Updated on 2021/11/01

写真a

 
ITOH Hayato
 
Organization
Graduate School of Informatics Department of Intelligent Systems 2 Designated assistant professor
Title
Designated assistant professor
Contact information
メールアドレス
External link

Degree 1

  1. 博士(工学) ( 2017.3   千葉大学 ) 

Research Interests 3

  1. Mathematical Engineering

  2. Pattern Recognition

  3. Computer-Aided Diagnosis

Research Areas 3

  1. Informatics / Statistical science

  2. Informatics / Intelligent informatics

  3. Informatics / Perceptual information processing

Research History 6

  1. Nagoya University   Graduate School of Informatics Department of Intelligent Systems   Designated assistant professor

    2020.11

      More details

    Country:Japan

  2. Nagoya University   Graduate School of Informatics Department of Intelligent Systems   Researcher

    2017.4 - 2020.11

      More details

    Country:Japan

  3. Czech Technical University in Prague   Faculty of Electronic Engineering Department of Cybernetics Center for Machine Perception   Research Visitor

    2013.4 - 2013.7

      More details

    Country:Czech Republic

  4. National Institute of Informatics   Special Collaboration with Research Students

    2013 - 2017

      More details

    Country:Japan

  5. National Institute of Informatics   Special Collaboration with Research Students

    2010 - 2011

      More details

    Country:Japan

  6. National Institute of Informatics   kenkyu-kensyusei

    2009 - 2010

      More details

    Country:Japan

▼display all

Education 3

  1. Chiba University   Graduate School of Advanced Integration Science   Information Sciences

    2012.4 - 2017.3

      More details

    Country: Japan

  2. Chiba University   Graduate School of Advanced Integration Science   Information Sciences

    2010.4 - 2012.3

      More details

    Country: Japan

  3. Chiba University   Faculty of Engineering   Department of Information and Image Sciences

    2006.4 - 2010.3

      More details

    Country: Japan

Committee Memberships 7

  1. JSCAS26   Local Organizer  

    2017.4 - 2017.10   

      More details

    Committee type:Academic society

  2. The workshop on mathematical and computational methods in biomedical imaging and image analysis (MCBMIIA2016)   System Administrator  

    2016.5 - 2016.11   

      More details

    Committee type:Academic society

  3. The workshop on mathematical and computational methods in biomedical imaging and image analysis (MCBMIIA2015)   System Administrator  

    2015.2 - 2015.11   

      More details

    Committee type:Academic society

  4. International Workshop on Computer Vision in Vehicle Technology: From Earth to Mars -In conjunction with ICCV 2013-   System Administrator  

    2013.4 - 2013.12   

      More details

    Committee type:Academic society

  5. International Workshop on Computer Vision in Vehicle Technology: From Earth to Mars -In conjunction with ECCV 2012-   System Administrator  

    2012.4 - 2012.10   

      More details

    Committee type:Academic society

  6. Joint IAPR International Workshops on Structural and Syntactic Pattern Recognition (SSPR 2012) and Statistical Techniques in Pattern Recognition (SPR 2012)   Local Organizer  

    2011.8 - 2012.11   

      More details

    Committee type:Academic society

  7. nternational Workshop on Computer Vision in Vehicle Technology: From Earth to Mars -In conjunction with ICCV 2011-   System Administrator  

    2011.1 - 2011.11   

      More details

    Committee type:Academic society

▼display all

Awards 3

  1. JSCAS Speech Award

    2018.10   The Japan Society of Computer Aided Surgery  

     More details

    Award type:Award from Japanese society, conference, symposium, etc.  Country:Japan

  2. 日本医用画像工学会 奨励賞

    2018.9   The Japanese Society of Medical Imaging Technology  

     More details

    Award type:Award from Japanese society, conference, symposium, etc. 

  3. MI-ken Encouragement Award

    2018.5   The Institute of Electronics, Information and Communication Engineers MI-ken  

     More details

    Award type:Award from Japanese society, conference, symposium, etc.  Country:Japan

 

Papers 70

  1. Binary polyp-size classification based on deep-learned spatial information Reviewed International coauthorship

    Hayato Itoh, Masahiro Oda, Kai Jiang, Yuichi Mori, Masashi Misawa, Shin-Ei Kudo, Kenichiro Imai, Sayo Ito, Kinichi Hotta, Kensaku Mori

    International Journal of Computer Assisted Radiology and Surgery   Vol. in Print   2021.9

     More details

    Authorship:Lead author   Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1007/s11548-021-02477-z

  2. Unsupervised colonoscopic depth estimation by domain translations with a Lambertian-reflection keeping auxiliary task Reviewed International coauthorship

    Itoh Hayato, Oda Masahiro, Mori Yuichi, Misawa Masashi, Kudo Shin-Ei, Imai Kenichiro, Ito Sayo, Hotta Kinichi, Takabatake Hirotsugu, Mori Masaki, Natori Hiroshi, Mori Kensaku

    INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY   Vol. 16 ( 6 ) page: 989 - 1001   2021.5

     More details

    Authorship:Lead author   Language:English   Publishing type:Research paper (scientific journal)   Publisher:International Journal of Computer Assisted Radiology and Surgery  

    DOI: 10.1007/s11548-021-02398-x

    Web of Science

    Scopus

    PubMed

  3. Robust endocytoscopic image classification based on higher-order symmetric tensor analysis and multi-scale topological statistics. Reviewed

    Hayato Itoh, Yukitaka Nimura, Yuichi Mori, Masashi Misawa, Shin-ei Kudo, Kinichi Hotta, Kazuo Ohtsuka, Shoichi Saito, Yutaka Saito, Hiroaki Ikematsu, Yuichiro Hayashi, Masahiro Oda, Kensaku Mori

    INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY   Vol. 15 ( 12 ) page: 2049 - 2059   2020

     More details

    Authorship:Lead author   Publishing type:Research paper (scientific journal)  

    DOI: 10.1007/s11548-020-02255-3

    researchmap

    Other Link: https://dblp.uni-trier.de/db/journals/cars/cars15.html#ItohNMMKHOSSIHO20

  4. Multilinear Subspace Method Based on Geodesic Distance for Volumetric Object Classification. Reviewed

    Hayato Itoh, Atsushi Imiya

        page: 672 - 683   2019

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-030-29888-3_55

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/caip/caip2019-1.html#ItohI19

  5. Discriminative Feature Selection by Optimal Manifold Search for Neoplastic Image Recognition. Reviewed

    Hayato Itoh, Yuichi Mori, Masashi Misawa, Masahiro Oda, Shin-ei Kudo, Kensaku Mori

        page: 534 - 549   2018

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-030-11018-5_43

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/eccv/eccv2018w4.html#ItohMMOKM18

  6. Relaxed Optimisation for Tensor Principal Component Analysis and Applications to Recognition, Compression and Retrieval of Volumetric Shapes Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai

    Imaging, Vision and Learning Based on Optimization and PDEs     page: 165 - 200   2018

     More details

    Authorship:Lead author   Publishing type:Part of collection (book)   Publisher:Springer International Publishing  

    DOI: 10.1007/978-3-319-91274-5_8

    researchmap

  7. Distances Between Tensor Subspaces. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 50 - 59   2018

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.3233/978-1-61499-929-4-50

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/appis/appis2018.html#ItohIS18

  8. Towards Automated Colonoscopy Diagnosis: Binary Polyp Size Estimation via Unsupervised Depth Learning. Reviewed

    Hayato Itoh, Holger R. Roth, Le Lu 0001, Masahiro Oda, Masashi Misawa, Yuichi Mori, Shin-ei Kudo, Kensaku Mori

        page: 611 - 619   2018

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-030-00934-2_68

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/miccai/miccai2018-2.html#ItohRLOMMKM18

  9. Dimension Reduction and Construction of Feature Space for Image Pattern Recognition. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    J. Math. Imaging Vis.   Vol. 56 ( 1 ) page: 1 - 31   2016

     More details

    Authorship:Lead author   Publishing type:Research paper (scientific journal)  

    DOI: 10.1007/s10851-015-0629-1

    researchmap

  10. Pattern recognition in multilinear space and its applications: mathematics, computational algorithms and numerical validations. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    Mach. Vis. Appl.   Vol. 27 ( 8 ) page: 1259 - 1273   2016

     More details

    Authorship:Lead author   Publishing type:Research paper (scientific journal)  

    DOI: 10.1007/s00138-016-0806-2

    researchmap

  11. Mathematical Aspects of Tensor Subspace Method. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 37 - 48   2016

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-49055-7_4

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/sspr/sspr2016.html#ItohIS16

  12. Topology-Preserving Dimension-Reduction Methods for Image Pattern Recognition. Reviewed

    Hayato Itoh, Tomoya Sakai 0002, Kazuhiko Kawamoto, Atsushi Imiya

        page: 195 - 204   2013

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-642-38886-6_19

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/scia/scia2013.html#ItohSKI13

  13. Artificial intelligence and computer-aided diagnosis for colonoscopy: where do we stand now? Reviewed International coauthorship

    Shin-Ei Kudo, Yuichi Mori, Usama M. Abdel-aal, Masashi Misawa, Hayato Itoh, Masahiro Oda, Kensaku Mori

    Translational Gastroenterology and Hepatology   Vol. 6   2021.10

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.21037/tgh.2019.12.14

  14. Impact of the clinical use of artificial intelligence-assisted neoplasia detection for colonoscopy: a large-scale prospective, propensity score-matched study (with video). Reviewed International coauthorship International journal

    Misaki Ishiyama, Shin-Ei Kudo, Masashi Misawa, Yuichi Mori, Maeda Yasuharu, Katsuro Ichimasa, Toyoki Kudo, Takemasa Hayashi, Kunihiko Wakamura, Hideyuki Miyachi, Fumio Ishida, Hayato Itoh, Masahiro Oda, Kensaku Mori

    Gastrointestinal endoscopy     2021.8

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    BACKGROUND AND AIMS: Recently, the use of computer-aided detection (CADe) for colonoscopy has been investigated to improve the adenoma detection rate (ADR). We aimed to assess the efficacy of a regulatory-approved CADe in a large-scale study with high numbers of patients and endoscopists. METHODS: This was a propensity score matched prospective study that took place at a university hospital between July 2020 and December 2020. We recruited patients aged ≥20 years who were scheduled for colonoscopy. Patients with polyposis, inflammatory bowel disease, or incomplete colonoscopy were excluded. We used a regulatory-approved CADe and conducted a propensity score matching-based comparison of the ADR between patients examined with and without CADe as the primary outcome. RESULTS: During the study period, 2,261 patients underwent colonoscopy with the CADe system or routine colonoscopy and 175 patients were excluded in accordance with the exclusion criteria. Thirty endoscopists (9 nonexperts and 21 experts) were involved in this study. Propensity score matching was conducted using 5 factors, resulting in 1,836 patients included in the analysis (918 patients in each group). The ADR was significantly higher in the CADe group than in the control group (26.4% vs 19.9%, respectively; relative risk [OR], 1.32; 95% confidence interval [CI], 1.12-1.57); however, there was no significant increase in the advanced neoplasia detection rate (3.7% vs 2.9%, respectively). CONCLUSIONS: The use of the CADe system for colonoscopy significantly increased the ADR in a large-scale prospective study including 30 endoscopists. (UMIN-CTR: UMIN000040677.).

