 8 ) J178 - J184 2016年01月

画像特徴を用いたLIDCデータベースからのすりガラス状候補陰影の検出法査読有り

横田佳祐, 金亨燮, タンジュークイ, 石川聖二, 橘理恵, 平野靖, 木戸尚治, 青木隆敏

映像情報メディア学会誌 70 ( 8 ) J178 - J184 2016年01月

記述言語：日本語掲載種別：研究論文（学術雑誌）

本稿では読影医師への負担軽減と読影精度の向上を目的とし，LIDCデータベース上の胸部CT画像からのGGO候補領域の自動抽出法を提案する．手法としては，まず肺野領域の抽出を行い，得られた肺野領域に対し，3D Line Filterによる血管・気管支領域の除去を行う．その後，濃度・勾配閾値処理により初期GGO候補領域を抽出する．そして初期GGO候補領域のセグメンテーションを行い，統計的特徴量を算出する．最後に，特徴量を基に識別器を構築し，最終的なGGO候補領域を決定する．本稿では，識別器としてニューラルネットワークとサポートベクターマシンの2種類を用い，両者の識別性能を比較する．実験では，提案法をLIDCデータベース上の胸部CT画像31症例に適用し，その結果に対する考察と有用性を述べる．

DOI： 10.3169/itej.70.J178

その他リンク： https://ci.nii.ac.jp/naid/130005165445

MY VISION システムを用いた指文字認識

平川学, タンジュークイ, 金亨燮, 石川聖二, Hamada SATOSI

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 29 ( 0 ) 41 - 44 2016年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

<p><i>This paper proposes a method of recognizing finger-spelled letters using the MY VISION system which contains a camera mounted on the chest of a human and a PC. A finger-speller performs the spelling to the camera. The images of the speller's hand are captured and hand areas are detected employing skin color. </i></p><p><i>For the detection, skin color calibration is performed in advance to cope with various illumination conditions and cluttered backgrounds. The shape and motion of the hand area are analyzed to recognize the finger-spelled letters. The PCA and the nearest neighbor method are employed for the recognition. The high recognition rate with the finger-spelled letters was achieved by the proposed system along with the high speed recognition. </i></p>

DOI： 10.24466/pacbfsa.29.0_41

その他リンク： https://ci.nii.ac.jp/naid/130007979809

高齢者見守りのための 3D-MHI を用いた人の動作認識

山下陽太郎, タンジュークイ, 金亨燮, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 29 ( 0 ) 33 - 36 2016年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

<p><i>Nowadays the population of elderly people who live alone has been growing continuously. When they encounter an accident at home, they won't be able to ask for help in the worst case. Under these circumstances, development of an automatic system to care elderly people living alone becomes increasingly important. We have been developing a computer vision system which finds abnormal motions of elderly people among their daily activities indoors. In this paper, we propose a new method of human motion representation called 3D-MHIs </i>(<i>3-dimensional-Motion History Image</i>)<i> based on FoE (Focus of Expansion) and a method of human motion recognition employing 3-D Hu moments by use of a single camera view. The method can recognize not only those motions parallel to a camera lens but also those motions toward depth direction. The effectiveness of the proposed method was experimentally shown.</i><i> </i></p>

DOI： 10.24466/pacbfsa.29.0_33

その他リンク： https://ci.nii.ac.jp/naid/130007979895

MSC-HOG 特徴量を用いた飛び出し歩行者の検出

小野祐汰, タンジュークイ, 金亨燮, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 29 ( 0 ) 37 - 40 2016年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

<p><i>In recent years, the development of Intelligent Transportation Systems (ITS) has become more active than ever and many methods of detecting people and automobiles using an in-vehicle camera have been proposed to date. However, a method of predicting a pedestrian's rush-out into a road still faces a mountain of challenges. In this paper, we propose a method of detecting a pedestrian and judging his/her direction of walk employing MSC-HOG, an improved version of the HOG, and AdaBoost. The prediction of a pedestrian's rush-out is carried out by tracking each pedestrian and judging the change of his/her walk directions. The effectiveness of the proposed method is verified by the experiments employing car videos.</i><i> </i></p>

DOI： 10.24466/pacbfsa.29.0_37

その他リンク： https://ci.nii.ac.jp/naid/130007979886

正規分布を用いたMY VISION 映像の振動補償

吉田直人, タンジュークイ, 金亨燮, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 29 ( 0 ) 45 - 48 2016年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

<p><i>In recent years,</i> <i>wearable devices have been developing rapidly as intelligent tools. According to MIC, the wearable devices market is expected to be 15 times larger in 2020 than 2013.</i> <i>Among them, a wearable camera with a PC has been drawing much attention as a substitute for a human real time visual processing system. However, because such a camera is used by being mounted on a user, the vibration generated by the movement of the user affects captured video images and hence resulting in the difficulty in the video processing.</i> <i>This paper proposes a method of compensating the vibration of a video images captured by a wearable camera by detecting the background movement obtained from optical flow analysis. Experimental results are shown and the method is evaluated.</i></p>

DOI： 10.24466/pacbfsa.29.0_45

その他リンク： https://ci.nii.ac.jp/naid/130007979807

移動カメラ映像からの物体の検出

松尾拓弥, タンジュークイ, 金亨燮, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 29 ( 0 ) 23 - 26 2016年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

<p><i>This paper presents a method of object detection using a moving camera. In order to perform background subtraction with a moving camera case, the proposed method estimates camera motion and creates a current background model from a previous background model using the camera motion. The background model is a set of a single Gaussian distribution. The method employs the upper and the lower bound of a threshold used for the judgment of background/foreground and color information to achieve precise detection of an object. We have confirmed the effectiveness of the proposed method by the experiments performed using some human motion videos. </i></p>

DOI： 10.24466/pacbfsa.29.0_23

その他リンク： https://ci.nii.ac.jp/naid/130007979883

スケール不変な顕著領域特徴に基づくCR画像からの指骨領域の自動位置合わせ査読有り

梶原，村上，金，タン，石川

バイオメディカル・ファジィ・システム学会誌 17 ( 2 ) 35 - 42 2015年12月

IGBTデバイスの目視検査支援システムの開発～不良品の細分類

結城，金，タン，石川，附田，大村

第３４回計測自動制御学会九州支部学術講演会 173 - 174 2015年11月

尤度を考慮した変形量に基づく指骨CR画像の位置合わせ

梶原，金，タン，石川，村上

第３４回計測自動制御学会九州支部学術講演会 53 - 54 2015年11月

遺伝的アルゴリズムを用いた手部CR像からの骨びらんの検出

村上，金，タン，石川

第２８回バイオメディカル・ファジィ・システム学会年次大会講演論文集 11 - 12 2015年11月

MSGVF SnakesによるMR画像からの指骨領域の自動抽出

重吉，村上，金，タン，石川

映像情報メディア学会技術報告（ヒューマンインフォメーション） 11 - 14 2015年11月

Development of a supporting system for visual inspection of IGBT device based on statistical feature and complex multi-resolution analysis 査読有り

Yuki, Kim, Tan, Ishikawa, Tsukuda, Omura

International Conference on Control, Automation and Systems 1551 - 1554 2015年10月

Posture estimation from Kinect image using RVM regression analysis 査読有り

Fujimura, Kim, Tan, Ishikawa

International Conference on Control, Automation and Systems 1540 - 1542 2015年10月

Development of image viewer for analyzing of temporal subtraction from chest CT images 査読有り

Kondo, Yoshino, Kim, Tan, Ishikawa, Murakami, Aoki, Tachibana, Hirano, Kido

International Conference on Control, Automation and Systems 1543 - 1546 2015年10月

Automatic segmentation of phalanges regions on MR images based on MSGVF Snakes 査読有り

Shigeyoshi, Murakami, Kim, Tan, Ishikawa

International Conference on Control, Automation and Systems 1547 - 1550 2015年10月

Comparison of feature extraction methods for head recognition 査読有り

Mudjirahardjo, P., Tan, J. K., Kim, H. S., Ishikawa, S.

Proc. 2015 Int. Electronics Symposium 124 - 128 2015年09月

Levenberg-Marquardt法を用いた３次元頭部CT・MR画像の自動位置合わせ

木崎，山村，金，タン，石川，山本

電信情報通信学会技術報告，ライフインテリジェンスとオフィス情報システム 51 - 54 2015年09月

Automatic Registration of Phalanges Regions in CR Images Based on Scale-Invariant Saliant Region Features 査読有り

Kajihara, Murakami, Kim, Tan, Ishikawa

IIAE International Conference on Inteligent Systems and Image Processing 2015 272 - 276 2015年09月

SVMによる胸部CT画像からのすりガラス状候補陰影の検出

金，横田，タン，石川，橘，平野，木戸，青木

映像情報メディア学会年次大会 2 pages 2015年08月

濃度勾配特徴を用いた下肢CTA画像からの血管領域の抽出

金，和田，タン，石川，山本

映像情報メディア学会年次大会 2 pages 2015年08月

スケール不変の顕著領域特徴に基づくCR画像からの支笏領域の自動位置合わせ法

梶原，村上，金，タン，石川

第34回日本医用画像工学会大会予稿集 10 pages 2015年07月

2段階AdaBoostを用いた経時的差分像からの結節状候補陰影の識別査読有り

田中修司, 金亨燮, タンジュークイ, 石川聖二, 村上誠一, 青木隆敏, 平野靖, 木戸尚治, 橘理恵

バイオメディカル・ファジィ・システム学会誌 17 ( 1 ) 9 - 16 2015年07月

記述言語：日本語掲載種別：研究論文（学術雑誌）

近年,肺癌による死亡者数の増加から,病変部の早期発見・治療が重要視されている.そのため,胸部MDCT画像を用いた精密検査などが進められている.MDCT画像は,癌検出が容易である反面,読影を行う画像枚数が多く,医師への負担増が懸念されている.そこでコンピュータ支援診断システムによる,読影医師への負担軽減などが期待されている.このシステムの一つとして,経時的変化を強調する経時的差分像技術がある.この技術より得られた差分像から結節状候補陰影を検出するための研究が行われ,医師への診断支援の実現が期待されている.本稿では,胸部MDCT画像の経時的差分像より直径20[mm]以下の結節状候補陰影を抽出し,特徴量解析に基づく2段階AdaBoostによる病変候補陰影を識別するためのシステムの開発を行い,実画像による性能評価を行ったところ,96.8[%]の識別率を得た.

その他リンク： http://ci.nii.ac.jp/naid/110009987542

Detecting a taxi from a video for visually handicapped people 査読有り国際誌

Nishimura, A., Tan, J. K., Kim, H., Ishikawa, S

Proc. SICE Annual Conf. 2015 2015年07月

画像特徴を用いたLIDCデータベースからのGGO候補領域の自動抽出

横田，金，タン，石川，橘，平野，木戸，青木

画像電子学会年次大会講演CDROM 4 pages 2015年06月

Groupwise surface correspondence using particle filtering 査読有り

Li G., Kim H., Tan J., Ishikawa S.

Proceedings of SPIE - The International Society for Optical Engineering 9443 2015年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

© 2015 SPIE.To obtain an effective interpretation of organic shape using statistical shape models (SSMs), the correspondence of the landmarks through all the training samples is the most challenging part in model building. In this study, a coarse-tofine groupwise correspondence method for 3-D polygonal surfaces is proposed. We manipulate a reference model in advance. Then all the training samples are mapped to a unified spherical parameter space. According to the positions of landmarks of the reference model, the candidate regions for correspondence are chosen. Finally we refine the perceptually correct correspondences between landmarks using particle filter algorithm, where the likelihood of local surface features are introduced as the criterion. The proposed method was performed on the correspondence of 9 cases of left lung training samples. Experimental results show the proposed method is flexible and under-constrained.

DOI： 10.1117/12.2179122

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84925431291&origin=inward

Human action representation and recognition: An approach to a histogram of spatiotemporal templates 査読有り

Ahsan S., Tan J., Kim H., Ishikawa S.

International Journal of Innovative Computing, Information and Control 11 ( 6 ) 1855 - 1867 2015年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

© 2015.The motion sequences of human actions have its own discriminating profile that can be represented as a spatiotemporal template like Motion History Image (MHI). A histogram is a popular statistic to present the underlying information in a template. In this paper a histogram oriented action recognition method is presented. In the proposed method, we use the Directional Motion History Images (DMHI), their corresponding Local Binary Pattern (LBP) images and the Motion Energy Image (MEI) as spatiotemporal template. The intensity histogram is then extracted from those images which are concatenated together to form the feature vector for action representation. A linear combination of the histograms taken from DMHIs and LBP images is used in the experiment. We evaluated the performance of the proposed method along with some variants of it using the renowned KTH action dataset and found higher accuracies. The obtained results justify the superiority of the proposed method compared to other approaches for action recognition found in literature.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84957063027&origin=inward

スケール不変な顕著領域特徴に基づくCR画像からの指骨領域の自動位置合わせ査読有り

梶原将太, 村上誠一, 金亨燮, タンジュークイ, 石川聖二

バイオメディカル・ファジィ・システム学会誌 17 ( 2 ) 35 - 42 2015年01月

記述言語：日本語掲載種別：研究論文（学術雑誌）

関節リウマチは代表的な骨疾患で,病態が進行すると関節の変形や機能障害が生じ患者のQOL(quality of life)を著しく低下させる.これらの診断には画像診断が有効であるが,医師の主観的な評価による診断精度のばらつきや,画像枚数の増加に伴う医師への負担増加などの問題が現存するため,これらの問題を克服し医師の負担を軽減することが重要な課題である.手のCR画像から定量的な評価支援を行うためのコンピュータ支援診断システムの必要性に応えるため,本稿では,手のCR画像からの指骨領域の自動位置合わせ法を提案する.指骨領域は同一被験者の過去と現在の手のCR画像から指骨領域をそれぞれ自動抽出したものを用いる.位置合わせ法としては,指骨領域の関心領域上のエントロピーに基づく顕著領域特徴を求め,過去と現在画像上の顕著領域特徴間の関係性から最適な変形量を導出し,剛体変形を与えることにより,両画像の位置合わせを行い,経時的な変化部分を検出する.提案手法を3症例の過去と現在画像に適用し,良好な結果を得た.

DOI： 10.24466/jbfsa.17.2_35

その他リンク： https://ci.nii.ac.jp/naid/110010047578

拡張型MHI を用いた人の動作認識査読有り

山下陽太郎, タンジュークイ, 金亨燮, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 28 ( 0 ) 7 - 8 2015年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

<p><i>In recent years, elderly people who live alone have been continually increasing. Its number is equivalent to 31.4 percent of elderly population in Japan. In the worst case, they cannot cry for help because of diseases or sudden injuries. Therefore a system for watching over elderly people is strongly requested in our society. In order to realize such a system, this paper proposes a method of recognizing human motions employing an expanded MHI which uses FoE (Focus of Expansion) to recognize the motion toward the depth direction. The proposed method was applied to the recognition of 6 human motions including 'fall down' and satisfactory results were obtained. </i></p>

DOI： 10.24466/pacbfsa.28.0_7

その他リンク： https://ci.nii.ac.jp/naid/130007998716

遺伝的アルゴリズムを用いた手部<sub> </sub>CR 像からの骨びらんの検出査読有り

村上誠一, 金亨燮, タンジュークイ, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 28 ( 0 ) 11 - 12 2015年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

<p><i>Erosion of the hands and feet is one of the major diagnostic criteria for rheumatoid arthritis (RA). However, a method to detect erosion using radiographs has not been reported until now. In this paper, we propose a method for the automatic detection of erosion using genetic algorithm from computed radiography (CR) images of the hand. The proposed method adopted for 13 images which obtained CR images of hand, and satisfactory results are obtained.</i><i> </i></p>

DOI： 10.24466/pacbfsa.28.0_11

その他リンク： https://ci.nii.ac.jp/naid/130007998608

色分布を用いた自己装着カメラからの指文字認識査読有り

平川学, 濱田聡, タンジュークイ, 金亨燮, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 28 ( 0 ) 207 - 208 2015年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

<p><i>This paper proposes a method of recognizing finger-spelled letters using a video obtained from a chest-mounted camera. A finger-speller performs the spelling to the camera mounted on his/her chest. The images of the speller's hand are captured and analyzed by a PC to recognize the letters. For the hand detection, the color distribution of the hand is examined initially to cope with various illumination conditions. The detected hand area is recognized using the PCA and the nearest neighbor method. Experimental results show satisfactory performance of the proposed method.</i><i> </i></p>

DOI： 10.24466/pacbfsa.28.0_207

その他リンク： https://ci.nii.ac.jp/naid/130007998560

画像特徴を用いた LIDC データベースからの GGO 候補領域の自動抽出

横田佳祐, 金亨燮, タンジュークイ, 石川聖二, 橘理恵, 平野靖, 木戸尚冶, 青木隆敏

画像電子学会年次大会予稿集 ( 一般社団法人画像電子学会 ) 43 ( 0 ) 7 - 7 2015年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

本稿では読影医師への負担軽減と読影精度の向上を目的とし，LIDCデータベース（Lung Image Database Consortium）[5]上の胸部CT画像からのGGO候補領域の自動抽出法を提案する．

DOI： 10.11371/aiieej.43.0_7

その他リンク： https://ci.nii.ac.jp/naid/130008081244

濃度勾配特徴を用いた下肢CTA画像からの血管領域の抽出

金亨燮, 和田幸大, タンジュークイ, 石川聖二, 山本晃義

映像情報メディア学会年次大会講演予稿集 ( 一般社団法人映像情報メディア学会 ) 2015 ( 0 ) 31B-1 2015年01月

担当区分：筆頭著者,　責任著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

In the field for visual screening, it is difficult task to detect blood vessel regions with high accuracy detection rates since blood vessel regions are in contact with bone region on the CTA images. In this paper, we propose a new method for segmentation of arterial area, which are obtained CTA based on gradient features, and satisfactory experimental results are obtained. Some experimental results are shown with discussion.

DOI： 10.11485/iteac.2015.0_31b-1

SVMによる胸部CT画像からのすりガラス状候補陰影の検出

金亨燮, 横田佳祐, タンジュークイ, 石川聖二, 橘理恵, 平野靖, 木戸尚冶, 青木隆敏

映像情報メディア学会年次大会講演予稿集 ( 一般社団法人映像情報メディア学会 ) 2015 ( 0 ) 31B-2 2015年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

In this paper, we develop a CAD (Computer Aided Diagnosis) system for assisting in visual screening. In the first stage, we calculate the statistical features such as density, shape and textures on segmented candidate regions. Finally, we implement a classifier based on support vector machine to distinguish final candidate regions. We applied the proposed method to 31 CT image sets in the Lung Image Database Consortium (LIDC) which is supplied by National Center Institute (NCI).

