研究者詳細 - 神谷　亨

論文 - 神谷　亨

分割表示751 件中 221 - 240 件目／全件表示 >>

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

Scopus

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

Scopus

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

Scopus

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

Scopus

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

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

CiNii Article

CiNii Research

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

Scopus

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

Scopus

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

Scopus

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

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85085516392&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

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85108782511&origin=inward

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

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85108848194&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

Scopus

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

Scopus

その他リンク： https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85069818135&origin=inward

　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