研究者詳細 - 神谷　亨

論文 - 神谷　亨

分割表示757 件中 1 - 20 件目／全件表示 >>

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

Kyutacar

Scopus

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

Kyutacar

Scopus

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.

Kyutacar

Scopus

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.

Kyutacar

Scopus

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.

Kyutacar

Scopus

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.

Kyutacar

Scopus

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.

Kyutacar

Scopus

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.

Kyutacar

Scopus

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.

Kyutacar

Scopus

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

Kyutacar

Scopus

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

Scopus

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

Scopus

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

Kyutacar

Scopus

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

　前のページ- 次のページ　

<神谷 亨>

論文 - 神谷 亨

<神谷　亨>

論文 - 神谷　亨