DOI： 10.11485/iteac.2015.0_31b-2

MSGVF SnakesによるMR画像からの指骨領域の自動抽出(視聴覚の基礎と応用,マルチモーダル,感性情報処理,一般)

重吉功嗣, 村上誠一, 金亨燮, タンジュークイ, 石川聖二

映像情報メディア学会技術報告 ( 一般社団法人映像情報メディア学会 ) 39.43 ( 0 ) 11 - 14 2015年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

手の骨の疾患,特に指骨の疾患における画像診断では,主に骨を直接撮影可能なX線やCTなどが用いられ,骨のすりへりや破壊の有無を診断する.しかし,MRIを用いた画像診断により,CTなどでは観察が困難な症状の診断に有効な場合があり,早期発見・診断が期待されている.このような背景から,近年の医療診断における画像診断の占める割合は向上しているが,画像診断では,医師による診断結果のばらつきや,画像枚数の増大による医師への負担増加が懸念されている.そのため,定量的な解析を行うための,コンピュータ支援診断システムが必要となる.本論文では,手のMR画像からの指骨の疾患の定量的な評価を行うための前段階としてMSGVF Snakesによる指骨領域のセグメンテーション手法を提案する.

DOI： 10.11485/itetr.39.43.0_11

MICCAI2014参加報告 (医用画像)

北坂孝幸, 中込啓太, 音丸格, 小田昌宏, 平野靖, 増谷佳孝, 金亨燮

電子情報通信学会技術研究報告 = IEICE technical report : 信学技報 ( 一般社団法人電子情報通信学会 ) 114 ( 311 ) 49 - 54 2014年11月

MICCAI2014参加報告

北坂、中込、音丸、小田、平野、増谷、金

信学技法 11 ( 311 ) 49 - 54 2014年11月

Temporal subtraction method for lung nodule detection on successive thoracic CT soft-copy images 査読有り

Aoki, Murakami, Kim, Fujii, Takahashi, Iki, Hayashida, Katsuragawa, Shiraishi, Koori

Radiology 271 ( 1 ) 255 - 261 2014年04月

Detection of a bicycle in video images using MSC-HOG feature 査読有り

Jung H., Ehara Y., Tan J., Kim H., Ishikawa S.

International Journal of Innovative Computing, Information and Control 10 ( 2 ) 521 - 533 2014年02月

記述言語：英語掲載種別：研究論文（学術雑誌）

Traffic accidents are decreasing due to the influence of technology advancement; however accidents still occur due to carelessness of drivers. Therefore, many researchers have been studying how to realize an advanced safety system. The Histograms of Oriented Gradients (HOG) feature is well known as a useful method of detecting a standing human in various kinds of backgrounds. Unlike a person, a bicycle can appear differently from various angles. In this paper, we propose a method of detecting a bicycle on the road using improved HOG feature named MSC-HOG feature and the Real-AdaBoost algorithm. Experimental results and evaluation show satisfactory performance of the proposed method. © 2014 ISSN 1349-4198.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84893415119&origin=inward

Videotaped obstacle extraction from a moving camera 査読有り

Qian S., Tan J., Kim H., Ishikawa S., Morie T., Shinomiya T.

International Journal of Innovative Computing, Information and Control 10 ( 2 ) 717 - 728 2014年02月

記述言語：英語掲載種別：研究論文（学術雑誌）

In automatic collision avoidance systems, the ability to detect obstacles is important. This paper proposes a method of automatic obstacles detection employing a camera mounted on a vehicle. Although various methods of obstacles detection have already been reported, they normally detect moving objects such as pedestrians and bicycles. In this paper, a method is proposed for detecting obstacles on a road, even if they are moving or static, by the use of background modeling and road region classification. Background modeling is often used to detect moving objects when a camera is static. In this paper, we apply it to a moving camera case in order to obtain foreground images. Then we calculate the camera motion parameters using the correspondence of feature points between two consecutive images and detect the road region using motion compensation. In this road region, we carry out regional classification. We can delete all objects which are not obstacles in the foreground images based on the result of the regional classification. In the performed experiments, it is shown that the proposed method is able to extract the shape of both static and moving obstacles in a frontal view from a car. © 2014 ISSN 1349-4198.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84893413251&origin=inward

B1-1 自己装着カメラ映像からの信号標識の検出(B1 学生,一般講演)

熊野貴大, タンジュークイ, 金亨燮, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 27 ( 0 ) 13 - 14 2014年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

In recent years, the walk support equipment for visually impaired persons has spread, and a studded paving block and a sound-type signal are installed on the sidewalk. However, there are still problems that the places of installation are few with the former, and the provided information becomes obscure with the latter in the noisy environment such as heavy traffic or heavy rain. In order to solve these difficulties, this paper proposes a method of detecting pedestrian signals and traffic signs at a pedestrian crossing from a self-wearable camera. Experimental results are shown and the method is evaluated.

DOI： 10.24466/pacbfsa.27.0_13

その他リンク： https://ci.nii.ac.jp/naid/110009895909

差分処理を用いた胸部CT画像上におけるすりガラス陰影の領域抽出査読有り

橘理恵, 平野靖, 徐睿, 木戸尚治, 金亨燮

Ｍｅｄｉｃａｌ　Ｉｍａｇｉｎｇ　Ｔｅｃｈｎｏｌｏｇｙ ( 日本医用画像工学会 ) 32 ( 3 ) 196 - 202 2014年01月

記述言語：日本語掲載種別：研究論文（学術雑誌）

胸部CT画像における肺腫瘍の中でも，すりガラス状陰影（ground glass opacity: GGO）を呈する腫瘍は良悪性鑑別が困難であるため経過観察されることが多い．この経過観察においては大きさの変化が診断の重要な指標のひとつとなる．そこで，本研究ではGGOを呈する腫瘍の領域をCT画像上から抽出する手法を提案する．GGOの中でも特に充実部を含まないpure GGOは二値化をベースとした抽出手法でのCT値に対する閾値を決めることが大変困難である．そのため，本手法ではシグモイド関数を用いて腫瘍を強調したのち，腫瘍のないスライスを背景とした背景差分処理を行うことで腫瘍の大まかな領域を抽出し，その後モルフォロジ処理等を用いて最終腫瘍領域の抽出を行った．LIDC（The Lung Image Database Consortium）の症例の中からGGOを呈する腫瘍を選択し，それらを用いて本手法の有用性を検討した結果を報告する．

DOI： 10.11409/mit.32.196

その他リンク： https://ci.nii.ac.jp/naid/130004679782

GGVFとシフトベクトルの平滑化による胸部MDCT画像の経時的差分法(視聴覚の基礎と応用,マルチモーダル,感性情報処理,一般)

芳野由利子, 金亨燮, タンジュークイ, 石川聖二, 村上誠一, 青木隆敏

映像情報メディア学会技術報告 ( 一般社団法人映像情報メディア学会 ) 38 ( 0 ) 53 - 56 2014年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

近年,MDCT画像の高精細化及び画像枚数の増大による読影負担を軽減し,病巣の認識を補助するCADの研究開発が進められている.その技術の一つとして,同一被験者の過去に撮影した画像と現在新しく撮影した画像との差分演算を行い,経時変化のみを強調する経時的差分像技術がある.経時的差分像技術では,撮影時の姿勢や呼吸などの体動による肺野領域の変形を考慮し,現在の画像に合わせて過去の画像を変形させ両画像間で差分演算を行う.しかし,肺野領域には血管などの組織が三次元的に分布しており,正確なレジストレーションが困難である.ミスレジストレーションによって画像内の隣接する各ピクセルのシフトベクトル(変形量と方向)の整合性が低い領域が発生し,それによって経時的差分画像上に多くのアーチファクトが発生する.本稿では,血管などの構造情報を考慮するGGVFを用いた画像間の位置合わせと,各ピクセルのシフトベクトルに対する平滑化を行い,レジストレーション精度向上及びアーチファクト軽減の有効性を検討する.

DOI： 10.11485/itetr.38.46.0_53

その他リンク： https://ci.nii.ac.jp/naid/110009885411

DSCと相互情報量を用いた3次元頭部CT・MR画像の自動位置合わせ法査読有り

山村雄太郎, 早田大地, 金亨燮, タンジュークイ, 石川聖二, 山本晃義

バイオメディカル・ファジィ・システム学会誌 ( バイオメディカル・ファジィ・システム学会 ) 16 ( 2 ) 19 - 27 2014年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

近年,異なるモダリティから得られる画像情報を融合し,一方のモダリティだけでは得られない情報を効率的に取らえることが可能になり,診断能の向上や効率化が図られるようになった.このような画像融合(フュージョン)では,画像の位置合わせが重要となるが,頭部のCT画像とMR画像を融合して新しい診断情報を得る場合,その多くがマニュアル操作によるため,医師の負担増加や結果のばらつきが問題となっている.本稿では,脳外科手術法の一つとして用いられる,サイバーナイフ手術時の治療計画作成時に必要となる,頭部CTとMR画像の位置合わせ法について述べる.手法としては,両画像からの皮膚領域の輪郭のみを強調した2値画像を生成し,その画像に対してDSC(Dice Similarity Coefficient)を最適化関数として位置合わせを行う.最後に,頭部の3次元画像に対して複数の関心領域(Volume Of Interest; VOI)を設置し,そのVOI内の相互情報量が最大となるようにアフィン変換を施すことにより最終的な位置合わせを行う.提案手法をCT,MR画像の各5症例に適用し,両画像の位置合わせを行った結果に対する考察と今後の展望について述べる.

DOI： 10.24466/jbfsa.16.2_19

その他リンク： https://ci.nii.ac.jp/naid/110009872264

CR画像からの指骨領域の自動抽出と経時変化抽出のための画像位置合わせ法査読有り

村上誠一, 保都祥道, 金亨燮, タンジュークイ, 石川聖二

バイオメディカル・ファジィ・システム学会誌 ( バイオメディカル・ファジィ・システム学会 ) 16 ( 2 ) 11 - 17 2014年01月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

手の骨の主な疾患として,関節リウマチや骨粗鬆症がある.現在,これらの疾患の診断には,CR, CT, MRIによって撮影された画像情報により,骨密度値の計測などの画像診断が主に行われている.しかし,画像診断を行う際の関心領域の設定には,医療従事者によるマニュアル操作が主として行われており,操作者の個人差の問題から自動による関心領域の設定法が求められているが,研究報告はほとんど見られないのが現状である.そのため,関心領域の設定結果の再現性や精度に欠け,病変部の見落としの危険性が懸念されている.これらの問題点から,コンピュータを用いた関心領域の自動抽出・定量評価を行うためのシステム構築への要望が高まっている.本稿では,指骨の経時変化の定量化を行うための関心領域の抽出と位置合わせ手法を提案する.手法としては,手のCR画像から関心領域である指骨の領域を自動抽出し,過去・現在画像の位置合わせを行い,両画像間での経時変化を定量化するための,コンピュータ支援診断(CAD)システムの開発を行う.提案法を,同一被験者の過去および現在の実CR画像セットに適用した結果について述べる.

DOI： 10.24466/jbfsa.16.2_11

その他リンク： https://ci.nii.ac.jp/naid/110009872263

B1-4 Improved Local Binary Patternsを用いたコーナ検出(B1 学生,一般講演)

津嶋隆, タンジュークイ, 金亨燮, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 27 ( 0 ) 19 - 20 2014年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

Detecting feature points from an image is an important technique in advanced image processing, and various feature point detection methods have been proposed to date. Attention is paid on the technique for detecting corner points in this paper. The proposed method detects corners of objects from an ILBP image by use of particular ILBP values and their spread patterns. The advantage of the proposed method is that, unlike Harris corner detector or FAST, it does not employ a threshold for the corner detection. Experimental results are shown with discussion.

DOI： 10.24466/pacbfsa.27.0_19

その他リンク： https://ci.nii.ac.jp/naid/110009895912

B1-3 画像上の顕著領域の検出法(B1 学生,一般講演)

桑田聖生, タンジュークイ, 金亨燮, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 27 ( 0 ) 17 - 18 2014年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

This paper proposes a method of detecting salient regions from an image. Detection of salient regions is important, because it is expected to improve the precision and the processing speed of object recognition through image processing. The proposed method employs human visual characteristics, i.e., complementary color harmony and Purkinje phenomenon to detect salient regions from an image. Experimental results show the effectiveness of the spatial redundancy considering global redundancy and local redundancy, based on the human visual characteristics.

DOI： 10.24466/pacbfsa.27.0_17

その他リンク： https://ci.nii.ac.jp/naid/110009895911

B1-2 自己視点映像による歩行者検出とそれに基づく特有情報の抽出(B1 学生,一般講演)

酒井隆一, タンジュークイ, 金亨燮, 石川聖二

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 27 ( 0 ) 15 - 16 2014年01月

記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

In recent years, researches on pedestrians detection have grown actively. Also, technology on pedestrians detection can be used in many fields. This paper proposes a method of peculiar information extraction based on pedestrians detection from self-wearable camera images. This method is more effective in cost than using more cameras, and applicable to various fields. Also, not only pedestrians detection but also their peculiar information extraction extends the flexibility of pedestrians' information processing. Experimental results are shown and the performance of the proposed method is evaluated

DOI： 10.24466/pacbfsa.27.0_15

その他リンク： https://ci.nii.ac.jp/naid/110009895910

Abnormal motion detection in an occlusive environment 査読有り

Mudjirahardjo P., Mudjirahardjo P., Tan J., Tan J., Kim H., Kim H., Ishikawa S., Ishikawa S.

Proceedings of the SICE Annual Conference 1398 - 1402 2013年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

We present a motion classification approach to detect movements of interest (abnormal motion) based on optical flow. By tracking all feature points of a moving human in successive frames, we calculate the coordinate space and create feature space. This is done directly from the intensity information without explicitly computing the underlying motions. It requires no foreground segmentation, no prior learning of activities, no motion recognition and no object detection. First, we determine the abnormal scene and speed by using the velocity histogram. Then by using k-means clustering over velocity orientation and magnitude, we determine the abnormal direction. The performance of the proposed method is experimentally shown.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84888608440&origin=inward

Development of a scaler stroke display system using computer vision 査読有り

Kuroiwa M., Kuroiwa M., Tan J., Tan J., Kim H., Kim H., Ishikawa S., Ishikawa S.

Proceedings of the SICE Annual Conference 1722 - 1723 2013年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper describes a dental trainings imulation sys tem with a jaw mode l using a computer vis ion technique for scaling and root planing (SRP) . The SRP is a cornerstone of treatment of periodontal diseases. We have difficulty in watching the movement of the working end of a scaler in a mouth. We propose a method of recovering the movement of the scaler employing a marker attached to the scaler. The performance of the proposed method is shown experimentally.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84888617888&origin=inward

Detection of underwater objects based on machine learning 査読有り

Tan Y., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 2104 - 2109 2013年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Side-scan and forward-looking sonars are some of the most widely used imaging systems for obtaining large scale images of the seafloor, and their use continues to expand rapidly with their increased deployment on autonomous underwater vehicles. However, it is difficult to extract quantitative information from the images generated from these processes, particularly for the detection and extraction of information on the objects within these images. We propose in this paper an algorithm for automatic detection of underwater objects in side-scan images based on machine learning employing adaptive boosting. Experimental results show that the method produces consistent maps of the seafloor.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84888617459&origin=inward

Detecting foreground objects by sequential background inference in a video captured by a moving camera 査読有り

Setyawan F., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 1699 - 1702 2013年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper proposes a technique to separate foreground objects from the background in a video taken with a moving camera. In this technique, the first image frame is considered as the first background. The next image frame is an image containing foreground objects. The next image frame is conditioned to have the same position with the first background. Adjustment of the position between the two image frames is done by determining the image features of the two images using Harris corner detector method. After image features have been obtained, feature point correspondence between the first background image and a subsequent image is searched for using Lucas-Kanade tracker. Outlier pairs are discarded by RANSAC. By the employment of a set of feature point pairs, the 2D projective transform is computed between the two images. Furthermore it is determined whether a pixel is included in the foreground or the background. Experimental results show satisfactory performance of the proposed method.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84888592154&origin=inward

Classifying 2D and 3D objects on a road employing the road plane 査読有り

Qian S., Tan J., Kim H., Ishikawa S., Morie T., Shinomiya T.

Proceedings of the SICE Annual Conference 1689 - 1692 2013年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Obstacles detection is an important technology in autonomous collision avoidance systems. We have already proposed an obstacle detection method based on background modeling. But this method detects 2D and 3D objects simultaneously. Since these 2D objects are not dangerous to driving, they will reduce the accuracy of detection if they are detected as obstacles. In order not to detect these 2D objects, this paper proposes a method for classifying 2D objects and 3D objects. The proposed method first estimates the camera motion parameters from the correspondences of feature points between two successive images. We calculate the 3D positions of the feature points on a detected object in the world coordinate system using triangulation. Then we estimate the parameters of the road plane using 3D positions of those feature points. Finally we calculate the distances from the 3D positions of the feature points to the road plane. Based on these distances, we can classify 2D objects and 3D objects. Experimental results show satisfactory performance of the proposed method.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84888611260&origin=inward

A novel pedestrian detector on low-resolution images: Gradient LBP using patterns of oriented edges 査読有り

Boudisa, Tan, Kim, Shinomiya, Ishikawa

IEICE Transactions on Information and Systems E96-D ( 12 ) 2882 - 2887 2013年12月

A global-local approach to saliency detection 査読有り

Boudissa A., Tan J., Kim H., Ishikawa S., Shinomiya T., Mikolajczyk K.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 8048 LNCS ( PART 2 ) 332 - 337 2013年09月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In this paper, we present a novel approach to saliency detection. We define a visually salient region with the following two properties; global saliency i.e. the spatial redundancy, and local saliency i.e. the region complexity. The former is its probability of occurrence within the image, whereas the latter defines how much information is contained within the region, and it is quantified by the entropy. By combining the global spatial redundancy measure and local entropy, we can achieve a simple, yet robust saliency detector. We evaluate it quantitatively and compare to Itti et al. [6] as well as to the spectral residual approach [5] on publicly available data where it shows a significant improvement. © 2013 Springer-Verlag.

DOI： 10.1007/978-3-642-40246-3_41

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84884496162&origin=inward

頭部CTA・MRA画像からの血管領域の抽出 (マルチメディア情報ハイディング・エンリッチメント)

前田真也, 山村雄太郎, 金亨燮, タンジュークイ, 石川聖二, 山本晃義

電子情報通信学会技術研究報告 = IEICE technical report : 信学技報 ( 一般社団法人電子情報通信学会 ) 113 ( 212 ) 81 - 84 2013年09月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

がんに対する治療法の一つとして放射線療法がある.放射線療法とは,腫瘍部に対して放射線を照射する治療法である.放射線治療では,正常組織の放射線被ばくを抑え,腫瘍部にのみ放射線を照射する必要がある.このような腫瘍位置の把握にはCT,MRなどの画像が用いられるが,各画像を重ね合わせて表示することにより,腫瘍位置を効果的に表示可能となる.画像間の重ね合わせについては,画像内の特徴点の情報を用いたものが代表的なものとしてあげられる.本稿では,放射線療法の対象のひとつである脳腫瘍の治療における,画像重ね合わせに用いる特徴点として,CTAおよびMRA画像から脳血管領域を抽出するための手法を提案する.頭部実CTA・MRA画像に対して提案手法を適用し,その有用性について検討した.