    DOI: 10.1016/j.gie.2021.07.022

    PubMed

    researchmap

  15. Artificial intelligence-assisted colonic endocytoscopy for cancer recognition: a multicenter study Reviewed International coauthorship

    Yuichi Mori, Shin-ei Kudo, Masashi Misawa, Kinichi Hotta, Ohtsuka Kazuo, Shoichi Saito, Hiroaki Ikematsu, Yutaka Saito, Takahisa Matsuda, Takeda Kenichi, Toyoki Kudo, Tetsuo Nemoto, Hayato Itoh, Kensaku Mori

    Endoscopy International Open   Vol. 9 ( 7 ) page: E1004 - E1011   2021.7

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

    DOI: 10.1055/a-1475-3624

  16. Development of a computer-aided detection system for colonoscopy and a publicly accessible large colonoscopy video database (with video) Reviewed

    Misawa Masashi, Kudo Shin-ei, Mori Yuichi, Hotta Kinichi, Ohtsuka Kazuo, Matsuda Takahisa, Saito Shoichi, Kudo Toyoki, Baba Toshiyuki, Ishida Fumio, Itoh Hayato, Oda Masahiro, Mori Kensaku

    GASTROINTESTINAL ENDOSCOPY   Vol. 93 ( 4 ) page: 960 - +   2021.4

     More details

    Publisher:Gastrointestinal Endoscopy  

    DOI: 10.1016/j.gie.2020.07.060

    Web of Science

    Scopus

    PubMed

  17. Artificial Intelligence System to Determine Risk of T1 Colorectal Cancer Metastasis to Lymph Node Reviewed

    Kudo Shin-ei, Ichimasa Katsuro, Villard Benjamin, Mori Yuichi, Misawa Masashi, Saito Shoichi, Hotta Kinichi, Saito Yutaka, Matsuda Takahisa, Yamada Kazutaka, Mitani Toshifumi, Ohtsuka Kazuo, Chino Akiko, Ide Daisuke, Imai Kenichiro, Kishida Yoshihiro, Nakamura Keiko, Saiki Yasumitsu, Tanaka Masafumi, Hoteya Shu, Yamashita Satoshi, Kinugasa Yusuke, Fukuda Masayoshi, Kudo Toyoki, Miyachi Hideyuki, Ishida Fumio, Itoh Hayato, Oda Masahiro, Mori Kensaku

    GASTROENTEROLOGY   Vol. 160 ( 4 ) page: 1075 - +   2021.3

     More details

  18. Dense-layer-based YOLO-v3 for detection and localization of colon perforations Reviewed

    Kai Jiang, Hayato Itoh, Masahiro Oda, Taishi Okumura, Yuichi Mori, Masashi Misawa, Takemasa Hayashi, Shin-Ei Kudo, Kensaku Mori

    Medical Imaging 2021: Computer-Aided Diagnosis     2021.2

     More details

    Publishing type:Research paper (international conference proceedings)   Publisher:SPIE  

    DOI: 10.1117/12.2582300

    researchmap

  19. Intestinal region reconstruction of ileus cases from 3D CT images based on graphical representation and its visualization Reviewed

    Hirohisa Oda, Yuichiro Hayashi, Takayuki Kitasaka, Yudai Tamada, Aitaro Takimoto, Akinari Hinoki, Hiroo Uchida, Kojiro Suzuki, Hayato Itoh, Masahiro Oda, Kensaku Mori

    Medical Imaging 2021: Computer-Aided Diagnosis     2021.2

     More details

    Publishing type:Research paper (international conference proceedings)   Publisher:SPIE  

    DOI: 10.1117/12.2581261

    researchmap

  20. Extremely imbalanced subarachnoid hemorrhage detection based on DenseNet-LSTM network with class-balanced loss and transfer learning Reviewed

    Zhongyang Lu, Masahiro Oda, Yuichiro Hayashi, Tao Hu, Hayato Itoh, Takeyuki Watadani, Osamu Abe, Masahiro Hashimoto, Masahiro Jinzaki, Kensaku Mori

    Medical Imaging 2021: Computer-Aided Diagnosis     2021.2

     More details

    Publishing type:Research paper (international conference proceedings)   Publisher:SPIE  

    DOI: 10.1117/12.2582088

    researchmap

  21. Single-shot three-dimensional reconstruction for colonoscopic image analysis Reviewed

    Hayato Itoh, Masahiro Oda, Yuichi Mori, Masashi Misawa, Shin-Ei Kudo, Kinnichi Hotta, Hirotsugu Takabatake, Masaki Mori, Hiroshi Natori, Kensaku Mori

    Medical Imaging 2021: Image-Guided Procedures, Robotic Interventions, and Modeling     2021.2

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)   Publisher:SPIE  

    DOI: 10.1117/12.2582660

    researchmap

  22. Current status and future perspective on artificial intelligence for lower endoscopy Reviewed

    Digestive Endoscopy   Vol. 33 ( 2 ) page: 273 - 284   2021.1

     More details

    Publisher:Digestive Endoscopy  

    DOI: 10.1111/den.13847

    Scopus

    PubMed

  23. Uncertainty Meets 3D-Spatial Feature in Colonoscopic Polyp-Size Determination Reviewed International coauthorship

    Hayato Itoh, Masahiro Oda, Kai Jiang, Yuichi Mori, Masashi Misawa, Shin-Ei Kudo, Kenichiro Imai, Sayo Ito, Kinichi Hotta, Kensaku Mori

    Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization   Vol. in Print   2021

     More details

    Authorship:Lead author   Language:English   Publishing type:Research paper (scientific journal)  

  24. Depth Estimation from Single-shot Monocular Endoscope Image Using Image Domain Adaptation And Edge-Aware Depth Estimation Reviewed

    Masahiro Oda, Hayato Itoh, Kiyohito Tanaka, Hirotsugu Takabatake, Masaki Mori, Hiroshi Natori, Kensaku Mori

    Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization   Vol. in Print   2021

     More details

    Language:English   Publishing type:Research paper (scientific journal)  

  25. Artificial Intelligence-assisted System Improves Endoscopic Identification of Colorectal Neoplasms Reviewed

    Shin-ei Kudo, Masashi Misawa, Yuichi Mori, Kinichi Hotta, Kazuo Ohtsuka, Hiroaki Ikematsu, Yutaka Saito, Kenichi Takeda, Hiroki Nakamura, Katsuro Ichimasa, Tomoyuki Ishigaki, Naoya Toyoshima, Toyoki Kudo, Takemasa Hayashi, Kunihiko Wakamura, Toshiyuki Baba, Fumio Ishida, Haruhiro Inoue, Hayato Itoh, Masahiro Oda, Kensaku Mori

    CLINICAL GASTROENTEROLOGY AND HEPATOLOGY   Vol. 18 ( 8 ) page: 1874 - +   2020.7

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:ELSEVIER SCIENCE INC  

    BACKGROUND & AIMS: Precise optical diagnosis of colorectal polyps could improve the cost-effectiveness of colonoscopy and reduce polypectomy-related complications. However, it is difficult for community-based non-experts to obtain sufficient diagnostic performance. Artificial intelligence-based systems have been developed to analyze endoscopic images; they identify neoplasms with high accuracy and low interobserver variation. We performed a multi-center study to determine the diagnostic accuracy of EndoBRAIN, an artificial intelligence-based system that analyzes cell nuclei, crypt structure, and microvessels in endoscopic images, in identification of colon neoplasms.METHODS: The EndoBRAIN system was initially trained using 69,142 endocytoscopic images, taken at 520-fold magnification, from patients with colorectal polyps who underwent endoscopy at 5 academic centers in Japan from October 2017 through March 2018. We performed a retrospective comparative analysis of the diagnostic performance of EndoBRAIN vs that of 30 endoscopists (20 trainees and 10 experts); the endoscopists assessed images from 100 cases produced via white-light microscopy, endocytoscopy with methylene blue staining, and endocytoscopy with narrow-band imaging. EndoBRAIN was used to assess endocytoscopic, but not white-light, images. The primary outcome was the accuracy of EndoBrain in distinguishing neoplasms from non-neoplasms, compared with that of endoscopists, using findings from pathology analysis as the reference standard.RESULTS: In analysis of stained endocytoscopic images, EndoBRAIN identified colon lesions with 96.9% sensitivity (95% CI, 95.8%-97.8%), 100% specificity (95% CI, 99.6%-100%), 98% accuracy (95% CI, 97.3%-98.6%), a 100% positive-predictive value (95% CI, 99.8%-100%), and a 94.6% negative-predictive (95% CI, 92.7%-96.1%); these values were all significantly greater than those of the endoscopy trainees and experts. In analysis of narrow-band images, EndoBRAIN distinguished neoplastic from non-neoplastic lesions with 96.9% sensitivity (95% CI, 95.8-97.8), 94.3% specificity (95% CI, 92.3-95.9), 96.0% accuracy (95% CI, 95.1-96.8), a 96.9% positive-predictive value, (95% CI, 95.8-97.8), and a 94.3% negative-predictive value (95% CI, 92.3-95.9); these values were all significantly higher than those of the endoscopy trainees, sensitivity and negative-predictive value were significantly higher but the other values are comparable to those of the experts.CONCLUSIONS: EndoBRAIN accurately differentiated neoplastic from non-neoplastic lesions in stained endocytoscopic images and endocytoscopic narrow-band images, when pathology findings were used as the standard. This technology has been authorized for clinical use by the Japanese regulatory agency and should be used in endoscopic evaluation of small polyps more widespread clinical settings.

    DOI: 10.1016/j.cgh.2019.09.009

    Web of Science

    researchmap

  26. PREDICTION OF LYMPH NODE METASTASIS IN T2 COLORECTAL CANCER BASED ON ARTIFICIAL INTELLIGENCE -PROPOSAL OF AN INDICATION FOR FUTURE FULL-THICKNESS ENDOSCOPIC RESECTION- Reviewed

    Katsuro Ichimasa, Shinei Kudo, Villard Benjamin, Kenta Nakahara, Yuichi Mori, Masashi Misawa, Maeda Yasuharu, Naoya Toyoshima, Noriyuki Ogata, Toyoki Kudo, Takemasa Hayashi, Kunihiko Wakamura, Hideyuki Miyachi, Naruhiko Sawada, Hayato Itoh, Masahiro Oda, Kensaku Mori, Fumio Ishida

    GASTROINTESTINAL ENDOSCOPY   Vol. 91 ( 6 ) page: AB43 - AB43   2020.6

     More details

    Language:English   Publisher:MOSBY-ELSEVIER  

    Web of Science

    researchmap

  27. USE OF ARTIFICIAL INTELLIGENCE TO PREVENT SEVERE PERFORATION DURING ENDOSCOPIC SUBMUCOSAL DISSECTION FOR COLORECTAL NEOPLASM: A PROOF-OF-CONCEPT STUDY Reviewed