その他リンク： https://ci.nii.ac.jp/naid/110009783192

頭部CTA・MRA画像からの血管領域の抽出 (画像工学)

前田真也, 山村雄太郎, 金亨燮, タンジュークイ, 石川聖二, 山本晃義

電子情報通信学会技術研究報告 = IEICE technical report : 信学技報 ( 一般社団法人電子情報通信学会 ) 113 ( 211 ) 81 - 84 2013年09月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

がんに対する治療法の一つとして放射線療法がある.放射線療法とは,腫瘍部に対して放射線を照射する治療法である.放射線治療では,正常組織の放射線被ばくを抑え,腫瘍部にのみ放射線を照射する必要がある.このような腫瘍位置の把握にはCT,MRなどの画像が用いられるが,各画像を重ね合わせて表示することにより,腫瘍位置を効果的に表示可能となる.画像間の重ね合わせについては,画像内の特徴点の情報を用いたものが代表的なものとしてあげられる.本稿では,放射線療法の対象のひとつである脳腫瘍の治療における,画像重ね合わせに用いる特徴点として,CTAおよびMRA画像から脳血管領域を抽出するための手法を提案する.頭部実CTA・MRA画像に対して提案手法を適用し,その有用性について検討した.

その他リンク： https://ci.nii.ac.jp/naid/110009817858

頭部CTA・MRA画像からの血管領域の抽出 (ライフインテリジェンスとオフィス情報システム)

前田真也, 山村雄太郎, 金亨燮, タンジュークイ, 石川聖二, 山本晃義

電子情報通信学会技術研究報告 = IEICE technical report : 信学技報 ( 一般社団法人電子情報通信学会 ) 113 ( 210 ) 81 - 84 2013年09月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

がんに対する治療法の一つとして放射線療法がある.放射線療法とは,腫瘍部に対して放射線を照射する治療法である.放射線治療では,正常組織の放射線被ばくを抑え,腫瘍部にのみ放射線を照射する必要がある.このような腫瘍位置の把握にはCT,MRなどの画像が用いられるが,各画像を重ね合わせて表示することにより,腫瘍位置を効果的に表示可能となる.画像間の重ね合わせについては,画像内の特徴点の情報を用いたものが代表的なものとしてあげられる.本稿では,放射線療法の対象のひとつである脳腫瘍の治療における,画像重ね合わせに用いる特徴点として,CTAおよびMRA画像から脳血管領域を抽出するための手法を提案する.頭部実CTA・MRA画像に対して提案手法を適用し,その有用性について検討した.

その他リンク： https://ci.nii.ac.jp/naid/110009784523

Road region estimation and obstacles extraction using a monocular camera 査読有り

Qian S., Tan J., Kim H., Ishikawa S., Morie T., Shinomiya T.

International Journal of Innovative Computing, Information and Control 9 ( 9 ) 3561 - 3572 2013年07月

記述言語：英語掲載種別：研究論文（学術雑誌）

An obstacles detection method is proposed which employs a camera mounted on a vehicle. Although various methods of obstacles detection have already been reported, they normally detect only moving obstacles and not static obstacles, and a detected obstacle is represented by a rectangular frame that surrounds this obstacle, which does not provide shape of the obstacle. In this paper, a method is proposed for detecting obstacles on a road, irrespective of moving or static, by use of background modeling and road region detection. The output of the proposed method is the shape of obstacles. Background modeling is often used to detect moving objects when a camera is static. In this paper, we apply it to a moving camera case in order to obtain foreground images. Then we extract the road region using SVM. In this road region, we carry out region classification. We can delete all the objects which are not obstacles in the foreground images based on the result of the region classification. In the performed experiments, it is shown that the proposed method is able to extract the shape of both static and moving obstacles when a car is driving. © 2013 ISSN 1349-4198.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84880054018&origin=inward

非剛体画像位置合わせ法による経時的差分像上の結節状候補陰影の検出と表示用ソフトウェアの開発 (パターン認識・メディア理解)

池田由利子, 時佐拓弥, 前田真也, 金亨燮, タンジュークイ, 石川聖二

電子情報通信学会技術研究報告 = IEICE technical report : 信学技報 ( 一般社団法人電子情報通信学会 ) 113 ( 75 ) 1 - 4 2013年06月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

近年,医療現場では医師の診断をサポートするコンピュータ支援診断(CAD)の開発が注目されている.その一つとして同一被験者の過去に撮影した画像と新しく撮影した画像間の差分演算を行い,それにより新しく発症した病変部や既存陰影を強調表示する経時的差分像を医師に提示することにより,画像診断の効率化を図る試みがなされている.本論文では、胸部CT画像を対象とした経時的差分像技術の非剛体レジストレーション手法により得られる経時的差分像における結節状候補陰影の検出を行うCADシステムを提案する.提案するCADシステムでは,ANN識別器による最終的な結節状陰影候補領域の検出を行い,3D表示するためのソフトウェアの開発を行った.

その他リンク： https://ci.nii.ac.jp/naid/110009779161

濃度勾配情報を用いた胸部MDCT画像における経時的差分手法査読有り

前田真也, 三宅徳朗, 金亨燮, タンジュークイ, 石川聖二, 村上誠一, 青木隆敏

計測自動制御学会論文集 = Transactions of the Society of Instrument and Control Engineers ( 計測自動制御学会 ) 49 ( 4 ) 461 - 468 2013年04月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

Recently, CAD system has been introduced as a visual screening method in medical field. Also, temporal subtraction method, which enhances temporal changes such as the abnormalities on the image, is implemented as one of technique for CAD by subtracting previous image from current one. Image registration techniques are required for the temporal subtraction method. If the accuracy of image registration is insufficiently, normal structure elements which should be removed are remained on subtraction image. In this study, a novel image registration method for temporal subtraction technique on thoracic MDCT images has been developed to reduce the subtraction artifacts by using intensity gradient information. The proposed method has been applied to thoracic MDCT images, and its efficiency has been indicated.

DOI： 10.9746/sicetr.49.461

その他リンク： https://ci.nii.ac.jp/naid/10031166608

Detection of artery regions in lower extremity arteries from non-enhanced MR imaging based on particle filter algorithms 査読有り

Koga Y., Yamamoto A., Yamamoto A., Kim H., Tan J., Ishikawa S.

Journal of Advanced Computational Intelligence and Intelligent Informatics 17 ( 2 ) 318 - 323 2013年03月

記述言語：英語掲載種別：研究論文（学術雑誌）

Recently, the arteries sclerosis obliterans (ASO) or called peripheral arterial disease (PAD) typically caused by chronic ischemia of limbs increases remarkably. As one of the diagnosis methods, the image diagnosis methods such as MR image are applied in medical fields. In this paper, we propose a vascular extraction method using fresh blood imaging (FBI) method, as well as apply it to computer aided diagnosis (CAD) system. Especially, to prevent the spread outside of the region and improve the segment accuracy of peripheral artery areas, we introduce particle filter algorithms. We performed our method on automatic artery regions detection using non-enhanced MR images. Furthermore, we compared the extracted results to gold standard data and analyzed accuracy by receiver operating characteristic (ROC). The effectiveness of our proposed method and satisfactory of its detected accuracy were confirmed.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84879329594&origin=inward

血管構造情報を用いた三次元胸部CT画像における非剛体レジストレーション法査読有り

前田真也, 金亨燮, タンジュークイ, 石川聖二, 村上誠一, 青木隆敏

電子情報通信学会論文誌. D, 情報・システム = The IEICE transactions on information and systems (Japanese edition) ( 一般社団法人電子情報通信学会 ) 96 ( 3 ) 733 - 742 2013年03月

担当区分：責任著者記述言語：日本語掲載種別：研究論文（学術雑誌）

胸部CT画像を用いた肺結節の発生や経時変化の定期的な観察は,肺がんの早期発見及び良悪性判別において重要である.定期診断においては,現時点で撮影された画像(現在画像)と,過去の検診で撮影された画像(過去画像)の比較読影が行われている.このとき,現在画像と過去画像間のレジストレーションを行い,肺野内の構造の対応関係を求めることにより,経時的な変化部分を認識しやすくなる.しかし,肺野領域内には血管などの細かな組織が三次元的に分布しており,細部における高精度なレジストレーションが困難となる.そこで本論文では,血管構造情報を考慮した新たな非剛体レジストレーション手法を提案する.提案手法では血管構造情報として,血管構造らしさ,及び血管方向の情報を用い,画像間の血管構造情報の類似度を定義する.血管構造情報の類似度を用いた非剛体レジストレーションを行うことにより,位置あわせ精度の改善を試みた.提案手法を胸部実MDCT画像に適用し,正規化相互相関値(NCC)及びSum of Squared Difference(SSD)によるレジストレーション精度の評価を行った.その結果,従来手法ではNCCが0.712,SSDが185であったのに対し,提案手法では,NCCが0.800,SSDが153となり精度の改善が確認された.

その他リンク： https://ci.nii.ac.jp/naid/110009593041

A parameterization based correspondence method for PDM building 査読有り

Li G., Kim H., Tan J., Ishikawa S.

Journal of Advanced Computational Intelligence and Intelligent Informatics 17 ( 1 ) 18 - 26 2013年01月

記述言語：英語掲載種別：研究論文（学術雑誌）

Place-march of corresponding landmarks is one of the major factors influencing 3D Points Distribution Model (PDM) quality. In this study, we propose a semi-automatic correspondence method based on surface parameterization theory. All the training sets are mapped into a spherical domain previously. The rotation transformation of training samples is regarded as spherical rotation of their maps. We solve it by comparing the density distribution of surface map of training sample with respect to the reference model. Simultaneously, the corresponding landmarks across the whole training set are marketed depending on the spherical coordinates on parameter domain. In this paper, we also compared the corresponding results with two constraint conditions of spherical conformal mapping: 3 datum points constrain and zero-mass constrain. Experimental results are given for left lung training sets of 3D shapes. The mean result with the 3 datum points constraint and the zero mass-center constraint was 21.65 mm and 20.19 mm respectively.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84879455289&origin=inward

Recognizing human actions using histogram of local binary patterns 査読有り

Ahsan S., Tan J., Kim H., Ishikawa S.

2013 IEEE/SICE International Symposium on System Integration, SII 2013 54 - 59 2013年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Human action recognition from video clips has become an active research field in recent years. Each action has its unique shape and a motion sequence can be suitably represented by a histogram. In this paper a histogram based action recognition method is presented. Motion history images are a good spatiotemporal template for action representation. In the present method, we use local binary patterns of directional motion history images for the histogram representation. We measured the performance of the proposed method along with some variants of it by employing KTH action dataset and found higher accuracy. The presented results also justify the superiority of the proposed method compared to other approaches for action recognition found in literature. © 2013 IEEE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84902492748&origin=inward

Automatic classification of seabed sediments based on HLAC 査読有り

Tan Y., Tan J., Kim H., Ishikawa S.

2013 IEEE/SICE International Symposium on System Integration, SII 2013 653 - 658 2013年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Understanding the distribution of seafloor sediment using a side-scan sonar is very important to grasp the distribution of seabed resources. This task is traditionally carried out by a skilled human operator. However, with the appearance of Autonomous Underwater Vehicles, automated processing is now needed to tackle the large amount of data produced and to enable on the fly adaptation of the missions and near real time update of the operator. We propose in this paper a method that applies a subspace method and higher-order local auto-correlation feature to the acoustic image provided by the side-scan sonar to classify seabed sediment automatically. In texture classification, the proposed method outperformed other methods such as gray level co-occurrence matrix and Local Binary Pattern operator. Experimental results show that the proposed method produces a consistent map of a seafloor. © 2013 IEEE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84902467004&origin=inward

C-1-1 パーティクルフィルタを用いたFBI画像からの下肢血管領域の自動抽出(治療支援(1))

金亨燮, 古賀結子, タンジュークイ, 石川聖二, 山本晃義

バイオメディカル・ファジィ・システム学会大会講演論文集 ( バイオメディカル・ファジィ・システム学会 ) 26 ( 0 ) 29 - 32 2013年01月

担当区分：筆頭著者記述言語：日本語掲載種別：研究論文（研究会，シンポジウム資料等）

Recently, the Arterio sclerosis obliterans (ASO) caused by chronic ischemia of limbs increases. As one of the diagnosis methods to detect the ASO, MR image which is obtained fresh blood imaging (FBI) is used for visual screening in medical fields. In this paper, we propose a method to detect blood vessel regions in legs from the FBI images. Especially, we developed for preventing the spread outside of the region and the segment accuracy in peripheral artery areas that is improved by using particle filter algorithms. We performed our proposed method on to real MR images which is obtained by FBI method and satisfactory detection results of artery regions are achieved.

DOI： 10.24466/pacbfsa.26.0_29

その他リンク： https://ci.nii.ac.jp/naid/110009830344

ブースティングによる機械学習に基づく海底物体の検出査読有り

丹康弘, タンジュークイ, 金亨燮, 石川聖二

日本船舶海洋工学会論文集 ( 社団法人日本船舶海洋工学会 ) 18 ( 0 ) 115 - 121 2013年01月

記述言語：日本語掲載種別：研究論文（学術雑誌）

Side-scan and forward looking sonars are some of the most widely used imaging systems for obtaining large scale images of a seafloor, and their use continues to expand rapidly with their increasing deployment on Autonomous Underwater Vehicles. However,it is difficult to extract quantitative information from the images generated from these processes, in particular, for the detection and extraction of information on the objects within these images. We propose in this paper an algorithm for automatic detection of underwater objects in side-scan images based on machine learning employing adaptive boosting. Experimental results show that the method produces consistent maps of a seafloor.

DOI： 10.2534/jjasnaoe.18.115

その他リンク： https://ci.nii.ac.jp/naid/130003393937

3-D recovery of a non-rigid object from a single camera view by piecewise recovery and synthesis 査読有り

Ishikawa S., Tan J., Kim H., Ishikawa S.

Proceedings - International Conference on Pattern Recognition 1443 - 1446 2012年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper proposes a novel technique for 3-D recovery of a non-rigid object, such as a human in motion, from a single camera view. To achieve the 3-D recovery, the proposed technique performs segmentation of an object under deformation into respective parts which are regarded as rigid. For high accuracy segmentation, multi-stage learning and local subspace affinity are employed in this stage. Each part recovers its 3-D shape by applying the factorization method to it. This is the initial solution. The shape of each part is then refined by applying a quadratic model to the initial solution. The entire 3-D recovery of the object is finally performed using the common points among the segmented parts. The experiments employing a synthetic non-rigid object and real human motion data show effectiveness of the proposed technique. © 2012 ICPR Org Committee.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84874574364&origin=inward

Detection of bone regions on legs from CTA image using concentration gradients 査読有り

Wada K., Yamamoto A., Kim H., Tan J., Ishikawa S.

International Conference on Control, Automation and Systems 1818 - 1821 2012年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Recently, radiologists can easily screening with high accuracy based on CTA (Computed Tomography Angiography) images and/or MRI (Magnetic Resonance Imaging) which is supported by medical imaging technology. On the other hand, visual screening makes burden to doctors which cause increasing the images and limited times. To avoid these problems, CAD (Computer Aided Diagnosis or Detection) system is developed by reducing the burdens and to improve the diagnostic accuracy. Problems of diagnosis on legs from CTA are burdens to doctors and detection of blood vessel by manual. In the field for visual screening, it is difficult task to detect blood vessel regions with high accuracy detection rates since blood vessel regions are in contact with bone region on the CTA images. In this paper, we propose a new method for segmentation of arterial area, which are obtained CTA based on image processing by concentration gradients, and satisfactory experimental results are obtained. Some experimental results are shown with discussion. © 2012 ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84872535204&origin=inward

Classification of lung nodules on temporal subtraction image based on statistical features and improvement of segmentation accuracy 査読有り

Miyajima T., Tokisa T., Maeda S., Kim H., Tan J., Ishikawa S., Murakami S., Murakami S., Aoki T.

International Conference on Control, Automation and Systems 1814 - 1817 2012年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Recently, thorax MDCT images are used in visual screening for early detection of lung nodules. Radiologists can easily detect lung nodules on images, but it has enormous images and load of radiologist for visual screening. To reduce the load of radiologist and improve the detection accuracy, a CAD (Computer Aided Diagnosis) system is expected from medical fields. In the medical image processing fields, some related works are reported to develop the CAD system including temporal subtraction technique as helpful technical issues. In this paper, we propose a classification of lung nodules on temporal subtraction image based on image processing technique. At first, the candidate regions including nodules are detected by the multiple threshold technique in terms of the pixel value on the temporal subtraction images. Then, we remove vessel regions on nodules by the most suitable threshold technique and watershed method. Also we remove the false positives which are caused by mis-registration using selective enhancement filter, rule-base method and artificial neural networks. In this paper, we illustrate some experimental result which applied our algorithm to 31 chest MDCT cases including lung nodules. © 2012 ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84872550943&origin=inward

Applying MSC-HOG feature to the detection of a human on a bicycle 査読有り

Jung H., Ehara Y., Tan J., Kim H., Ishikawa S.

International Conference on Control, Automation and Systems 514 - 517 2012年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Traffic accidents are decreasing under the influence of technology advancement. But the problems still remain that accidents occur due to carelessness of drivers. Therefore many researchers have been still studying to realize an advanced safety system. The Histograms of Oriented Gradients (HOG) feature is well known as a useful method of detecting a standing human in various kinds of the background. Unlike a human, a bicycle changes its appearance variously according to viewpoints. Hence, it is more difficult than detecting a human. In this paper, we propose a method of detecting a human on a bicycle using the Multiple-size Cell HOG (MSC-HOG) feature and the RealAdaboost algorithm. Experimental results and evaluation show satisfactory performance of the proposed method. © 2012 ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84872579239&origin=inward

A temporal subtraction method for thoracic CT images using non rigid warping technique 査読有り

Tokisa T., Kim H., Tan J., Ishikawa S., Moon Y., Yoon S., Kim W.

International Conference on Control, Automation and Systems 1809 - 1813 2012年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

A temporal subtraction technique which is subtracted from previous image to current one is introduced as powerful tools in medical fields to diagnose abnormalities. It provided a computer aided diagnosis (CAD) tools on visual screening. Radiologist can detect lesions on image by compare the two images. It is because the subtraction image can enhance the temporal changes, such as shaped of new lesions and/or the temporal changes in existing abnormalities by removing most of the normal background structures by subtraction of a previous image from a current one. There are some technical reports to register the different images until now. But subtraction artifacts are still remained which are caused by mis-registration. In this paper, we propose a new method for temporal subtraction method on thoracic MDCT images using non-rigid image warping techniques based on free form deformation (FFD). We applied our method to two clinical cases of chest CT image sets and compare to conventional methods in terms of computational cost and accuracy. © 2012 ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84872541007&origin=inward

A camera-computer system to support safe walking of a blind person 査読有り

Kanayama A., Tan J., Kim H., Ishikawa S.

International Conference on Control, Automation and Systems 511 - 513 2012年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

It is necessary for a blind person to know the locations of the pedestrians or moving objects around him/her on a walk road for his/her safe walking. This paper proposes a system for detecting pedestrians or moving objects in front of a person on a walk road by a camera mounted on his/her body. The system judges if a moving object in front of the person is going to hit him/her in the near future by analyzing the motion vectors acquired on the camera images. In case there is a fear of hit, the system tells it to the person by sound. In this way, the proposed system helps safe walking of a blind person. Some experimental results are shown with discussion. © 2012 ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84872511199&origin=inward

Temporal subtraction method for abdominal contrast and non-contrast image based on image matching techniques 査読有り

Minashima M., Ogihara S., Kim H., Tan J., Ishikawa S., Murakami S., Murakami S., Aoki T.

International Conference on Control, Automation and Systems 1805 - 1808 2012年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Recently, the death rate due to the liver cancer rises remarkably. To reduce the rate, the early detection of the disease is important. To detect diseases in early stage which are concern cancer, image diagnosis such as CT image is used in medical fields. On the other hand, the burden to a radiologist becomes increase. Therefore, the development of a system reducing the burden of the radiologist is important. In order to diagnose abnormalities based on medical imaging there are some reports. But, there is no report which is concern with detecting abnormality on liver disease based on temporal subtraction technique for abdominal CT image. As one of the methods to analyze abnormalities on visual screening, temporal subtraction technique is useful. This technique subtracts past image to current one. To obtain the good performance based temporal subtraction technique, image registration is most important task. In this paper, we propose a registration method for liver CT image using voxel matching techniques. We describe our registration method from two CT image which obtained deference time series and shows experimental results with discussion. © 2012 ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84872554258&origin=inward

A method for cardiac region segmentation on ultrasound images based on particle filter algorithm 査読有り

Kim H., Sugandi B., Tan J., Ishikawa S.

Proc. of the IADIS Int. Conf. Computer Graphics, Visualization, Computer Vision and Image Processing 2012, CGVCVIP 2012, Part of the IADIS MCCSIS 2012 157 - 162 2012年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper presents a contour segmentation of cardiac region on ultrasound images viewed as a tracking framework. We develop a tracking model for contour segmentation and estimate the image contour using particle filter algorithm. The tracking model is initialized by projecting equip spaced radii from center point of the contour to the tracking boundary. The motion of each particle on the radii is governed by a system model. The samples likelihood is measured based on the gradient intensity or edge of the contour. Then the estimated contour is measured based on the mean estimate of the samples likelihood. Our proposed method is implemented to the ultrasound images of a cardiac and the satisfactory results are achieved. © 2012 IADIS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84887293035&origin=inward

Human motion recognition employing infrared radiation camera images 査読有り

Hiroshima T., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 392 - 395 2012年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper proposes a method of motion recognition in real night environment employing an infrared radiation camera. We record a video employing an infrared radiation camera to perform motion recognition at night. We use Directional Motion History Images (DMHIs) and Directional Motion Energy Images (DMEIs) for human motion representation. For the recognition of the motions, we employ the eigenspace method. In the experiment, it was confirmed that the human motion recognition at night is possible. © 2012 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84869392576&origin=inward

Japanese finger-spelling recognition using a chest-mounted camera 査読有り

Nagasue A., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 909 - 912 2012年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper proposes a technique for recognizing Japanese finger-spelling using a sign language user's chest-mounted camera. Unlike existent systems, the technique employs a chest-mounted camera attached to a sign language user himself/herself and recognizes his/her sign language through the captured images of Japanese finger-spelling. We use a hand area picture of his/her hand and the MHIs (Motion History Images) for the Japanese finger-spelling recognition. For the recognition method, we employ the ICA (Independent Component Analysis). Furthermore, in order to recognize the Japanese finger-spelling from an animation (an image sequence), a character segmentation technique is also proposed. The performance of the proposed system is shown experimentally. © 2012 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84869417788&origin=inward

Arterial hemodynamic analysis on non-enhanced magnetic resonance angiogram using optical flow 査読有り

Yamamoto A., Kim H., Tan J., Ishikawa S.

Artificial Life and Robotics 17 ( 1 ) 102 - 106 2012年10月

記述言語：英語掲載種別：研究論文（学術雑誌）

© ISAROB 2012. Peripheral arterial disease (PAD) is one of the reasons caused to the lower extremity atherosclerotic disease. Its diagnosis is needed to obtain much kind of the information of vascular morphology as well as the blood flow information based on hemodynamics. The diagnosis of the PAD using magnetic resonance imaging (MRI) equipment without contrast medium is available as a useful visual screening in clinical practice. In this paper, we propose a novel method for visualizing hemodynamics to arterial images obtained by a non-contrast enhanced magnetic resonance angiography (MRA) based on the Lucas–Kanade optical flow with the image pyramid processing, and satisfied experimental results are obtained.

DOI： 10.1007/s10015-012-0022-8

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84863670045&origin=inward

Motion history image: its variants and applications 査読有り

Ahad, Tan, Kim, Ishikawa

Machine Vision and Applications 23 255 - 281 2012年03月

経時的差分像技術を用いた胸部CT画像からの結節状候補陰影領域の自動抽出

時佐、三宅、前田、金、タン、石川、村上、青木

第4回呼吸機能イメージング研究会学術集会 77 2012年02月

An effective directional motion database organization for human motion recognition 査読有り

Eftakhar, Tan, Kim, Ishikawa

Intl. Journal of Innovative Computing, Information and Control 8 ( 2 ) 1359 - 1370 2012年02月

An effective directional motion database organization for human motion recognition 査読有り

Ashik Eftakhar S., Tan J., Kim H., Ishikawa S.

International Journal of Innovative Computing, Information and Control 8 ( 2 ) 1359 - 1370 2012年02月

記述言語：英語掲載種別：研究論文（学術雑誌）

Automatic recognition of human motions has an increasing demand in the recent visionary world. However, with the registration of large number of motions from varying viewpoints, the necessity for an effective motion database for recognition has become a vital issue. In the context of motion database development, this paper proposes a directional database organization for human motion recognition. This organization partitions the motion database into several sub-databases on the basis of camera orientation. Separate feature spaces are constructed, and correspondingly directional sub-databases are built, leading to the constitution of the complete motion database. The directionally similar but semantically different motions are properly distinguished. To show the ro-bustness of the proposed organization for recognizing human motions, a set of motions captured from varying viewpoints is analyzed. An eigenspace representation is employed as a generic feature space that sufficiently characterizes the motion features. Motion History Image (MHI) and Exclusive-OR (XOR) image representations are used as motion templates where MHI is found performing better than XOR image. The experimental re-sults show high-level of satisfactory performance and claim the significant improvement over earlier developed systems. © 2012 ICIC International.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84863081707&origin=inward

Japanese Finger-spelling Recognition Using a Chest-mounted Camera 査読有り

Nagasue, Tan, Kim, Ishikawa

The 17th International Symposium on Artificial Life and Robotics 1059 - 1062 2012年01月

Automatic segmentation of liver region from non-contrast and contrast CT images employing tree-structural image transformations 査読有り

Komatsu M., Li G., Kim H., Tan J., Ishikawa S., Yamamoto A., Yamamoto A.

Proceedings of the 16th International Symposium on Artificial Life and Robotics, AROB 16th'11 763 - 766 2011年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In the medical image processing field, segmentation from the CT image is one of the most important problems for analyzing the abnormalities and diagnosis on visual screening. Many related segmentation techniques have been developed for automatic extraction of ROI. It is however, there are still no fully automatic segmentation methods that are generally applicable to ROI based on CT image set. In this paper, we present a technique for automatic extraction of liver region on the MDCT images employing automatic construction of tree-structural image transformation (ACTIT). We propose a new technique for extraction of organs employing ACTIT with non-contrast and contrast image set in order to introduce temporal change information. We apply the proposed technique to three abdominal image set and satisfactory segmentation results are achieved. © 2011 ISAROB.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84866687359&origin=inward

Moving objects detection at an intersection by sequential background extraction 査読有り

Sonoda S., Tan J., Kim H., Ishikawa S., Morie T.

International Conference on Control, Automation and Systems 1752 - 1755 2011年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Recently, there are many traffic accidents in turning right at an intersection. They are mainly caused by a driver's oversight of pedestrians and motorcycles that are occluded by oncoming cars. Therefore a system is necessary to detect moving objects such as oncoming cars and pedestrians at an intersection, and warn a vehicle driver. This paper describes a technique for detecting moving objects in turning right at an intersection when vehicle is stopping. Moving objects are detected by Mixture of Gaussians (MoG). In addition, we distinguish cars from pedestrians using the difference of the area size and the aspect ratio of detected objects. The object which is classified as a pedestrian is tracked using Lucas-Kanade Tracker. If the detected cars and pedestrians overlap or a car completely obscures pedestrians, we perform the estimation of pedestrian's location by using the information on past frames. By doing this, it is possible to detect pedestrians that drivers are actually difficult to see. The performance of the proposed technique was examined employing car videos and satisfactory results were obtained. © 2011 ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84863041517&origin=inward

Extraction of individual pedestrians employing stereo camera images 査読有り

Kawabe M., Tan J., Kim H., Ishikawa S., Morie T.

International Conference on Control, Automation and Systems 1744 - 1747 2011年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

A traffic accident occurs mainly because of carelessness of a driver. To prevent such an accident, many systems which detect pedestrians from vehicle are proposed. But most of these systems give a warning to a driver every time when a pedestrian is detected. The effect of warning becomes lower if the frequency of warning increases in a busy area. The objective of this study is to develop a technique for detecting pedestrians from a vehicle and to arrange them according to their potential risks by analyzing their behaviors. In this paper, we propose a technique for segmenting pedestrians employing stereo camera images to analyze a pedestrian's behavior. First, foreground regions are extracted employing background estimation using the mixture of Gaussian model. After extracting foreground regions in an input image, these regions are segmented by using the distance information. By using this method, each shape of the detected pedestrians can be obtained even when pedestrians are overlapping with each other. The proposed technique was examined experimentally employing real video images and satisfactory results were obtained. © 2011 ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84863049839&origin=inward

Direction-oriented human motion recognition with prior estimation of directions 査読有り

Eftakhar S., Tan J., Kim H., Ishikawa S.

IECON Proceedings (Industrial Electronics Conference) 4226 - 4231 2011年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

As the facilities are becoming available with the advent of the state-of-art technologies, the necessity of man-machine interaction systems is growing day-by-day. Within various applications of such a system, one of the most promising applications in the field of computer vision is the understanding and the interpretation of human motion or behavior in a scene. Direction-oriented motion capture of a person performing some tasks is an important issue in developing a human motion recognition system, since an intelligent system should also be incorporated with the directional information. We propose a direction-oriented motion recognition approach that makes use of the directional information by prior estimation. This reduces the processing time of the system by excluding unnecessary searching for the most similar motions. In this approach, direction-wise motions are clustered within the feature space in order to make the direction estimation easier. Each motion is converted to individual template, namely Motion History Image (MHI) and Exclusive-OR (XOR) Image, by extracting distinguishable features from the video-clips containing the motions. A Structured Motion Database (SMoDB) is developed to match an unlabeled motion against the pre-stored motions. Experiments are conducted on an Avatar dataset and significant improvements in the results are noticed. © 2011 IEEE.

DOI： 10.1109/IECON.2011.6120002

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84863028335&origin=inward

Detection of pedestrians employing a wide-angel camera 査読有り

Matsuda R., Tan J., Kim H., Ishikawa S.

International Conference on Control, Automation and Systems 1748 - 1751 2011年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Recently, the number of accidents that pedestrians have law violation is in the tendency of decrease in Japan. However, accidents caused by pedestrians crossing a crosswalk or dashing into a crosswalk still have high ratio, and both accident sources account for 15% of the whole number of accidents caused by a pedestrian. Although many researches in ITS in which pedestrians are detected from in-vehicle cameras have been actively done to solve these problems, they usually employ standard cameras, and those pedestrians who exist outside of the camera view cannot be detected. In this paper, we employ a wide-angle camera which has wider view than a general camera and propose a technique for detecting pedestrians from the wide-angle image. Since, in a wide-angle camera image, every object becomes smaller, we propose a technique for detecting pedestrians employing optical flows converging to a FOE (Focus of Expansion). Experimental results show satisfactory performance of the technique. © 2011 ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84863031172&origin=inward

Detection of a bicycle and its driving directions using HOG feature 査読有り

Jung H., Tan J., Kim H., Ishikawa S.

Proceedings of the 16th International Symposium on Artificial Life and Robotics, AROB 16th'11 781 - 784 2011年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Studies on car vision have currently been practiced around recognizing a human enthusiastically. The Histograms of Oriented Gradients (HOG) feature has been proposed as useful feature for recognizing a human standing in various kinds of background. On the other hand, although a bicycle is important transportation vehicle in urban environment, its automatic recognition or detection is not an easy task for a computer vision system, because bicycle's appearance can change dramatically according to viewpoints and a person riding on the bicycle is a non-rigid object. Thus, automatic bicycle detection is an important research subject in an intelligent perception system using car vision. In this paper, we propose a method of detecting a bicycle and its driving direction using the HOG feature and RealAdaboost algorithm. Experimental results show satisfactory performance of the proposed method. © 2011 ISAROB.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84866651335&origin=inward

Detecting human flows on a road different from main flows 査読有り

Park M., Tan J., Nakashima Y., Kim H., Ishikawa S.

Proceedings of the 16th International Symposium on Artificial Life and Robotics, AROB 16th'11 793 - 796 2011年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Automatic detection of human flows on a road by a computer vision system is of great importance mainly in surveillance systems, where human flows are observed by a camera and a computer analyzes the videos that the camera provides to detect a person having a different flow of movement, such as a person walking toward a certain direction while most of the people walk in the opposite direction, or a person running in a group of walking people. This paper describes a technique for finding a person having a different behavior or motion from others. The idea of the paper is to classify motion flows (or optical flows) extracted from a video into respective groups having respective directions of the motion by analyzing the motion flows. Experimental results show effectiveness of the proposed technique. © 2011 ISAROB.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84866643383&origin=inward

A 3D Matching Method for Organic Training Samples Alignment Based on Surface Curvature Distribution 査読有り

Li, Kim, Tan, Ishikawa, Yamamoto

Open Journal of Medical Imaging 1 ( 2 ) 43 - 47 2011年12月

Multiple objects tracking method based on particle filter 査読有り

Sugandi, Kim, Tan, Ishikawa

Recent Research in Circuits, Systems, Mechanics and Transportation Systems 64 - 69 2011年12月

3-D modeling of dynamic remote environments employing the images from cell-phone cameras and a communication network 査読有り

Ozaki M., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 48 - 51 2011年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper describes a system for 3-D modeling of dynamic remote environments employing the images provided from cell-phone cameras and a communication network. In the system, we employ two cell-phones to acquire images. Each image is sent to a lab by an e-mail and employed for the 3-D modeling. For the algorithm of 3-D recovery, we employ a mobile stereo vision method, since this method is based on the factorization method and a steady solution can be expected. In the experiment, we tested the influence of the restoration error by using two cell-phones and obtained an excellent result. © 2011 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=81255179892&origin=inward

Multiple persons' action recognition by fast human detection 査読有り

Eftakhar S., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 1639 - 1644 2011年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Recognition of the actions of several persons in a scene is gaining importance for human action analysis. Such a kind of system capable of recognizing human actions in a scene is proposed in this paper. We have adopted a fast human detector using Histograms of Oriented Gradients (HOG) and Support Vector Machine (SVM) to robustly trace individual persons. Person-wise bounding blobs are extracted and individual features are tracked in subsequent frames. The recognition is performed individually by comparing with the known motion templates using a high-speed action database. The technique guarantees improved performance in terms of detection and recognition. © 2011 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=81255147356&origin=inward

Improved approach for action recognition based on local and global features 査読有り

Ahad M., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 1645 - 1649 2011年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper presents an improved spatio-temporal (XYT) approach for local interest point-based global action representation, considering the history of moving points in an action. The presented spatio-temporal representation demonstrate robust results and we compare the developed method with previous other method. This is a SURF-based method where we extract visual features to select candidate points based on the SURF detector. Afterwards, motion features are extracted by exploiting the local interest points and by employing optical flow. RANSAC is employed to reduce the unwanted outliers and improve the performance of the method. Based on an outdoor action dataset, we have found that the developed method demonstrate satisfactory recognition results. © 2011 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=81255210497&origin=inward

Buildings recovery employing Manhattan-world constraint 査読有り

Ohyama Y., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 431 - 434 2011年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

The structure from Motion (SfM) algorithm is an established method of shape recovery from a single video sequence. However, limitation of the method is the accuracy of the recovery due to poor texture of a recovered plane or a large scale object. We propose a technique for improving the precision of the recovery by applying the Manhattan-world constraint to the SfM algorithm, which assumes that the buildings are composed of vertical and horizontal planes. We show its effectiveness by the experiments performed in a real-life environment. © 2011 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=81255178501&origin=inward

Action dataset - A survey 査読有り

Ahad M., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 1650 - 1655 2011年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Human action understanding and recognition have various demands for different applications in the field of computer vision and human-machine interaction. Due to these issues, more than a decade, extensive researches are going on in this arena - to recognize various actions and activities. Researchers have been exploiting various action datasets and some of them become prominent. Though there are some good datasets, unfortunately, to have a strong survey on these datasets has been a long due. This paper attempts this and presents the key datasets and analyzes them in different perspectives. © 2011 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=81255147357&origin=inward

3-D recovery of a non-rigid object from a single camera view 査読有り

Ishikawa S., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 447 - 450 2011年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper proposes a technique for 3-D recovery of a non-rigid object, such as a moving person, from a single camera view. Recovery of a non-rigid object is not possible from a single camera view without any condition. In this paper, we propose a single camera technique for recovering a non-rigid object under the condition that the object is composed of a set of rigid objects. The experiments employing real motion data show effectiveness of the proposed technique. © 2011 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=81255128065&origin=inward

階層的画像位置合わせによる胸部マルチスライスCT画像の経時的差分処理の高速化査読有り

前田、金、タン、石川、山本

バイオメディカル・ファジィ・システム学会誌 13 ( 2 ) 1 - 7 2011年10月

A non-rigid alignment method for triangular mesh surface of lung field 査読有り

Li, Kim, Tan. Ishikawa, Yamamoto

International Workshop on Smart Info-Media Systems in Asia 150 - 153 2011年10月

A shape-based segmentation method of multi-organs on CT images using genetic algorithm 査読有り

Li G., Takahashi H., Kim H., Tan J., Ishikawa S., Yamamoto A., Yamamoto A.