    Taishi Okumura, Shinei Kudo, Takemasa Hayashi, Yuichi Mori, Masashi Misawa, Masahiro Abe, Yuta Sato, Shinichi Kataoka, Yuta Kouyama, Tatsuya Sakurai, Maeda Yasuharu, Yushi Ogawa, Katsuro Ichimasa, Hiroki Nakamura, Tomoyuki Ishigaki, Naoya Toyoshima, Noriyuki Ogata, Toyoki Kudo, Tomokazu Hisayuki, Kunihiko Wakamura, Hideyuki Miyachi, Toshiyuki Baba, Fumio Ishida, Hitomi Oishi, Hayato Itoh, Kensaku Mori

    GASTROINTESTINAL ENDOSCOPY   Vol. 91 ( 6 ) page: AB247 - AB248   2020.6

     More details

    Language:English   Publisher:MOSBY-ELSEVIER  

    Web of Science

    researchmap

  28. How Far Will Clinical Application of AI Applications Advance for Colorectal Cancer Diagnosis? Reviewed

    Yuichi Mori, Shin-ei Kudo, Masashi Misawa, Kenichi Takeda, Toyoki Kudo, Hayato Itoh, Masahiro Oda, Kensaku Mori

    JOURNAL OF THE ANUS RECTUM AND COLON   Vol. 4 ( 2 ) page: 47 - 50   2020.4

     More details

    Language:English   Publisher:JAPAN SOC COLOPROCTOLOGY  

    Integrating artificial intelligence (AI) applications into colonoscopy practice is being accelerated as deep learning technologies emerge. In this field, most of the preceding research has focused on polyp detection and characterization, which can mitigate inherent human errors accompanying colonoscopy procedures. On the other hand, more challenging research areas are currently capturing attention: the automated prediction of invasive cancers. Colorectal cancers (CRCs) harbor potential lymph node metastasis when they invade deeply into submucosal layers, which should be resected surgically rather than endoscopically. However, pretreatment discrimination of deeply invasive submucosal CRCs is considered difficult, according to previous prospective studies (e.g., <70% sensitivity), leading to an increased number of unnecessary surgeries for large adenomas or slightly invasive submucosal CRCs. AI is now expected to overcome this challenging hurdle because it is considered to provide better performance in predicting invasive cancer than non-expert endoscopists. In this review, we introduce five relevant publications in this area. Unfortunately, progress in this research area is in a very preliminary phase, compared to that of automated polyp detection and characterization, because of the lack of number of invasive CRCs used for machine learning. However, this issue will be overcome with more target images and cases. The research field of AI for invasive CRCs is just starting but could be a game changer of patient care in the near future, given rapidly growing technologies, and research will gradually increase.

    DOI: 10.23922/jarc.2019-045

    Web of Science

    researchmap

  29. Visualising decision-reasoning regions in computer-aided pathological pattern diagnosis of endoscytoscopic images based on CNN weights analysis Reviewed

    Hayato Itoh, Zhongyang Lu, Yuichi Mori, Masashi Misawa, Masahiro Oda, Shin-ei Kudo, Kensaku Mori

    MEDICAL IMAGING 2020: COMPUTER-AIDED DIAGNOSIS   Vol. 11314   2020

     More details

    Authorship:Lead author   Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:SPIE-INT SOC OPTICAL ENGINEERING  

    Purpose of this paper is to present a method for visualising decision-reasoning regions in computer-aided pathological pattern diagnosis of endocytoscopic images. Endocytoscope enables us to perform direct observation of cells and their nuclei on the colon wall at maximum 500-times ultramagnification. For this new modality, computer-aided pathological diagnosis system is strongly required for the support of non-expert physicians. To develop a CAD system, we adopt convolutional neural network (CNN) as the classifier of endocytoscopic images. In addition to this classification function, based on CNN weights analysis, we develop a filter function that visualises decision-reasoning regions on classified images. This visualisation function helps novice endocytoscopists to develop their understanding of pathological pattern on endocytoscopic images for accurate endocytoscopic diagnosis. In numerical experiment, our CNN model achieved 90 % classification accuracy. Furthermore, experimental results show that decision-reasoning regions suggested by our filter function contain characteristic pit patterns in real endocytoscopic diagnosis.

    DOI: 10.1117/12.2549532

    Web of Science

    researchmap

  30. Visualizing intestines for diagnostic assistance of ileus based on intestinal region segmentation from 3D CT images Reviewed

    Hirohisa Oda, Kohei Nishio, Takayuki Kitasaka, Hizuru Amano, Aitaro Takimoto, Hiroo Uchida, Kojiro Suzuki, Hayato Itoh, Masahiro Oda, Kensaku Mori

    MEDICAL IMAGING 2020: COMPUTER-AIDED DIAGNOSIS   Vol. 11314   2020

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:SPIE-INT SOC OPTICAL ENGINEERING  

    This paper presents a visualization method of intestine (the small and large intestines) regions and their stenosed parts caused by ileus from CT volumes. Since it is difficult for non-expert clinicians to find stenosed parts, the intestine and its stenosed parts should be visualized intuitively. Furthermore, the intestine regions of ileus cases are quite hard to be segmented. The proposed method segments intestine regions by 3D FCN (3D U-Net). Intestine regions are quite difficult to be segmented in ileus cases since the inside the intestine is filled with fluids. These fluids have similar intensities with intestinal wall on 3D CT volumes. We segment the intestine regions by using 3D U-Net trained by a weak annotation approach. Weak-annotation makes possible to train the 3D U-Net with small manually-traced label images of the intestine. This avoids us to prepare many annotation labels of the intestine that has long and winding shape. Each intestine segment is volume-rendered and colored based on the distance from its endpoint in volume rendering. Stenosed parts (disjoint points of an intestine segment) can be easily identified on such visualization. In the experiments, we showed that stenosed parts were intuitively visualized as endpoints of segmented regions, which are colored by red or blue.