ICIC Express Letters 5 ( 9 A ) 3189 - 3194 2011年09月

記述言語：英語掲載種別：研究論文（学術雑誌）

Due to utilizing the priori information of shape, the Active Shape Model (ASM) approach is considered as a robust and effective tool for region of interest (ROI) searching in medical image segmentation. However, in the practice, since the pose parameters and the shape parameters have strong coupling, the analytic method is difficult to perform. In this paper, we propose a parameter optimization method by genetic algorithm (GA) and apply it to the segmentation of lung field and cardiac field on CT images. Region texture derived from training set is proposed as the fitness function in genetic algorithm. Segmentation experiments were performed on lung and cardiac field segmentation on CT images. The lung field segmentation results perform well with average error of 91.0%. At last, we argue the limitation of linear Principal Component Analysis (PCA) to represent shape variations through cardiac field segmentation results. © ICIC International 2011.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=80052456864&origin=inward

A Shape-Based Segmentation Method of Multi-organs on CT Images Using Genetic Algorithm 査読有り

Li, Takahashi, Kim, Tan, Ishikawa and Yamamoto

ICIC Express Letters 5 ( 9(A) ) 3189 - 3194 2011年09月

Automatic Detection of Phalangeal Region on CR Images Using Active Contour Model 査読有り

Yamakawa, Murakami, Kim, Tan, Ishikawa, Aoki

International Symposim on Advanced Intelligent Systems 362 - 365 2011年09月

Detection of Calcification on Carotid Artery in Dental CT Image 査読有り

Shimizu, Kim, Tan, Ishikawa, Tanaka, Kitou, Morimoto

International Symposim on Advanced Intelligent Systems 358 - 361 2011年09月

Segmentation Method for Mandibular Region in Dental CT Images Using MPR Image Information 査読有り

Kuroki, Kim, Tan, Ishikawa, Tanaka, Kitou, Morimoto

International Symposim on Advanced Intelligent Systems 63 - 66 2011年09月

Detection of Artery Regions on Fresh Blood Imaging Using Particle Filter Algorithms 査読有り

Koga, Yamamoto, Kim, Tan, Ishikawa

International Symposim on Advanced Intelligent Systems 59 - 62 2011年09月

A Method for Image Registration on Head CT and MR Image by Using Real-coded Genetic Algorithm 査読有り

Hayata, Yamamura, Kim, Tan, Ishikawa, Yamamoto

International Symposim on Advanced Intelligent Systems 55 - 58 2011年09月

Automatic Detection Method for Candidate Regions of Lung Nodule from the Temporal Subtraction Images 査読有り

Tokisa, Miyake, Maeda, Kim, Tan, Ishikawa, Murakami, Aoki

International Symposim on Advanced Intelligent Systems 51 - 54 2011年09月

Automatic Segmentation Method of Internal Organ Regions Using Graph Cuts 査読有り

Tani, Li, Kim, Tan, Ishikawa, Yamamoto

International Symposim on Advanced Intelligent Systems 47 - 50 2011年09月

Mulple persons’ recognition by fast human detection 査読有り

Efackhar, Tan, Kim, Ishikawa

SICE Annual Conference 2011 1639 - 1644 2011年09月

Detection of lung nodules in thoracic MDCT images based on temporal changes from previous and current images 査読有り

Maeda S., Tomiyama Y., Kim H., Miyake N., Itai Y., Kooi Tan J., Ishikawa S., Yamamoto A.

Journal of Advanced Computational Intelligence and Intelligent Informatics 15 ( 6 ) 707 - 713 2011年08月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Temporal subtraction enhances temporal change by subtracting images captured at different times. Medical images captured currently (current images) and in previous examination (previous images) are subtracted to enhance new lesions and temporal change in existing lesion shadows. Temporal subtraction using chest MultiDetector-Row Computed Tomography (MDCT) images and currently being developed is to be applied to nodule detection in pulmonary regions. Nodule detection using conventional temporal subtraction, however, yields many false-positive results for those 20 mm or less in diameter, requiring improvement. We discuss improvements in nodule detection accuracy using temporal subtraction, first extracting rough nodules from temporal subtraction images as candidate shadows. Features are then acquired from current, previous, and temporal subtraction images. We use intensity features in previous images and shape features in the current images and in features used in conventional methods. Using acquired features, we build a neural network classifier, then extract final pulmonary candidates in unknown shadows.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=80051941017&origin=inward

胸部CT画像のための経時的差分像生成法と臨床用アプリケーションの開発査読有り

金、三宅、前田、タン、石川、村上、青木

CT検診 18 ( 2 ) 107 - 113 2011年08月

オトガイ孔の位置情報とヘッセ行列を用いた歯科CT画像からの下顎管領域の抽出査読有り

黒木，三戸，金，タン，石川，田中，鬼頭，森本

バイオメディカル・ファジィ・システム学会誌 13 ( 1 ) 109 - 112 2011年06月

過去・現在の胸部MDCT像セットを用いた経時的差分像技術の開発査読有り

三宅，金，前田,タン，石川，村上，青木，山本

バイオメディカル・ファジィ・システム学会誌 13 ( 1 ) 73 - 80 2011年06月

Reduction of processing times for temporal subtraction on lung CT image employing octree algorithms 査読有り

Maeda S., Kim H., Itai Y., Tan J., Ishikawa S., Yamamoto A.

International Journal of Innovative Computing, Information and Control 7 ( 5 B ) 2603 - 2610 2011年05月

記述言語：英語掲載種別：研究論文（学術雑誌）

The temporal subtraction image, which can be obtained by subtracting previous image from current one, is useful for visual screening in clinical field. The temporal subtraction technique removes normal structures, e.g., blood vessel. Hence, it can enhance interval changes such as the new lesions and the changes of existing abnormalities on medical images. Recently, several temporal subtraction methods have been proposed for thoracic medical images. In temporal subtraction, image registration technique is required for correcting displacement between current image and previous one. However, efficient image registration technique of temporal subtraction for MDCT (Multi Detector-row CT) has not been proposed because of the complication of deformation in 3 dimensional region. In this paper, we propose a new efficient computer aided diagnosis (CAD) algorithms for detection of lung nodules which are obtained by temporal subtraction for thoracic MDCT images. We have tried to reduce the computational time for the temporal subtraction image by use of octree algorithms on 3-dimensional image space. To evaluate our method, we have applied the method to 4 MDCT dataset and confirmed its efficiency. © 2011.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=79956078276&origin=inward

Reduction of processing times for temporal subtraction on lung CT image employing octree algorithms 査読有り

Maeda, Kim, Itai, Tan, Ishikawa, Yamamoto

Intl. Journal of Innovative Computing, Information and Control 7 ( 5(B) ) 2603 - 2610 2011年05月

テクスチャ解析を用いた頚椎側面画像におけるモーションアーチファクト検出のための定量的手法の開発

村上、金、タン、石川

北九州医工学術者協会論文誌 21 3 - 6 2011年04月

3次元胸部CT像からの経時的差分像の生成法と解析用GUIの開発

三宅, 金，前田，タン，石川，村上，青木，桂川，白石

北九州医工学術者協会論文誌 21 22 - 25 2011年04月

SURF-based Spatio-Temporal History Image Method for Action Representation 査読有り

Md. Atiqur Rahman Ahad, J.K. Tan, H. Kim, S. Ishikawa

IEEE International Conference on Industrial Technology 397 - 402 2011年03月

Registration of head ct and mr image sets based on optimal retrieval of volume of interests 査読有り

Yamamura Y., Harada K., Kim H., Kooi Tan J., Ishikawa S., Yamamoto A., Yamamoto A.

ICIC Express Letters, Part B: Applications 2 ( 1 ) 101 - 106 2011年02月

記述言語：英語掲載種別：研究論文（学術雑誌）

Image registration is one of the important techniques for diagnosis or treat- ment using medical images. Especially, it is very efficient to detect abnormalities by comparing two or more images. The images, which are provided from multi modalities or taken several different times, are used for diagnosis. One of the multi modal images is CT and MR imaging of the head for diagnosis and surgical planning indicates that physicians and surgeons gain important information from these modalities. In general, to simulate before the surgical operation, manual registration techniques are mainly per- formed on MR and CT images of the head. However, manual registration of the object may require several hours for analysis based on anatomical knowledge. In this paper, we describe a new method for image registration technique of head images obtained from CT and MRI. Our new method uses an optimal retrieval on neighbor volumes of interest (VOIs) and maximization of mutual information. This method also can reduce the pro-cessing time without decreasing the registration accuracy. © 2011 ISSN 2185-2766.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=79251503899&origin=inward

Registration of head CT and MR image sets based on optimal retrieval of volume of interests 査読有り

Yamamura,Harada,Kim,Tan,Ishikawa,Yamamoto

ICICIC Express Letters Part B: Applications 2 ( 1 ) 101 - 106 2011年02月

A Method for Reduction of Subtraction Artifacts on Temporal Subtraction Images by Use of Generalized Gradient Vector Flow Technique 査読有り

Kim,Miyake,Tan,Ishikawa,Murakami,Aoki

Joint Meeting combining The 3rd meeting of Japanese Society of Pulmonary Functional Imaging and 5th International Workshop for Pulmonary Functional Imaging 162 - 162 2011年01月

Automatic Segmentation of Liver Region from Non-contrast and Contrast CT images Employing Tree-structural Image Transformations 査読有り

Komatsu,Li,Kim,Tan,Ishikawa,Yamamoto

The 16th International Symposium on Artificial Life and Robotics 763 - 766 2011年01月

ベクトル集中度とボクセルマッチング法を用いた胸部CT画像の位置合わせ査読有り

三宅、金、前田、タン、石川、村上、青木

進化計算シンポジウム２０１０ 240 - 247 2010年12月

Recognizing human motions from surrounding viewpoints employing hierarchical eigenspaces 査読有り

Eftakhar,Tan,Kim,Ishikawa

Intl. Conf. Control, Automation, Robotics and Vision 2179 - 2184 2010年12月

GGVFからのベクトル集中度合を考慮した胸部CT像の経時的差分法

三宅、金、前田、タン、石川、村上、青木、山本

第29回SICE九州支部学術講演会 191 - 194 2010年12月

ヘッセ行列を用いた歯科CT画像からの下顎管領域の抽出

黒木、金、タン、石川、田中、鬼頭、森本

第29回SICE九州支部学術講演会 197 - 200 2010年12月

Active Shape Modelを用いた胸部CT画像からの心臓領域の自動抽出

高橋、小松、李、金、タン、石川、山本

第29回SICE九州支部学術講演会 195 - 196 2010年12月

非剛体レジストレーション：画像レジストレーション法の胸部ＣＴへの応用

金　亨燮

日本医用画像工学会e-ニュースレター ( 日本医用画像工学会 ) 7 32 - 33 2010年12月

A color-based particle filter for multiple objects tracking in outdoor environment 査読有り

Sugandi B., Kim H., Tan J., Ishikawa S.

Proceedings of the 15th International Symposium on Artificial Life and Robotics, AROB 15th'10 604 - 609 2010年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Tracking of multiple objects is more challenging than tracking a single object. Some problems arise in multiple object tracking that do not exist in single object tracking such as objects occlusion, appearing new object and disappearing already existed object, updating the occluded object, etc. In this paper, we present an approach to handle multiple objects tracking in the presence of occlusions, background clutter and appearance change. The occlusion is handled considering the predicted trajectories of the objects based on a dynamic model and likelihood measures. We propose also a target-model update conditions, ensuring the proper tracking multiple objects. The proposed method is implemented in the probabilistic framework such as particle filter in conjunction with a color feature. The particle filter has proven very successful for non-linear and non-Gaussian estimation problems. It approximates a posterior probability density of the state such as the object position by using samples or particles, where each state is denoted as the hypothetical state of the tracked object and its weight. The observation likelihood of the objects is modeled based on color histogram. The sample weight is measured based on Bhattacharya coefficient that measures the similarity between each sample's histogram with a specified target model. The algorithm can track the multiple objects in the present of occlusion and noise successfully. Some experimental results show the effectiveness of our method to track the multiple objects. © 2010 ISAROB.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84866680510&origin=inward

Segmentation of artery areas on none-enhanced fresh blood imaging based on dot enhancement filter and 3-D region growing method 査読有り

Yamamoto A., Nishizaki T., Kim H., Tan J., Ishikawa S.

Proceedings of the 15th International Symposium on Artificial Life and Robotics, AROB 15th'10 493 - 496 2010年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Peripheral arterial disease (PAD) is characterized by lower extremities arterial obstruction due to athe rosclerosis and manifests in lower extremities as intermittent claudicating, limb ischemia, or gangrene. The dia gnosis of the PAD using magnetic resonance imaging (MRI) equipment without contrast medium is available as a useful visual screening in clinical practice. In this paper, we propose a new method for segmentation of arterial images, which are obtained from non-contrast enhanced magnetic resonance angiography (MRA) based on dot enhancement filter and 3-D region growing methods, and satisfactory experimental results are obtained. © 2010 ISAROB.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84866686962&origin=inward

Segmentation method for phalanges in CR image by use of DCT 査読有り

Hozu Y., Murakami S., Murakami S., Kim H., Tan J., Ishikawa S., Aoki T.

ICCAS 2010 - International Conference on Control, Automation and Systems 2082 - 2085 2010年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

In this paper, we propose a CAD (Computer aided diagnosis) system to analyze the RA (rheumatoid arthritis) and osteoporosis by using image processing techniques from the CR images. To analyze the RA, we develop a segmentation method for phalanges in CR Image by use of DCT (Discrete Cosine Transform) for detection of temporal change. The temporal change is detected using the difference image between previous image and current one. The DCT is performed to emphasize the edge of the difference image. Finally, the phalanges are extracted by performing Snakes. The primary objective of this study is to segment phalanges by making temporal subtraction images. We apply our proposed technique to eight cases of CR images and satisfactory segmentation results are achieved. A new index that diagnoses the progress level of the disease of phalanges can be offered as a second opinion. ©ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78751512171&origin=inward

Segmentation method for cardiac region in CT images based on active shape model 査読有り

Takahashi H., Komatsu M., Kim H., Tan J., Ishikawa S., Yamamoto A., Yamamoto A.

ICCAS 2010 - International Conference on Control, Automation and Systems 2074 - 2077 2010年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Recently, multi detector row computed tomography (MDCT) has been introduced into medical fields. By the development of MDCT, images with high quality are provided into medical fields. So many related image processing techniques are proposed into medical image processing fields for extraction of abnormal area. In the medical image processing field, segmentation is one of the most important problems for analyzing the abnormalities and recognition of internal structures before the operation. For this reason, many approaches are proposed for detection of abnormal area on CT images. Before detection of abnormal areas, segmentation of organs in CT images is one of the most important problems for analyzing of disease. However, poor contrast, image noises and motion artifacts make this segmentation problem difficult in particular in cardiac region. Moreover, there are still no fully automatic segmentation methods for cardiac region on CT images. In this paper, we present automatic extraction technique for detection of cardiac region. Our proposed technique combines active shape model (ASM) and genetic algorithm (GA). We apply our proposed technique to five real CT images and satisfactory segmentation results are achieved. ©ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78751513157&origin=inward

Modeling of human postures using stereo camera 査読有り

Yoshida S., Tan J., Kim H., Ishikawa S.

ICCAS 2010 - International Conference on Control, Automation and Systems 1432 - 1435 2010年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper proposes an automatic human posture acquisition technique by a robot. Decrease of working population will be a vital problem in our society. Escaping from the work in danger is also important for a human being to raise QOL. To solve all theses issues, it may be quite useful to realize a robot which works in place of a human. Based on this background, our research aims at making a technique for a robot to acquire various human postures/motions automatically. In this paper, a technique is presented for obtaining human postures automatically by employing a human skeleton model and a stereo camera system. Performance of the technique is shown experimentally. ©ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78751531236&origin=inward

Method for reduction of metal artifacts in dental CT images by use of frequency analysis 査読有り

Mito K., Kim H., Tan J., Ishikawa S., Tanaka T., Morimoto Y.

ICCAS 2010 - International Conference on Control, Automation and Systems 1271 - 1274 2010年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

CAD (Computer Aided Diagnosis) system that has 3-D viewer function to display the DICOM (Digital Imaging and Communication in Medicine) image has been developed in recent years. Now, the function of CAD system for artificial implant to which development is advanced includes the removal of artifact in the input DICOM image, region extraction of bone, major vessel, and nerve and mounting of function to presume place where implant is stricken. However, it is still required a software with high performance because of taking the processing time to do a series of operation before simulating by the strike of the implant on the operation on a clinical site. In this paper, we propose the function to extract the area of bone separately for teeth and mandibles automatically as addition to the system that can do a series of operation before simulating by the strike of the implant. In addition, we proposed also the function to extract each tooth separately. The main aim of this paper is to develop the integrative system for the dental screening that can be operated in a personal computer. ©ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78751559566&origin=inward

Human motion recognition using directional motion history images 査読有り

Murakami M., Tan J., Kim H., Ishikawa S.

ICCAS 2010 - International Conference on Control, Automation and Systems 1445 - 1449 2010年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Nowadays many persons are needed to observe images from surveillance cameras, because many surveillance cameras are installed in a town or in buildings. They are working under strain, because they must always watch the images from surveillance cameras to find a person with abnormal motion. Therefore reduction of the load is necessary. Many existing researches on human motion recognition are only recognition and the information of the result is not used. We assume that human motion is a set of basic motions. So, an overall motion can be understood using the recognition result of basic motions. The goal of the present research is to develop a method of human motion representation and translation using directional motion history images (DMHIs). In this paper, we describe a method of recognizing basic motions using the DMHIs. We perform the recognition by the Histograms of Oriented Gradients (HOG) feature. In the experiment, the number of bins and local area (cell) sizes for calculating the HOG feature are changed and the most suitable values are inspected. ©ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78751548634&origin=inward

Human behavior analysis with optical flow and median-filtered Temporal Motion Segmentation method 査読有り

Ahad M., Tan J., Kim H., Ishikawa S.

Proceedings of the 15th International Symposium on Artificial Life and Robotics, AROB 15th'10 1022 - 1025 2010年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

We focus on human activity analysis so that an intelligent system (e.g., a robot) can easily understand some important activities and help thereby. Hence we present an improved method for activity analysis, called Medianfiltered Temporal Motion Segmentation (MfTMS) method, which can segment and understand motion temporally from the video sequence. It is based on the computation of optical flow and thereafter split it into four different channels. Later median filtering is applied and we produce four motion-history templates based on the directional motion vectors. Based on the total pixel volumes on these history templates and their related variations, various directions of the action primitives are segmented temporally. We conduct experimentations both indoor and outdoor environments and achieved sound performance. This segmentation method can assist an intelligent system or a robot to understand activities and take decisions afterwards. It is a simple but robust approach. © 2010 ISAROB.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84866648695&origin=inward

Detection of blood vessel regions using weighted MIP from non-enhanced MR imaging 査読有り

Koga Y., Yamamoto A., Yamamoto A., Kim H., Tan J., Ishikawa S.

ICCAS 2010 - International Conference on Control, Automation and Systems 2078 - 2081 2010年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Recently, arterio sclerosis obliterans (ASO) caused by the chronic ischemia of limbs increase remarkably. As one of the diagnosis methods of this disease, the image diagnosis such as MR image is used on medical fields. In this paper, we propose a method to develop medical diagnostic support system using fresh blood imaging (FBI) method, in which the structure of blood vessel regions in legs could be analyzed from non-enhanced MR images. As the method, an initial point of the vessel regions in the artery areas is acquired from weighted MIP images automatically. Then extend the artery areas using region growing method. Especially, we developed for the position relationship of the crossed blood vessel and the segment accuracy in peripheral artery areas that is improved by using weighted MIP images that the result could not be gotten in conventional MIP images. To evaluate our proposed method, we performed our proposed method and satisfactory results are achieved. ©ICROS.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78751514149&origin=inward

Solving boundary problem of the motion database for improved human motion recognition 査読有り

Eftakhar S., Tan J., Kim H., Ishikawa S.

ICECE 2010 - 6th International Conference on Electrical and Computer Engineering 698 - 701 2010年12月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Development of a high accuracy and robust human motion recognition system is the fundamental requirement to apply in the real-life applications. Therefore, the recognition systems are being investigated to improve the system's performance and to enhance with more functionality. As a result of such an investigation, the formerly adopted structured motion database, capable of suitable retrieval of known motion patterns, is modified for obtaining more improved performance over the earlier systems [11]. The earlier systems has the boundary problems for the motion points those lie on the edge of the searching boundary within the query space at the time of retrieving similar motions from within the motion database. This problem might lead to the misrecognition of human motions in high degree. To overcome this problem, we make use of two sets of query spaces, namely original and shifted query space set which is to be switched based on the input motion. Motion History Image (MHI) and Exclusive-OR (OR) image are used as the motion representations, and the directional eigenspace strategy is used for constructing vector spaces. The results guarantee the significant improvement of the proposed system. ©2010 IEEE.

DOI： 10.1109/ICELCE.2010.5700789

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=79951802289&origin=inward

Creating an entire object model employing virtual see-through cameras 査読有り

Sone T., Piao M., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 2504 - 2506 2010年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper proposes a novel 3-D modeling technique of an object. The existent 3-D modeling techniques such as stereo vision or factorization recovers 3-D shapes of part of an object observed commonly from multiple orientations. This is followed by registration among recovered partial shapes in order to obtain an entire 3-D model of the object, resulting in some geometrical errors in the linked shape. The proposed technique creates entire shape of an object simultaneously without partial shape registration by the employment of virtual see-through cameras that surround the object and observes its rear part as well as the frontal part. All the surrounding cameras are calibrated initially using the captured images. The frontal side of an object concerned recovers its 3-D shape by the factorization method applied to the image data obtained from the frontal cameras. The recovered frontal shape is virtually projected onto the image planes of the rear cameras. This procedure realizes that all the surrounding cameras observe all the shape of the object, irrespective of frontal or rear. Then the factorization method is applied again to this situation to yield an entire 3-D model of the object. A nonlinear optimization method is employed in the technique to achieve the modeling of higher precision. The proposed technique is applied to the 3-D modeling of a mini basket ball match and their activity is successfully modeled in a 3-D way. © 2010 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78649296463&origin=inward

Three-dimensional modeling of a non-rigid object employing a single camera 査読有り

Horinouchi T., Jung H., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 2972 - 2975 2010年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

This paper proposes a technique for recovering 3-D shape of a non-rigid object employing a single camera. Normally two or more cameras are necessary for recovering 3-D shape of a non-rigid object. Shape recovery of a non-rigid object employing a single camera has recently been studied by a few researchers, but they deal only with small deformation. The technique proposed in this paper achieves the shape recovery of larger deformation by regarding a non-rigid object as a set of rigid parts. Each rigid part recovers its 3-D shape by the application of the factorization method and they are connected mutually to represent an overall deformation of the object. The present technique also considers the shape recovery of an occluded part caused by ego motion. Effectiveness of the proposed technique is shown by experiment. © 2010 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78649236530&origin=inward

Viewpoint-oriented human activity recognition in a cluttered outdoor environment 査読有り

Eftakhar S., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 1506 - 1511 2010年11月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Motion analysis in a cluttered outdoor scenario is a real challenging task. In this paper, a suitable human activity recognition approach is proposed that adapts with the noisy environment and with variable viewpoints. The recognition approach encompasses some motion processing methodologies having high precision for specific analysis, among those are: motion segmentation, optical flow computation, Motion History Image (MHI) generation, structured database development, and so on. Estimating the subject's moving body region, 4-directional MHIs of each activity are generated. An efficient pre-developed motion database aids in recognition of the activities. The proposed approach proves its significant precision in recognition for various activities from varying viewpoints. © 2010 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78649234809&origin=inward

Lower-dimensional feature sets for template-based motion recognition approaches 査読有り

Ahad M., Tan J., Kim H., Ishikawa S.

Journal of Computer Science 6 ( 8 ) 920 - 927 2010年11月

記述言語：英語掲載種別：研究論文（学術雑誌）

Problem statement: In template-based motion recognition approaches, feature sets are computed from the template for classification. Hu invariants are widely employed for this purpose since its inception. However, development of lower-dimensional feature vector sets is required for faster computation along with robust recognition. The concept of reduced size of Hu moment is really interesting. From its inception, seven higher orders Hu moments have been employed by many researchers without considering why seven and why not less numbers. Approach: In this study, we analyzed with various feature sets with different number of Hu moments and rationalized that based on the characteristics of central moments, it is not necessary to employ all the seven moments in every applications and, in that way, we can reduce the computational cost and make it faster. Results: Based on various feature vectors sets, it is evident that we can use lower dimensional feature vectors for our Directional Motion History Image (DMHI) method and other methods. Conclusion: Therefore, we can conclude that we do not need all seven invariants, rather 1st two or three invariants seem enough-as we are not reproducing the image. Higher invariants are noisy and hence can be ignored. The 0th order moment for Energy images provide enough information about the mass area and hence, no need to calculate the other seven invariants. © 2010 Science Publications.

DOI： 10.3844/jcssp.2010.920.927

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78049482012&origin=inward

Method for Reduction of Metal Artifacts in Dental CT Images by Use of Frequency Analysis 査読有り

Mito,Kim,Tan,Ishikawa,Tanaka,Morimoto

International Conference on Control, Automation and Systems 1271 - 1274 2010年10月

Medeling of Human Postures Using Stereo Camera 査読有り

Yoshida,Tan,Kim,Ishikawa

International Conference on Control, Automation and Systems 1432 - 1435 2010年10月

Segmentation Method for Phalanges in CR Image by Use of DCT 査読有り

Hozu,Murakami,Kim,Tan,Ishikawa,Aoki

International Conference on Control, Automation and Systems 2082 - 2085 2010年10月

Detection of Blood Vessel Regions Using Weighted MIP from Non-enhanced MR Imaging 査読有り

Koga,Yamamoto,Kim,Tan,Ishikawa

International Conference on Control, Automation and System 2078 - 2081 2010年10月

Segmentation Method for Cardiac Region in CT Images Based on Active Shape Model 査読有り

Takahashi,Komatsu,Kim,Tan,Ishikawa,Yamamoto

International Conference on Control, Automation and Systems 2074 - 2077 2010年10月

Human Motion Recognition Using Directional Motion History Image 査読有り

Murakami,Tan,Kim,Ishikawa

International Conference on Control, Automation and Systems 1445 - 1449 2010年10月

濃度・形状特徴量を用いたスリガラス状陰影の自動識別法

塩澤，金，タン，石川

バイオメディカル・ファジィ・システム学会第23回年次大会講演論文集 59 - 62 2010年10月

異なる時期に得られる胸部MDCT画像の経時的差分像技術の開発

三宅、金、前田、タン、石川、村上、青木、山本

バイオメディカル・ファジィ・システム学会第23回年次大会講演論文集 55 - 58 2010年10月

Snakes法によるCR画像からの指骨領域の自動抽出

山川，村上，金，タン，石川，青木

バイオメディカル・ファジィ・システム学会第23回年次大会講演論文集 125 - 128 2010年10月

VTKを用いたDICOM表示用GUIの開発

竹内，金，タン，石川，山本

バイオメディカル・ファジィ・システム学会第23回年次大会講演論文集 75 - 76 2010年10月

Occlusion handling in object tracking based on particle filter approach 査読有り

Sugandi,Kim,Tan,Ishikawa

7th Intl. Conference & Expo on Emerging Technologies for a Smarter World 2010年09月

Action recognition by employing combined directional motion history and energy images 査読有り

Rahman Ahad M., Tan J., Kim H., Ishikawa S.

2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops, CVPRW 2010 73 - 78 2010年09月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Human action understanding and analysis for various applications are still in infancy due to various factors. In this paper, for recognizing various complex activities, a combined cue for motion representation and later recognition is demonstrated based on the optical flowbased four directional motion history and basic energy images. Optical flow between consecutive frames are computed to create the update function and to segment the moving regions. These motion vectors are split into four different channels. From these channels, the corresponding four directional history templates are computed. These along with frame-subtracted energy motion templates represent the final motion information of an action sequence. From these templates, feature vectors are calculated according to the seven Hu invariants. We develop a 35-dimensional feature vector for each action. For classification, k-nearest neighbor classification scheme is employed. For partitioning scheme, we employ leave-one-out cross-validation method. Both indoor and outdoor dataset provide satisfactory recognition results. These analysis, representation can be used for robot vision, interactive systems, computer games, behavior understanding, etc. © 2010 IEEE.

DOI： 10.1109/CVPRW.2010.5543160

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=77956548614&origin=inward

Segmentation of artery areas on non-enhanced fresh blood imaging based on a dot enhancement filter and the 3-D region-growing method 査読有り

Yamamoto A., Nishizaki T., Kim H., Tan J., Ishikawa S.

Artificial Life and Robotics 15 ( 1 ) 48 - 52 2010年09月

記述言語：英語掲載種別：研究論文（学術雑誌）

Peripheral arterial disease (PAD) is characterized by arterial obstruction in the lower extremities due to atherosclerosis, and manifests itself as intermittent claudicating, limb ischemia, or gangrene. Diagnosis of PAD using magnetic resonance imaging (MRI) equipment without contrast medium is available as a useful visual screening in clinical practice. In this article, we propose a new method for the segmentation of arterial images which are obtained from non-contrast-enhanced magnetic resonance angiography (MRA) based on a dot enhancement filter and 3-D region-growing methods, and satisfactory experimental results are obtained. © 2010 International Symposium on Artificial Life and Robotics (ISAROB).

DOI： 10.1007/s10015-010-0764-0

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=77956106053&origin=inward

胸部MDCT画像の経時差分法の開発と病変部候補領域の抽出

金　亨燮

立石科学技術振興財団助成研究成果集 ( 19 ) 29 - 32 2010年09月

Detection of Candidate Regions for Lung Nodules on Temporal Subtraction Image Based on Artificial Neural Networks 査読有り

Maeda,Kim,Tomiyama,Tan,Ishikawa,Yamamoto

Intl. Conf. Biological and Biomedical Engineering 1037 - 1040 2010年08月

Tracking of Multiple Moving Objects Under Outdoor Environment Using Color-based Particle Filter 査読有り

Sugandi,Kim,Tan,Ishikawa

Proceedings of 2010 3rd IEEE International Conference on Computer Science and Information Technology 103 - 107 2010年07月

形状および濃度特徴量を用いた胸部CT画像上のスリガラス状候補陰影の抽出

金、勝間田、前田、板井、タン、石川

画像電子学会2010年度年次大会講演論文集CD-ROM 2010年06月

濃度情報に基づく実数値GAを用いた頭部CTA・MRA画像の位置合わせ

金、西野、原田、山村、タン、石川、山本

画像電子学会2010年度年次大会講演論文集CD-ROM - 32 2010年06月

3次元リージョングローイングとテンプレート・マッチングを用いたCT画像からの肺臓・心臓領域の自動抽出査読有り

小松，高橋，金，タン，石川，山本

バイオメディカル・ファジィ・システム学会誌 12 ( 1 ) 93 - 98 2010年05月

Automatic Detection of GGO Candidate Regions in Lung CT Image Employing Neural Networks 査読有り

Kim,Katsumata,Maeda,Itai,Tan,Ishikawa

The 3rd Annual Meeting of the Asian Association for Algorithms and Computation 5 - 5 2010年04月

Segmentation of artery areas on none-enhanced fresh blood imaging based on dot enhancement filter and 3-D region growing method 査読有り

Yamamoto,Nishizaki,Kim,Tan,Ishikawa

Artificial Life and Robotics 15 ( 1 ) 48 - 52 2010年04月

仮想透視カメラ群による物体の全周囲復元法と人動作の立体モデル化への応用査読有り

タン，曽根，石川，金，四宮

バイオメディカル・ファジィ・システム学会誌 12 ( 1 ) 71 - 77 2010年04月

A color-based particle filter for multiple objects tracking in outdoor environment 査読有り

Sugandi,Kim,Tan,Ishikawa

Artificial Life and Robotics 15 ( 1 ) 41 - 47 2010年04月

Segmentation of artery areas on none-enhanced fresh blood imaging based on dot enhancement filter and 3-D region growing method 査読有り

Yamamoto,Nishizaki,Kim,Tan,Ishikawa

Artificial Life and Robotic 15 ( 1 ) 48 - 52 2010年04月

Human behavior analysis with optical flow and median-filtered temporal motion segmentation method 査読有り

Ahad,Tan,Kim,Ishikawa

International Symposium on Artificial Life and Robotics 1022 - 1025 2010年02月

Analysis of Motion Self-Occlusion Problem Due to Motion Overwriting for Human Activity Recognition 査読有り

Ahad,Tan,Kim,Ishikawa

Journal of Multimedia 5 ( 1 ) 36 - 46 2010年02月

局所的接続要素のマッチング法による経時的差分像上のアーチファクトの低減

板井、金、石川、桂川

第2回呼吸機能イメージング研究会学術集会抄録集 69 2010年01月

Automatic detection of GGO candidate regions by using density and shape features 査読有り

Kim H., Katsumata Y., Itai Y., Tan J., Ishikawa S.

International Journal of Innovative Computing, Information and Control 6 ( 1 ) 255 - 262 2010年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Various imaging equipments have been introduced into medical fields. Es-pecially, high resolution helical computed tomography (HRCT) is one of the most useful diagnosis systems because it provides a high resolution image to medical doctors as a clear image. Radiologist can easily detect abnormalities by use of the clear images. Detection of abnormal areas such as lung nodule, ground glass opacity on multi detector computed tomography images is a difficult task for radiologists. It is because subtle lesions such as small lung nodules tend to be low in contras, and a large number of computed tomogra-phy images require a long visual screening times. To detect the abnormalities by use of computer aided diagnosis (CAD) system, some technical methods for detecting the ab-normalities have been proposed in medical field. Despite of these efforts, their approach did not succeed because of difficulty of image processing in detecting the ground glass opacity (GGO) areas exactly. Thus they did not reach to the stage of automatic detec-tion employing unknown thoracic MDCT data sets. In this paper, we develop a CAD system for automatic detecting of GGO areas from thoracic MDCT images by use of five statistical features which are obtained four density features and one shape feature. The proposed technique applied on 31 MDCT image sets. 79.4 [%] of recognition rates and 1.07 of false positive rates was achieved. Some experimental results are shown along with a discussion. © 2010 ISSN.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=76549095216&origin=inward

Human motion recognition using directional motion history images 査読有り

Murakami M., Tan J., Kim H., Ishikawa S.

Proceedings of the SICE Annual Conference 1512 - 1514 2010年01月

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）

Nowadays many persons are needed to observe images from surveillance cameras, because many surveillance cameras are installed in a town or in buildings. They are working under stress, because they must always watch the images from surveillance cameras to find a person with abnormal motion. Therefore reduction of the load is necessary. Many existing researches on human motion recognition are only recognition and the information of the result is not used. We assume that human motion is a set of basic motions. So, we think we are able to translate an overall motion using the recognition result of basic motions. The goal of the present research is to develop a method of human motion representation and translation using directional motion history images (DMHIs). In this paper, we describe a method of recognizing basic motions using the DMHIs. We perform the recognition by matching of a graph. When the matching of the graph is performed, the intervals of chosen image frames are changed and the most suitable interval is inspected. © 2010 SICE.

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=78649280229&origin=inward

A simple approach for low-resolution activity recognition 査読有り

Md. Atiqur Rahman Ahad,Tan,Kim,Ishikawa

Intl. Journal for Computational Vision and Biomechanics 3 ( 1 ) 2010年01月

Automatic detection of GGO candidate regions by using density and shape features 査読有り

Kim,Katsumata,Itai,Tan,Ishikawa

International Journal of Innovative Computing, Information and Control 6 ( 1 ) 255 - 262 2010年01月

Development of a voxel matching technique for substantial reduction of subtraction artifacts in temporal subtraction images obtained from thoracic MDCT 査読有り

Itai,Kim,Ishikawa,Katsuragawa,Doi

Journal of Digital Imaging 23 ( 1 ) 31 - 38 2010年01月

Automatic detection and tracking of moving object employing a particle filter 査読有り

Sugandi,Kim,Tan,Ishikawa

International Conference on Innovative Computing, Information and Control 5 pages 2009年12月

Automatic detection of lung nodules in temporal subtraction image by use of shape and density features 査読有り

Miyake,Kim,Itai,Tan,Ishikawa,Katsuragawa

International Conference on Innovative Computing, Information and Control 4 pages 2009年12月

Development of registration method for head CTA and MRA images based on 2-D projection image 査読有り

Horita,Harada,Kim,Tan,Ishikawa,Yamamoto

ICIC Express Letters 3 ( 4 ) 957 - 962 2009年12月

Automatic classification of GGO regions employing linear discriminant function from the thoracic CT images 査読有り

Shiozawa,Kim,Tan,Ishikawa

ICIC Express Letters 3 ( 4 ) 963 - 968 2009年12月

3次元リージョングローイングとテンプレート・マッチングを用いたCT画像からの肺臓・心臓領域の自動抽出

高橋、小松、金、タン、石川、山本

第28回SICE九州支部学術講演会 213 - 214 2009年11月

歯科用CT画像からのメタルアーチファクトの低減法と画像ビュアの開発

木塚、三戸、金、タン、石川、田中、森本

第28回SICE九州支部学術講演会 217 - 218 2009年11月

CR画像からの指骨の経時変化の抽出法

保都、村上、片岡、金、タン、石川

第28回SICE九州支部学術講演会 205 - 206 2009年11月

Detection of GGO candidate regions by using edge enhancement filter and statistical features 査読有り

Kim,Syed,Tan,Ishikaw

International Journal of Innovative Computing, Information and Control 5 ( 11 ) 4267 - 4274 2009年11月

An image registration method for head CT and MR imaging based on optimal retrieval 査読有り

Yamamura,Kim,Tan,Ishikawa,Yamamoto

International Journal of Innovative Computing, Information and Control 5 ( 11 ) 4249 - 4254 2009年11月

関心領域の細分化による胸部CT画像からの高速な経時差分画像の生成法

前田,　金,　タン,　石川,　山本

バイオメディカル・ファジィ・システム学会第22回年次大会講演論文集 179 - 182 2009年10月

3D region growing法を用いた非造影MR画像からの血管領域の自働抽出

山本，西崎，金，タン，石川

バイオメディカル・ファジィ・システム学会第22回年次大会講演論文集 191 - 194 2009年10月

A moving object tracking based on color information employing a particle filter algorithm 査読有り

Sugandi,Kim,Tan,Ishikawa

Artificial Life and Robotics 14 ( 1 ) 39 - 42 2009年10月

A particle filter with adaptive model for object tracking 査読有り

Sugandi,Kim,Tan,Ishikawa

SICE International Conference 2009 6 pages 2009年08月

Human activity analysis: Concentrating on motion history image and its variants 査読有り

Ahad.,Tan,Kim,Ishikawa

SICE International Conference 2009 5401 - 5406 2009年08月

Entire 3-d modeling of an object by surround cameras 査読有り

Sone,Tan,Kim,Ishikawa

SICE International Conference 2009 4011 - 4014 2009年08月

Robust human motion recognition employing adaptive database structure 査読有り

Eftakhar,Tan,Kim,Ishikawa

SICE International Conference 2009 3989 - 3994 2009年08月

Recovering 3-D shape of a non-rigid object by a single mobile camera 査読有り

Horinouchi,Tan,Kim,Ishikawa

SICE International Conference 2009 3846 - 3849 2009年08月

Detecting pedestrians from a video by streo cameras 査読有り

Inumaru,Tan,Kim,Ishikawa

SICE International Conference 2009 663 - 666 2009年08月

Reduction of processing times for temporal subtraction from the thoracic MDCT image employing octree algorithms 査読有り