    DOI: 10.1117/12.2548910

    Web of Science

    researchmap

    Other Link: https://dblp.uni-trier.de/db/journals/corr/corr2003.html#abs-2003-01290

  31. Stable polyp-scene classification via subsampling and residual learning from an imbalanced large dataset. Reviewed International journal

    Hayato Itoh, Holger Roth, Masahiro Oda, Masashi Misawa, Yuichi Mori, Shin-Ei Kudo, Kensaku Mori

    Healthcare technology letters   Vol. 6 ( 6 ) page: 237 - 242   2019.12

     More details

    Authorship:Lead author   Language:English   Publishing type:Research paper (scientific journal)  

    This Letter presents a stable polyp-scene classification method with low false positive (FP) detection. Precise automated polyp detection during colonoscopies is essential for preventing colon-cancer deaths. There is, therefore, a demand for a computer-assisted diagnosis (CAD) system for colonoscopies to assist colonoscopists. A high-performance CAD system with spatiotemporal feature extraction via a three-dimensional convolutional neural network (3D CNN) with a limited dataset achieved about 80% detection accuracy in actual colonoscopic videos. Consequently, further improvement of a 3D CNN with larger training data is feasible. However, the ratio between polyp and non-polyp scenes is quite imbalanced in a large colonoscopic video dataset. This imbalance leads to unstable polyp detection. To circumvent this, the authors propose an efficient and balanced learning technique for deep residual learning. The authors' method randomly selects a subset of non-polyp scenes whose number is the same number of still images of polyp scenes at the beginning of each epoch of learning. Furthermore, they introduce post-processing for stable polyp-scene classification. This post-processing reduces the FPs that occur in the practical application of polyp-scene classification. They evaluate several residual networks with a large polyp-detection dataset consisting of 1027 colonoscopic videos. In the scene-level evaluation, their proposed method achieves stable polyp-scene classification with 0.86 sensitivity and 0.97 specificity.

    DOI: 10.1049/htl.2019.0079

    PubMed

    researchmap

  32. Artificial intelligence and colonoscopy: Current status and future perspectives Reviewed

    Shin-ei Kudo, Yuichi Mori, Masashi Misawa, Kenichi Takeda, Toyoki Kudo, Hayato Itoh, Masahiro Oda, Kensaku Mori

    DIGESTIVE ENDOSCOPY   Vol. 31 ( 4 ) page: 363 - 371   2019.7

     More details

    Language:English   Publisher:WILEY  

    Background and Aim Application of artificial intelligence in medicine is now attracting substantial attention. In the field of gastrointestinal endoscopy, computer-aided diagnosis (CAD) for colonoscopy is the most investigated area, although it is still in the preclinical phase. Because colonoscopy is carried out by humans, it is inherently an imperfect procedure. CAD assistance is expected to improve its quality regarding automated polyp detection and characterization (i.e. predicting the polyp's pathology). It could help prevent endoscopists from missing polyps as well as provide a precise optical diagnosis for those detected. Ultimately, these functions that CAD provides could produce a higher adenoma detection rate and reduce the cost of polypectomy for hyperplastic polyps. Methods and Results Currently, research on automated polyp detection has been limited to experimental assessments using an algorithm based on ex vivo videos or static images. Performance for clinical use was reported to have >90% sensitivity with acceptable specificity. In contrast, research on automated polyp characterization seems to surpass that for polyp detection. Prospective studies of in vivo use of artificial intelligence technologies have been reported by several groups, some of which showed a >90% negative predictive value for differentiating diminutive (<= 5 mm) rectosigmoid adenomas, which exceeded the threshold for optical biopsy. Conclusion We introduce the potential of using CAD for colonoscopy and describe the most recent conditions for regulatory approval for artificial intelligence-assisted medical devices.

    DOI: 10.1111/den.13340

    Web of Science

    researchmap

  33. Artificial intelligence and upper gastrointestinal endoscopy: Current status and future perspective Reviewed

    Yuichi Mori, Shin-ei Kudo, Hussein E. N. Mohmed, Masashi Misawa, Noriyuki Ogata, Hayato Itoh, Masahiro Oda, Kensaku Mori

    DIGESTIVE ENDOSCOPY   Vol. 31 ( 4 ) page: 378 - 388   2019.7

     More details

    Language:English   Publisher:WILEY  

    With recent breakthroughs in artificial intelligence, computer-aided diagnosis (CAD) for upper gastrointestinal endoscopy is gaining increasing attention. Main research focuses in this field include automated identification of dysplasia in Barrett's esophagus and detection of early gastric cancers. By helping endoscopists avoid missing and mischaracterizing neoplastic change in both the esophagus and the stomach, these technologies potentially contribute to solving current limitations of gastroscopy. Currently, optical diagnosis of early-stage dysplasia related to Barrett's esophagus can be precisely achieved only by endoscopists proficient in advanced endoscopic imaging, and the false-negative rate for detecting gastric cancer is approximately 10%. Ideally, these novel technologies should work during real-time gastroscopy to provide on-site decision support for endoscopists regardless of their skill; however, previous studies of these topics remain ex vivo and experimental in design. Therefore, the feasibility, effectiveness, and safety of CAD for upper gastrointestinal endoscopy in clinical practice remain unknown, although a considerable number of pilot studies have been conducted by both engineers and medical doctors with excellent results. This review summarizes current publications relating to CAD for upper gastrointestinal endoscopy from the perspective of endoscopists and aims to indicate what is required for future research and implementation in clinical practice.

    DOI: 10.1111/den.13317

    Web of Science

    researchmap

  34. ARTIFICIAL INTELLIGENCE-ASSISTED POLYP DETECTION SYSTEM FOR COLONOSCOPY, BASED ON THE LARGEST AVAILABLE COLLECTION OF CLINICAL VIDEO DATA FOR MACHINE LEARNING

    Masashi Misawa, Shinei Kudo, Yuichi Mori, Tomonari Cho, Shinichi Kataoka, Yasuharu Maeda, Yushi Ogawa, Kenichi Takeda, Hiroki Nakamura, Katsuro Ichimasa, Naoya Toyoshima, Noriyuki Ogata, Toyoki Kudo, Tomokazu Hisayuki, Takemasa Hayashi, Kunihiko Wakamura, Toshiyuki Baba, Fumio Ishida, Hayato Itoh, Masahiro Oda, Kensaku Mori

    GASTROINTESTINAL ENDOSCOPY   Vol. 89 ( 6 ) page: AB646 - AB647   2019.6

     More details

    Language:English   Publisher:MOSBY-ELSEVIER  

    DOI: 10.1016/j.gie.2019.03.1134

    Web of Science

    researchmap

  35. PERFORMANCE OF NON-EXPERT ENDOSCOPISTS IN OPTICAL BIOPSY OF DIMINUTIVE COLORECTAL POLYPS WITH REAL-TIME USE OF ARTIFICIAL INTELLIGENCE Reviewed

    Yuichi Mori, Shinei Kudo, Masashi Misawa, Shinichi Kataoka, Kenichi Takeda, Kenichi Suzuki, Katsuro Ichimasa, Yushi Ogawa, Yasuharu Maeda, Takemasa Hayashi, Kunihiko Wakamura, Toyoki Kudo, Fumio Ishida, Haruhiro Inoue, Hayato Itoh, Masahiro Oda, Kensaku Mori

    GASTROINTESTINAL ENDOSCOPY   Vol. 89 ( 6 ) page: AB89 - AB89   2019.6

     More details

    Language:English   Publisher:MOSBY-ELSEVIER  

    DOI: 10.1016/j.gie.2019.04.075

    Web of Science

    researchmap

  36. COMPUTER-AIDED DIAGNOSIS FOR SAMLL COLORECTAL LESIONS: A MULTI-CENTER VALIDATION "ENDOBRAIN STUDY" DESIGNED TO OBTAIN REGULATORY APPROVAL

    Kinichi Hotta, Shinei Kudo, Yuichi Mori, Hiroaki Ikematsu, Yutaka Saito, Kazuo Ohtsuka, Masashi Misawa, Hayato Itoh, Masahiro Oda, Kensaku Mori

    GASTROINTESTINAL ENDOSCOPY   Vol. 89 ( 6 ) page: AB76 - AB76   2019.6

     More details

    Language:English   Publisher:MOSBY-ELSEVIER  

    DOI: 10.1016/j.gie.2019.04.051

    Web of Science

    researchmap

  37. Investigation of extracting interlobular septa with Hessian analysis and Radial Structure Tensor combined with roundness error in micro-CT volume

    Xiaotian Zhao, Hirohisa Oda, Shota Nakamura, Yuichiro Hayashi, Hayato Itoh, Masahiro Oda, Kensaku Mori

    INTERNATIONAL FORUM ON MEDICAL IMAGING IN ASIA 2019   Vol. 11050   2019

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:SPIE-INT SOC OPTICAL ENGINEERING  

    Micro-CT is a nondestructive scanning device that is capable of capturing three dimensional structures at mu m level. With the spread of this device uses in medical fields, it is expected that this device may bring further understanding of the human anatomy by analyzing three-dimensional micro structure from volume of in vivo specimens captured by micro-CT. In the topic of micro structure analysis of lung, the methods for extracting surface structures including the interlobular septa and the visceral pleura were not commonly studied. In this paper, we introduce a method to extract sheet structure such as the interlobular septa and the visceral pleura from micro-CT volumes. The proposed method consists of two steps: Hessian analysis based method for sheet structure extraction and Radial Structure Tensor combined with roundness evaluation for hollow-tube structure extraction. We adopted the proposed method on complex phantom data and a medical lung micro-CT volume. We confirmed the extraction of the interlobular septa from medical volume from experiments.

    DOI: 10.1117/12.2521646

    Web of Science

    researchmap

  38. Polyp-Size Classification with RGB-D features for Colonoscopy Reviewed

    Hayato Itoh, Holger R. Roth, Yuichi Mori, Masashi Misawa, Masahiro Oda, Shin-ei Kudo, Kensaku Mori

    MEDICAL IMAGING 2019: COMPUTER-AIDED DIAGNOSIS   Vol. 10950   page: 1095015 - 1095015   2019

     More details

    Authorship:Lead author   Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:SPIE-INT SOC OPTICAL ENGINEERING  

    Measurement of a polyp size is an essential task in colon cancer screening, since the polyp-size information has critical roles for decision on colonoscopy. However, an estimation of a polyp size from a single view of colonoscope without a measurement device is quite difficult even for expert physicians. To overcome this difficulty, automated size estimation techniques would be desirable for clinical scenes. This paper presents polyp-size classification method with a single colonoscopic image for colonoscopy. Our proposed method estimates depth information from a single colonoscopic image with trained model and utilises the estimated information for the classification. In our method, the model for depth information is obtained by deep learning with colonoscopic videos. Experimental results show the achievement of binary and trinary polyp-size classification with 79% and 74% accuracy from a single still image of a colonoscopic movie.