Maeda,Kim,Itai,Tan,Ishikawa,Yamamoto

6th International Symposium on Management Engineering 2009 6 pages 2009年08月

構造マッチング法による胸部ＣＴ画像からの結節状陰影の検出

板井、金、石川、桂川

第28回日本医用画像工学会大会抄録集 2009年08月

２次元投影画像を用いたCTA・MRA画像からの頭部の位置合わせ法

堀田、原田、金、タン、石川、山本

第28回日本医用画像工学会大会抄録集 2009年08月

ニューラルネットワークによる経時的差分技術を利用した胸部CT画像からの結節状陰影の自動抽出法

三宅、板井、金、タン、石川、桂川

第28回日本医用画像工学会大会抄録集 2009年08月

エッジ情報に基づく頭部ＭＲ・ＣＴ画像の位置合わせ

原田、金、タン、石川、山本

第25回ファジィシステムシンポジウム講演論文集CD-ROM 2009年07月

Automatic detection of GGO in lung images by using statistical features and neural networks 査読有り

Kim,Katsumata,Tan,Ishikawa

The 24th International Technical Conference on Circuits/Systems, Computers and Communitions 705 - 708 2009年07月

Automatic Detection of Spinal Deformity Based on Statistical Features from the Moire Topographic Images 査読有り

Kim,Tan,Ishikawa,Shinomiya

International Journal of Computing 8 ( 1 ) 72 - 78 2009年07月

A color features-based method for object tracking employing a particle filter algorith 査読有り

Sugandi,Kim,Tan,Ishikawa

2nd Global Conference on Power Control & Optimization 206 - 211 2009年06月

仮想透視カメラ群を用いた立体の全周形状一括復元査読有り

守、曽根、タン、金、石川

バイオメディカル・ファジー・システム学会誌 11 ( 1 ) 17 - 22 2009年06月

Temporal Motion Recognition and Segmentation Approach 査読有り

Ahad,Tan,Kim,Ishikawa

Int. J. of Imaging Systems and Technology 19 ( 2 ) 91 - 99 2009年06月

Human Motion Recognition Employing Large Motion-Database Structure 査読有り

Ashik Eftakhar,Tan,Kim,Ishikaw

International Journal of Advanced Computer Engineering 2 ( 1 ) 17 - 23 2009年06月

Real Time Tracking and Identification of Moving Persons by Using a Camera in Outdoor Environments 査読有り

Sugandi,Kim,Tan,Ishikawa

International Journal of Innovative Computing, Information and Control 5 ( 5 ) 1179 - 1188 2009年05月

Ｍedia-Art Employing Virtual Shadows with Shape Recognition 査読有り

Hashimoto,Tan,Kim,Ishikawa

International Journal of Innovative Computing, Information and Control 5 ( 3 ) 601 - 607 2009年03月

大学/研究室/研究会紹介九州工業大学大学院聞か知能工学専攻知能制御工学講座　画像処理研究室の紹介

金　亨燮

画像通信 32 ( 1 ) 63 - 64 2009年03月

Automatic detection of blood vessels from CTA images employing morphological operation 査読有り

Itai,Yamamoto,Kim,Tan,Ishikawa

Artificial Life and Robotics 13 ( 2 ) 428 - 433 2009年02月

Performance analysis on an efficient human motion database with various motion representations 査読有り

Ashik,Tan,Kim,Ishikawa

ICCIT2008 2008年12月

Action recognition with various speeds and timed-DMHI feature vectors 査読有り

Ahad,Tan,Kim,Ishikawa

ICCIT2008 2008年12月

Solutions to motion self-occlusion problem in human activity analysi 査読有り

Ahad,Tan,Kim,Ishikawa

ICCIT2008 2008年12月

Reduction of FPs for Lung Nodules in MDCT by Use of Temporal Subtraction with Voxel-Matching Technique 査読有り

Itai,Kim,Ishikawa,Katsuragawa,Doi

Intennational Conference on Neural Infomation Processing 2008年11月

Face direction estimation based on eigen space technique 査読有り

Okubo,Sugandi,Kim,Tan,Ishikawa

International Conference on Control, Automation and Systems 1264 - 1267 2008年10月

Detection of blood vessels on CTA images of the legs 査読有り

Kozono,Yamamoto,Itai,Kim,Tan,Ishikawa

International Conference on Control, Automation and Systems 2008 1276 - 1279 2008年10月

Optimal registration method based on ICP algorithm from head CT and MR image sets 査読有り

Harada,Kim,Tan,Ishikawa,Yamamoto

International Conference on Control, Automation and Systems 2008 1268 - 1271 2008年10月

Segmentation of internal organs based on active contour models 　from CT images 査読有り

Kizuka,Kim,Tan,Ishikawa,Yamamoto

International Conference on Control, Automation and Systems 2008 1272 - 1275 2008年10月

uman Activity Recognition: Various Paradigms 査読有り

Ahad,Tan,Kim,Ishikawa

International Conference on Control, Automation and Systems 2008 1896 - 1901 2008年10月

Extraction of multi organs by use of level set method from the CT images 査読有り

Komatsu,Kim,Tan,Ishikawa,Yamamot

International Conference on Control, Automation and Systems 2008 1260 - 1263 2008年10月

3次元Elastic matching 法による腹部CT像からの血管領域の自動抽出法査読有り

小松、金、石川、タン、山本

バイオメディカル・ファジー・システム学会誌 10 ( 2 ) 65 - 72 2008年10月

可変証明下における逐次更新型色ヒストグラムを用いた適応的色抽出とそれを用いた人物の追跡査読有り

上村、タン、金、石川

バイオメディカル・ファジー・システム学会誌 10 ( 2 ) 47 - 54 2008年10月

濃度特徴を用いた胸部MDCT像からのスリガラス状陰影の自動抽出査読有り

金、板井、タン、石川

バイオメディカル・ファジー・システム学会誌 10 ( 2 ) 57 - 63 2008年10月

包囲カメラ群を用いた立体の全周復元法査読有り

守、曽根、タン、金、石川

バイオメディカル・ファジー・システム学会誌 10 ( 2 ) 39 - 45 2008年10月

濃度特徴を用いた胸部CT画像からのGGOの自動抽出査読有り

勝間田、板井、前田、金、タン、石川

バイオメディカル・ファジー・システム学会誌 10 ( 2 ) 81 - 86 2008年10月

FBI法による非造影ＭＲ画像からの血管領域の抽出と３次元表示査読有り

山本、西﨑、金、タン、石川、中村

バイオメディカル・ファジー・システム学会誌 10 ( 2 ) 73 - 79 2008年10月

Three-dimensional information projection system using a hand-held screen 査読有り

Hashimoto,Tan,Kim,Ishikawa

Proc. 2008 IEEE Int. Conf. Systems, Man and Cybernetics SMC2008 1385 - 1389 2008年10月

Template-based human motion recognition for complex activities 査読有り

Ahad,Ogata,Tan,Kim,Ishikawa

Proc. 2008 IEEE Int. Conf. Systems, Man and Cybernetics 673 - 678 2008年10月

産学連携によるコンピュータ画像診断支援システムの開発

金　亨燮

第42回北九州医工学術者協会会議講演発表 2008年10月

Motion recognition approach to solve overwriting in complex actions 査読有り

Ahad,Ogata,Tan,Kim,Ishikawa

Proc. 8th Int. Conf. Automatic Face and Gesture Recognition (IEEE FG2008) 2008年09月

経時的差分画像を用いたCT画像からの結節状陰影の検出査読有り

板井，金，石川，桂川，土井

Medical Imaging Technology 26 ( 4 ) 217 - 224 2008年09月

ICP法を用いた頭部CT・MR画像の位置合わせ査読有り

原田，金，タン，石川，山村，山本

Medical Imaging Technology 26 ( 4 ) 246 - 250 2008年09月

A simple real-time approach for action separation into action primitives 査読有り

Ahad,Uemura,Tan,Kim,Ishikawa

Proc. 1st Intl. Workshop Tracking Humans for the Evaluation of heir Motion in Image Sequences 69 - 78 2008年09月

Moment-based human motion recognition from the representation of DMHI templates 査読有り

Ahad.,Ogata.,Tan.,Kim.,Ishikawa

SICE Annual Conference 578 - 583 2008年08月

An efficient approach to human motion recognition employing large motion-database structure 査読有り

Ashik.,Tan.,Kim.,Ishikawa

Proc. of SICE Annual Conference 2239 - 2243 2008年08月

Recognizing facial expression for man-machine interaction 査読有り

Hirata,Tan,Kim,Ishikawa

SICE International Conference 2009 1621 - 1624 2008年08月

A Complex Motion Recognition Technique Employing Directional Motion Templates 査読有り

Ahad,Ogata,Tan,Kim,Ishikawa

International Journal of Innovative Computing, Information and Control 4 ( 8 ) 1943 - 1954 2008年08月

Automatic Detection of Spinal Deformity by Use of Density Features from Moire Topographic Image 査読有り

Kim,Nakano,Tan,Ishikawa,Shinomiya

Intl. Conference Computer Graphics and Visualization 2008 239 - 243 2008年07月

Automatic Detection of Multi Organs on the CT Images Using the Ribs Information and a Level Set Method 査読有り

Komatsu,Toyota,Kim,Tan,Ishikawa,Yamamoto

International Technical Conference on Circuits/Systems, Computers and Communitions 5 - 8 2008年07月

Real Time Object Tracking and Identification Using a Camera 査読有り

Budi,Kim,Tan,Ishikawa

International Technical Conference on Circuits/Systems, Computers and Communitions 1585 - 1588 2008年07月

A Method for Reducing of Computational Time on Multi-Modal Image Registration Employing Wavelet Transformation from a Set of Head CT and MR Imaging 査読有り

Yamamura,Kim,Tan,Ishikawa,Yamamoto

The International Journal of Innovative Computing, Information and Control 4 ( 7 ) 1527 - 1536 2008年07月

Automatic Detection of GGO Candidate Regions by Using Artificial Neural Network from Thoracic MDCT Images 査読有り

Katsumata,Itai,Kim,Tan,Ishikawa

International Conference on Innovative Computing, Information and Control (ICICIC2008) 5 - 5 2008年06月

A multi modal image registration method by use of optimal retrieval on the VOIs 査読有り

Yamamura,Kim,Tan,Ishikawa,Yamamoto

International Conference on Innovative Computing, Information and Control (ICICIC2008) 4 - 4 2008年06月

A block matching technique for object tracking employing peripheral increment sign correlation image 査読有り

Sugandi,Kim,Tan,Ishikawa

International Conference on Computer and Communication Engineering 113 - 117 2008年05月

Human motion representation and recognition by directional motion history images 査読有り

Fukumoto,Tan,Kim,Ishikawa

Artificial Life & Robotics 13 ( 1 ) 326 - 330 2008年04月

Human motions representation and recognition by directional motion history images 査読有り

Fukumoto,Ogata,Tan,Kim,Ishikawa

未入力 13 ( 1 ) 326 - 330 2008年04月

Motion capture employing an uncalibrated camera 査読有り

Tan,Ishikawa,Kouno,Ohbuchi,Kim

Artificial Life and Robotics 13 ( 1 ) 311 - 314 2008年04月

Motion capture employing an uncalibrated camera 査読有り

Tan,Ishikawa,Kouno,Ohbuchi,Kim

Artificial Life & Robotics 13 ( 1 ) 311 - 314 2008年04月

Complex motion separation and recognition using directional motion templates 査読有り

Ahad,Ogata,Tan,Kim,Ishikawa

Intl. Workshop on Combinatorial Image Processing 73 - 82 2008年04月

胸部ＣＴのテンポラルサブトラクション

金　亨燮

日本放射線技術学会画像分科会画像通信 31 ( 1 ) 41 - 44 2008年04月

View-based human motion recognition in the presence of outliers 査読有り

Ahad,Ogata,Tan,Kim,Ishikawa

International Journal of Biomedical Soft Computing and Human Science 13 ( 1 ) 71 - 78 2008年03月

A New Registration Method with Voxel-Matching Technique for Temporal Subtraction Image 査読有り

Itai,Kim,Ishikawa,Katsuragawa,Doi

Proc. of SPIE Medical Imaging 2008: Computer-Aided Diagnosis 2008年02月

Entire shape recovery employing virtual see-through cameras 査読有り

Mori,Sone,Tan,Kim,Ishikawa

International Symposium on Artificial Life and Robotics 577 - 580 2008年02月

Three-dimensional human motion modeling by back projection based on image-based camera calibration 査読有り

Masaoka,Tan,Kim,Ishikawa

International Symposium on Artificial Life and Robotics 734 - 737 2008年02月

Automatic Detection of Blood Vessels from CTA Images Employing Morphological Operation 査読有り

Itai,Yamamoto,Kim,Tan,Ishikawa

International Symposium on Artificial Life and Robotics 215 - 218 2008年02月

ＭＲＲＦ法を用いた複雑背景下における人物の姿勢認識査読有り

大場，タン，金，石川

バイオメディカル・ファジー・システム学会誌 10 ( 1 ) 51 - 57 2008年02月

高速な３次元物体復元法の開発査読有り

松田，タン，金，石川

バイオメディカル・ファジー・システム学会誌 10 ( 1 ) 58 - 64 2008年02月

アフィンカメラモデルに基づく逆投影を用いた物体の立体モデル化査読有り

正岡，タン，金，石川

バイオメディカル・ファジー・システム学会誌 10 ( 1 ) 122 - 125 2008年02月

左右非対称の特徴を用いたモアレ画像からの脊柱側彎症自動識別査読有り

金，タン，石川，大塚，清水，四宮

画像電子学会 36 2008年01月

Tracking of Moving Persons using Multi-camera Employing Peripheral Increment Sign Correlation Image 査読有り

Budi,Kim,Tan,Ishikawa

ICIC Express Letters, An International Journal of Research and Surveys 1 ( 2 ) 177 - 184 2007年12月

Comparative analysis between two view-based methods: MHI and DMHI 査読有り

Ahad,M.A.R.,Ogata,T.,Tan,J.K.,Kim,H.S.,Ishikawa,S.

Proc. of ICCIT 2007年12月

An Image Warping Method for Temporal Subtraction Images Employing Smoothing of Shift Vectors on MDCT Images 査読有り

Itai,Kim,Ishikawa,Katsuragawa,Ishida,Kawashita,Awai、Doi

Intennational Conference on Neural Infomation Processing 2007年11月

胸部MDCT画像からの経時差分画像上のアーチファクトの低減法査読有り

前田，板井，金，石川，桂川

Medical Imaging Technology 25 ( 5 ) 399 - 403 2007年11月

3次元画像マッチング手法による胸部マルチスライスCT画像を用いた経時差分技術の開発査読有り

板井，金，石川，桂川，石田，川下，粟井

画像電子学会 36 2007年11月

A smart automated complex motion recognition technique 査読有り

Ahad,M.A.R.,Ogata,T.,Tan,J.K.,Kim,H.S.,Ishikawa,S.

Proc. of ACCV WS 142 - 149 2007年11月

Performance of multi-directional MHI for human motion recognition in the presence of outliers 査読有り

Ahad,M.A.R.,Ogata,T.,Tan,J.K.,Kim,H.S.,Ishikawa,S.

Proc. of the 33rd Annual Conf. of the IEEE Industrial Society (IECON) 2366 - 2370 2007年11月

Recognizing Human Postures in a Complex Background Employing MRRF Method 査読有り

Ooba,Tan,Kim,Ishikawa

International Conference on Control, Automation and Systems 2007 1159 - 1162 2007年10月

An Automatic Detection Method of Spinal Deformity from Moire Topographic Images Employing Asymmetric Degree of Shoulder and Waistline 査読有り

Tanoue,Nakano,Kim,Tan,Ishikawa,Otsuka,Shimizu,Shinomiya

International Conference on Control, Automation and Systems 2007 1282 - 1285 2007年10月

A Method for Reducing of Computational Time on Image Registration Employing Wavelet Transform 査読有り

Yamamura,Kim,Tan,Ishikawa,Yamamoto

International Conference on Control, Automation and Systems 2007 1286 - 1291 2007年10月

Development of an Autonomous Robot for Face Tracking 査読有り

Nishina,Tan,Kim,Ishikawa

International Conference on Control, Automation and Systems 2007 1178 - 1181 2007年10月

A 3-D Visualization Method for Non-enhanced Fresh Blood Imaging on MR Imaging Employing a 3-D Region Growing Technique 査読有り

Nishizaki,Yamamoto,Kim,Ishikawa,Nakamura

International Conference on Control, Automation and Systems 2007 1292 - 1295 2007年10月

Automatic Detection of GGO Candidate Regions Employing Four Statistical Features on Thoracic MDCT Image 査読有り

Katsumata,Itai,Maeda,Kim,Tan,Ishikawa

International Conference on Control, Automation and Systems 2007 1278 - 1281 2007年10月

Development of a Fast 3-D Object Reconstruction Technique 査読有り

Matsuda,Tan,Kim,Ishikawa

International Conference on Control, Automation and Systems 2007 2303 - 2306 2007年10月

Automatic Detection of Ground Glass Opacity from the Thoracic MDCT Images by Using Density Features 査読有り

Kim,Nakashima,Itai,Maeda,Tan,Ishikawa

International Conference on Control, Automation and Systems 2007 1274 - 1277 2007年10月

Detection of blood vessels on the abdominal CT images employing elastic matching technique 査読有り

Komatsu,Itai,Kim,Yamamoto,Ishikawa

The first International Symposium on Information and Computer Elements 239 - 242 2007年09月

Directional Motion Templates: New Appearance-based Recognition Method 査読有り

Ahad,Ogata,Tan,Kim,Ishikawa

The first International Symposium on Information and Computer Elements 303 - 308 2007年09月

Tracking of moving persons using multi camera under the network environment 査読有り

Budi,Yamamoto,Kim,Tan,Ishikawa

The first International Symposium on Information and Computer Elements 309 - 314 2007年09月

Robust human motion recognition based on multi-directional motion representation 査読有り

Ahad,Ogata,Tan,Kim,Ishikawa

The Second International Conference on Innovative Computing, Information and Control B10-08 4 - 4 2007年09月

Tracking of Moving Object by Using a Low Resolution Image 査読有り

Budi Sugandi,Kim,Tan,Ishikawa

The Second International Conference on Innovative Computing, Information and Control 4 - 4 2007年09月

A media-art employing virtual shadows with shape recognition 査読有り

Hashimoto,Tan,Kim,Ishikaw

The Second International Conference on Innovative Computing, Information and Control A07-06 4 - 4 2007年09月

モーションデータベースを用いた人の動作の高速認識査読有り

タン、河野、石川、金、四宮

画像電子学会誌 36 ( 5 ) 738 - 746 2007年09月

立体モデル化と複合現実感の技法を用いた遮蔽空間の可視化

石川，金，タン，橋本

九工大サテライトベンチャービジネスラボラトリ年報11号 11 33 - 36 2007年09月

胸部CT画像上の経時差分画像の生成法－遺伝的アルゴリズムを用いた局所シフトベクトルの導出－

板井，前田，金，石川，桂川

九工大サテライトベンチャービジネスラボラトリ年報 11 69 - 72 2007年09月

Ground-Glass Opacity Detection by Using Correlation between Successive Slice Images 査読有り

Kim,Maekado,Tan,Ishikawa,Tsukuda

International Journal on Artificial Intelligence Tools 16 ( 4 ) 583 - 592 2007年08月

経時差分法を用いた腹部CT画像上の血管領域の抽出査読有り

小松，金，山本，タン，石川

バイオメディカル・ファジー・システム学会誌 9 ( 1 ) 11 - 17 2007年07月

複合現実感と3次元モデル化を用いた遮蔽空間の可視化査読有り

橋本、タン、金、石川

バイオメディカル・ファジー・システム学会誌 9 ( 1 ) 43 - 51 2007年07月

Automatic Classification of Spinal Deformity by Using Four Symmetrical Features on the Moire Images 査読有り

Kim,Nakano,Tan,Ishikawa,Otsuka,Shimizu,Shinomiya

International Conference on Informatics in Control, Automation and Robotics 99 - 106 2007年05月

A segmentation method of lung areas by using SNAKES and automatic detection of abnormal shadow on the areas 査読有り

Itai,Kim,Ishikawa,Yamamoto,Nakamura

The International Journal of Innovative Computing, Information and Control 3 ( 2 ) 277 - 284 2007年04月

3D temporal subtraction on multislice CT images using nonlinear warping technique 査読有り

Ishida,Katsuragawa,Kawashita,Kim,Itai,Awai,Li,Doi

Proc. of SPIE Medical Imaging 2007: Computer-Aided Diagnosis 6514 651431.1 - 6514631.8 2007年02月

Development of temporal subtraction multislice CT images by using a 3D local matching with a genetic algorithm 査読有り

Itai,Kim,Ishikawa,Katsuragawa,Ishida,Doi

Radiological Society of North America 779 - 779 2006年12月

Automatic Judgment of Spinal Deformity Based on Back Propagation on Neural Network 査読有り

Kim,Tan,Ishikawa,Khalid,Ohtsuka,Shimizu,Shinomiya

The International Journal of Innovative Computing, Information and Control 2 ( 6 ) 1271 - 1279 2006年12月

A method for image registration by maximization of mutual information 査読有り

Yamamura,Kim,Yamamoto

SICE-ICASE International Joint Conference 2006 1469 - 1472 2006年10月

Detection of blood vessel on the abdominal CT images employing temporal subtraction method 査読有り

Komatsu,Maeda,Kim,Yamamoto,Okuda

SICE-ICASE International Joint Conference 2006 1473 - 1476 2006年10月

Automatic segmentation of liver region employing rib cage and its 3-D displaying 査読有り

Maeda,Komatsu,Kim,Yamamoto,Okuda

SICE-ICASE International Joint Conference 2006 1465 - 1468 2006年10月

Automatic Detection of Ground-Glass Opacity Shadows by Three Characteristics on MDCT Images 査読有り

Kim,Mori,Itai,Ishikawa,Yamamoto,Nakamura

World Congress on Medical Physics and Biomedical Engineering 2006 2196 - 2199 2006年09月

Spinal deformity detection from moire topographic image based on evaluating asymmetric degree 査読有り

Kim,Ushijima,Tan,Ishikawa,Otsuka,Shimizu,Shinomiya

World Congress on Medical Physics and Biomedical Engineering 2006 2192 - 2195 2006年09月

3-D Elastic Matching Technique for Temporal Subtraction Employing Thorax MDCT Images 査読有り

Itai,Kim,Ishikawa,Katsuragawa,Ishida,Kawashita,Awai,Li,Doi

World Congress on Medical Physics and Biomedical Engineering 2006 2188 - 2191 2006年09月

２次元投影像を用いた相互情報量に基づく３次元頭部CT，MR画像の位置合わせ査読有り

山村，金，石川，山本

Medical Imaging Technology 24 ( 4 ) 294 - 299 2006年09月

Temporal subtraction on 3D CT images by using nonlinear image warping technique 査読有り

Ishida,Katsuragawa,Kawashita,Kim,Itai,Awai,Li,Doi

International Journal of Computer Assisted Radiology and Surgery 468 - 468 2006年07月

胸部CT像からのスライス間相関を用いたスリガラス状陰影の自動抽出査読有り

金,前門,タン,石川,佃

画像電子学会 35 ( 2 ) 110 - 116 2006年03月

胸部CT画像における経時的サブトラクション

桂川、石田、川下、金、板井、粟井、土井

文部科学省科学研究費補助金特定領域研究「多次元医用画像の知的診断支援」第3回シンポジウム論文集 123 - 130 2006年01月

胸部CT画像における経時的サブトラクション

桂川,石田,川下,金,板井,粟井,土井

文部科学省科学研究費補助金特定領域研究「多次元医用画像の知的診断支援」第3回シンポジウム論文集 123 - 130 2006年01月

Automatic Extraction of Ground-Glass Opacity Shadows on CT Images of the Thorax by Correlation between Successive Slices 査読有り

Kim,Maekado,Tan,Ishikawa,Tsukuda

IEEE International Conference on Tools with Artificial Intelligence 607 - 612 2005年11月

Automatic segmentation of lung areas based on SNAKES and extraction of abnormal areas 査読有り

Itai,Kim,Ishikawa,Katsuragawa,Nakamura,Yamamoto

IEEE International Conference on Tools with Artificial Intelligence 377 - 381 2005年11月

Automatic extraction of abnormal areas on CT images of the lung regions 査読有り

Itai,Komatsu,Kim,Ishikawa,Katsuragawa,Nakamura,Yamamoto

International Symposium on Advanced Intelligent Systems 390 - 393 2005年10月

Extraction of the liver region on the abdominal CT image employing temporal subtraction method 査読有り

Komatsu,Itai,Kim,Ishikawa,Katsuragawa,Okuda,Yamamoto

International Symposium on Advanced Intelligent Systems 400 - 403 2005年10月

胸部MDCT像からの肺野領域のセグメンテーションとスライス間ボクセル密度計測による病巣部候補領域の自動抽出査読有り

板井，金，石川，桂川，中村，山本

バイオメディカル・ファジー・システム学会誌 7 ( 1 ) 33 - 40 2005年10月

ステガノグラフィによるDICOM画像の埋め込み

金，大西，石川，桂川，新見，山本

信学技報 MI2005-36 31 - 34 2005年09月

Spinal Deformity Detection Employing Back Propagation on Neural Network 査読有り

Kim,Tan,Ishikawa,Khalid,Max Viergever,Otsuka,Shinomiya

International Conference on Advances in Pattern Recognition 719 - 725 2005年08月

肺野領域内の病変部候補領域の自動抽出と評価

板井，小松，金，石川，中村，山本

第32回北九州医工学術者協会誌 16 75 - 76 2005年06月

可動モーションキャプチャ法を用いた負荷歩行のモデル化

タン，野田，倪，金，石川

第３２回北九州医工学術者協会誌 16 69 - 70 2005年06月

経時差分を用いた腹部CT画像上の領域抽出と3次元表示

小松，板井，金，石川，桂川，奥田，山本

第32回北九州医工学術者協会誌 16 71 - 74 2005年06月

Segmentation of Lung Area and Extraction of Abnormal Area on CT Images of the Thorax 査読有り

Mori,Kim,Ishikawa,Nakamura,Yamoto

The 30th Annual Conference of the IEEE Industrial Electronics Society 2004年11月

Automatic Spinal deformity Detection Based on Asymmetric Degree of Moire Topographic Images 査読有り

Ushjima,Kim,Ishikawa,Otsuka,Shimizu,Nakada,Shinomiya

The 30th Annual Conference of the IEEE Industrial Electronics Society 2004年11月

Automatic extraction of liver regions and classification of liver cirrhosis from the abdominal CT images 査読有り

Kim,Honda,Ishikawa,Okuda,Yamamoto,Shigehiko Katsuragawa,Max A. Viergever

The 30th Annual Conference of the IEEE Industrial Electronics Society 2004年11月

Scoliosis Detection Based on Difference of Apexes Position and Angle on Moire Topographic Images 査読有り

Hyoungseop Kim,Hideki Ushijima,Seiji Ishikawa,Yoshinori Otsuka,Hisashi Shimizu,Takasi Shinomiya,and Max A. Viergever

Computer Assisted Radiology and Surgery 1294 - 1294 2004年06月

腹部MDCT像上の肝臓領域の自動抽出と異常陰影の抽出

金，藤原，石川，奥田，山本

第２９回北九州医工学術者協会誌 15 77 - 78 2004年04月

腹部CT像からの肝臓領域と異常陰影候補領域の抽出査読有り

金，萩原，石川，四宮，佃

バイオメディカル・ファジー・システム学会誌 6 ( 1 ) 42 - 49 2004年04月

胸部ＣＴ像からの肺野領域とスリガラス状陰影の自動抽出査読有り

金，前門，石川，佃

バイオメディカル・ファジー・システム学会誌 6 ( 1 ) 50 - 56 2004年04月

Automatic Spinal Deformity Detection Based on Neural Network 査読有り

Kim,Ishikawa,Khalid,Otsuka,Shimizu,Nakada,Shinomiya,Viergever

Medical Image Computing & Computer Assisted Intervention 802 - 809 2003年11月

Automatic Extraction of Ground-glass Opacities on Lung CT Images by Histogram Analysis 査読有り

Maekado Masaki，Kim Hyoungseop，Ishikawa Seiji， Tsukuda Masaaki

International Conference on Control, Automation and Systems 2352 - 2355 2003年10月

Discrimination of Spinal Deformity Employing Discriminant Analysis on the Moire Images 査読有り

Hyoungseop Kim,Seiji Ishikawa,Yoshinori Otsuka,Hisashi Shimizu,Yasuhiro Nakada,Takashi Shinomiya

International Conference on Control, Automation and Systems 1990 - 1993 2003年10月

On a multi-agent system for assiting human intention 査読有り

Hajime Tawaki,Joo Kooi Tan,Hyoungseop Kim,Seiji Ishikawa

International Conference on Control, Automation and Systems 1126 - 1129 2003年10月

判別分析によるモアレ画像からの脊柱側彎症自動識別査読有り

金　亨燮，石川　聖二，大塚　嘉則，清水　尚，中田　康弘，四宮　孝史

バイオメディカル・ファジー・システム学会誌 5 ( 1 ) 1 - 8 2003年08月

VRによる融合空間の生成と応用―第Ⅱ期報告

石川，金，タン，廣橋，山口

平成14年度九州工業大学サテライトベンチャービジネスラボラトリー研究成果報告資料集 65 - 70 2003年04月

左右濃度差評価に基づくモアレ画像からの脊柱側彎症自動識別査読有り

金，石川，大塚，清水，四宮

Medical Imaging Technology 21 ( 2 ) 131 - 138 2003年03月

VRによる融合空間の生成と応用―第Ⅱ期報告

石川，金，タン，廣橋，山口

平成14年度九州工業大学サテライトベンチャービジネスラボラトリー研究成果報告資料集 65 - 69 2003年03月

多変量解析によるモアレ画像からの脊柱側彎症の自動識別

金，沼田，石川，大塚，清水，四宮

電子情報通信学会信学技報MBE2002-79 25 - 28 2002年12月

NNによる脊柱側彎症自動識別の試み

金，石川，Marzuki Khalid，大塚，清水，四宮

画像電子学会第198回研究会講演予稿集 47 - 50 2002年11月

3-D deformable objects modeling employing mobile cameras

Ishikawa,S.,Tan,J. K.,Kim,H. S.,Khalid,M

サテライトベンチャービジネスラボラトリー年報 6 104 - 107 2002年08月

Creating a virtual human from motion capture data

Ishikawa,S.,Tan,J. K.,Kim,H. S.,Khalid,M.,Hirohashi,D.,Yamaguchi,I.

サテライトベンチャービジネスラボラトリー年報 6 70 - 73 2002年08月

A method of 3-D object modeling employing mobile cameras

Ishikawa,S.,Tan,J. K.,Kim,H. S.,Khalid,M.

九州工業大学サテライトベンチャービジネスラボラトリー研究成果報告会資料集 65 - 68 2002年03月

Creating a virtual human from motion capture data

Ishikawa,S.,Tan,J. K.,Kim,H. S.,Khalid,M.,Hirohashi,D.,Yamaguchi,I

九州工業大学サテライトベンチャービジネスラボラトリー研究成果報告会資料集 69 - 70 2002年03月

Automatic Scoliosis Detection Based on Local Centroids Evaluation on Moire Topographic Images of Human Backs 査読有り

Kim,Ishikawa,Ohtsuka,Shimizu,Shinomiya,Max

IEEE Transactions on Medical Imaging Vol.20 ( 12 ) No.12 1314 - 1320 2001年12月

VRによる融合空間の生成と応用―仮想オブジェクト生成のためのモーションキャプチャ法

タンジュークイ,石川聖二,金亨燮,加藤了三

九州工業大学サテライト・ベンチャー・ビジネス・ラボラトリー年報 5 106 - 109 2001年05月

VRによる融合空間の生成と応用―仮想オブジェクトのオクルージョンの実現法

石川聖二,タンジュークイ,金亨燮,山下貴夫，加藤了三

九州工業大学サテライト・ベンチャー・ビジネス・ラボラトリー年報 5 110 - 113 2001年05月

CT像からの肝臓領域の自動抽出と容積の計測査読有り

金　亨燮、野田　耕司、石川　聖二、佃　正明

バイオメディカル・ファジー・システム学会誌 3 ( 1 ) 59 - 65 2001年04月

Development of a Mobile Robot for Handicapped People 査読有り

IGAWA,KIM,ISHIKAWA

International Conference on Control, Automation and Systems 59 - 62 2001年04月

Producing a Virtual Object with Realistic Motion for a Mixed Reality Space 査読有り

Hirohashi,Tan,Kim,Ishikawa

International Conference on Control, Automation and Systems 1084 - 1084 2001年04月

Spinal Deformity Detection Based on Characteristic in Local Region from moire images 査読有り

Hyoungseop Kim,Seiji Ishikawa,Yoshinori Othsuka,Hisashi Shimizu,Takashi Shinomiya

International Conference on Control, Automation and Systems 818 - 818 2001年04月

Solving a Dress Problem for a Human Model Recognition 査読有り

未入力

Proceeding of SICE Annual Conference CD-ROM: 212 A-3 2001年04月

Automatic Extraction of Liver Regions from Abdominal CT Images 査読有り

KIM,NODA,ISHIKAWA、 TSUKUDA

2nd International Symposium on Advanced Intelligent Systems 76 - 80 2001年04月

ヒト前十字靭帯の3次元画像の再構築査読有り

山本，石川，TAN，金，廣川，　川田

日本臨床バイオメカニクス学会誌 21 151 - 156 2000年09月

Analysis of brushing strokes using a motion capture technique 査読有り

Naito,Hagiwara,Hitaka,Ishikawa,Tan,Kim,Yokota

The American Academy of Periodontology 86th Annual Meeting Abstract Book 1910 - 1910 2000年04月

Man-Machine collaborative work based on visual communication 査読有り

Ishikawa,Kim,Inoue,Shiramura

Proc. TENCON2000 321 - 325 2000年04月

マルチエージェント空間における人と機械の協調作業法-視的対話による手渡し作業

未入力

九州工業大学研究報告(工学) 72 1 - 8 2000年04月

Man-Machine Cooperation in a Multi-Agent Space-Objects Delivery by Visual Communication

未入力

未入力 Proc. TENCON2000 721 - 8 2000年04月

モアレ画像からの重心位置評価に基づく脊柱側彎症の識別査読有り

金　亨燮，石田　和史，石川　聖二，大塚　嘉則，清水　尚

電子情報通信学会論文誌 J-83 ( 3 ) 879 - 886 2000年03月

VRによる融合空間の生成と応用

福田健博,石川聖二,金亨燮,廣橋大輔,加藤了三

九州工業大学サテライト・ベンチャー・ビジネス・ラボラトリー年報 4 94 - 101 2000年03月

VRによる融合空間の生成と応用―仮想オブジェクトの融合表示Ⅱ

福田健博,廣橋大輔,金亨燮,石川聖二,加藤了三

九州工業大学サテライト・ベンチャー・ビジネス・ラボラトリー年報 3 74 - 77 1999年05月

Spinal Deformity Detection Based on 2-D evaluation of Asymmetry of Moire Patterns of the Human Back 査読有り

Kim,Motoie,Ishikawa,Ohtsuka,Shimizu

Proceedings of 1999 International Technical Conference on Circuits/System, Computer and Communications 673 - 676 1999年04月

人体背面モアレ像上の2つの特徴軸を用いた脊柱側彎の自動分類査読有り

金，石川，加藤，大塚

日本医用画像工学会誌 17 ( 2 ) 148 - 154 1999年03月

近似的な回転対称形の対称中心の抽出法査読有り

金，本村，石川

電子情報通信学会論文誌 J82 ( 2 ) 158 - 161 1999年01月

VRによる融合空間の生成と応用―実空間と仮想空間の融合法

加藤了三,石川聖二,金亨燮,稲田智久,落丸直詩,福田健博

九州工業大学サテライト・ベンチャー・ビジネス・ラボラトリー年報 2 41 - 41 1998年05月

Finding the Center of Rotational Symmetry from Noisy Forms 査読有り

HyoungSeop KIM,Nachi MOTOMURA,Seiji ISHIKAWA

Proceedings of 3rd Asian Conference on Computer Vision 354 - 354 1998年04月

Automatic Judgment of Spinal Deformity from Moire Images Employing Asymmetry of Local Centroids Location 査読有り

KIM,ISHIDA,ISHIKAWA,OTSUKA

Proceedings of IAPR Workshop on Machine Vision Applications 242 - 242 1998年04月

Recognizing Asymmetric Moire Patterns for Human Spinal Deformity Detection 査読有り

HyoungSeop KIM,Hiroshi UENO,Seiji ISHIKAWA,Yoshinori OTSUKA

Proceedings of Korea Automatic Control Conference 568 - 568 1997年04月

Use of Handy Communication System with Digital Camera, “Color Zaurus”(MI-10DC, Sharp Co.Ltd), for Home-visit Nursing 査読有り

Matsuoka,Hisanaga,Kamiya,Baba,Fujishima,Yokohama,Ishikawa,Kim

Proceedings of the 3rd International Conference Aspects of Telemedicine 137 - 137 1997年04月

Judging Spinal Deformity by Two Characteristic Axes on a Human Back 査読有り

ISHIKAWA,EGUCHI,YAMAGUCHI,KIM,OTSUKA

Proceedings of Korea Automatic Control Conference 438 - 441 1996年04月

Extracting the Axis of Potential Axial Symmetry Employing Variance Minimization 査読有り

HyoungSeop KIM,Seiji ISHIKAWA

Proceedings of Korea Automatic Control Conference 434 - 434 1996年04月

Extracting Gall Bladders from Ultrasound Images 査読有り

KIM,ISHIKAWA,KATO,TSUKUDA,MATSUOKA

Proceedings of Korea Automatic Control Conference 248 - 251 1995年04月

　前のページ-次のページ　

1