    DOI: 10.1117/12.2513093

    Web of Science

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/micad/micad2019.html#ItohRMMOKM19

  39. Spaciousness Filters for Non-contrast CT Volume Segmentation of the Intestine Region for Emergency Ileus Diagnosis Reviewed

    Hirohisa Oda, Kohei Nishio, Takayuki Kitasaka, Benjamin Villard, Hizuru Amano, Kosuke Chiba, Akinari Hinoki, Hiroo Uchida, Kojiro Suzuki, Hayato Itoh, Masahiro Oda, Kensaku Mori

    UNCERTAINTY FOR SAFE UTILIZATION OF MACHINE LEARNING IN MEDICAL IMAGING AND CLINICAL IMAGE-BASED PROCEDURES   Vol. 11840   page: 104 - 114   2019

     More details

    Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:SPRINGER INTERNATIONAL PUBLISHING AG  

    This paper proposes enhancement filters for shape-specific regions, based on radial structure tensor (RST) analysis, which we name "spaciousness filters". RST analysis can be used in a similar way to Hessian analysis for classifying intensity structures. However, RST is insufficient for enhancing regions having little contrast or non-typical morphology. Our proposed filters enhance such regions by extending the ray search scheme of RST analysis to work as a filter evaluating spaciousness. We show applications to the abdominal CT of ileus patients having specific shapes. The intestines (including small intestines) of those patients consist of air, liquid and feces portions, and are not contrast-enhanced by barium. Enhancement of liquid and walls play key roles in the sufficient segmentation of intestines and division between neighboring regions. Experimental results on 7 clinical cases showed that the proposed intestine segmentation method produced higher Dice score (0.68) than traditional RST analysis (0.44), even without specific refinement processes like machine-learning-based false positive reduction.

    DOI: 10.1007/978-3-030-32689-0_11

    Web of Science

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/miccai/clip2019.html#OdaNKVACHUSIOM19

  40. Tubular Structure Segmentation Using Spatial Fully Connected Network with Radial Distance Loss for 3D Medical Images. Reviewed

    Chenglong Wang, Yuichiro Hayashi, Masahiro Oda, Hayato Itoh, Takayuki Kitasaka, Alejandro F. Frangi, Kensaku Mori

        page: 348 - 356   2019

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-030-32226-7_39

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/miccai/miccai2019-6.html#WangHOIKFM19

  41. Real-Time Use of Artificial Intelligence in Identification of Diminutive Polyps During Colonoscopy A Prospective Study Reviewed

    Yuichi Mori, Shin-ei Kudo, Masashi Misawa, Yutaka Saito, Hiroaki Ikematsu, Kinichi Hotta, Kazuo Ohtsuka, Fumihiko Urushibara, Shinichi Kataoka, Yushi Ogawa, Yasuharu Maeda, Kenichi Takeda, Hiroki Nakamura, Katsuro Ichimasa, Toyoki Kudo, Takemasa Hayashi, Kunihiko Wakamura, Fumio Ishida, Haruhiro Inoue, Hayato Itoh, Masahiro Oda, Kensaku Mori

    ANNALS OF INTERNAL MEDICINE   Vol. 169 ( 6 ) page: 357 - +   2018.9

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:AMER COLL PHYSICIANS  

    Background: Computer-aided diagnosis (CAD) for colonoscopy may help endoscopists distinguish neoplastic polyps (adenomas) requiring resection from nonneoplastic polyps not requiring resection, potentially reducing cost.Objective: To evaluate the performance of real-time CAD with endocytoscopes (x520 ultramagnifying colonoscopes providing microvascular and cellular visualization of colorectal polyps after application of the narrow-band imaging [NBI] and methylene blue staining modes, respectively).Design: Single-group, open-label, prospective study. (UMIN [University hospital Medical Information Network] Clinical Trial Registry: UMIN000027360).Setting: University hospital.Participants: 791 consecutive patients undergoing colonoscopy and 23 endoscopists.Intervention: Real-time use of CAD during colonoscopy.Measurements: CAD-predicted pathology (neoplastic or nonneoplastic) of detected diminutive polyps (5 mm) on the basis of real-time outputs compared with pathologic diagnosis of the resected specimen (gold standard). The primary end point was whether CAD with the stained mode produced a negative predictive value (NPV) of 90% or greater for identifying diminutive rectosigmoid adenomas, the threshold required to "diagnose-and-leave" nonneoplastic polyps. Best- and worst-case scenarios assumed that polyps lacking either CAD diagnosis or pathology were true- or false-positive or true- or false-negative, respectively.Results: Overall, 466 diminutive (including 250 rectosigmoid) polyps from 325 patients were assessed by CAD, with a pathologic prediction rate of 98.1% (457 of 466). The NPVs of CAD for diminutive rectosigmoid adenomas were 96.4% (95% CI, 91.8% to 98.8%) (best-case scenario) and 93.7% (CI, 88.3% to 97.1%) (worst-case scenario) with stained mode and 96.5% (CI, 92.1% to 98.9%) (best-case scenario) and 95.2% (CI, 90.3% to 98.0%) (worst-case scenario) with NBI.Limitation: Two thirds of the colonoscopies were conducted by experts who had each experienced more than 200 endocytoscopies; 186 polyps not assessed by CAD were excluded.Conclusion: Real-time CAD can achieve the performance level required for a diagnose-and-leave strategy for diminutive, nonneoplastic rectosigmoid polyps.

    DOI: 10.7326/M18-0249

    Web of Science

    researchmap

  42. Artificial Intelligence-Assisted Polyp Detection for Colonoscopy: Initial Experience Reviewed

    Masashi Misawa, Shin-ei Kudo, Yuichi Mori, Tomonari Cho, Shinichi Kataoka, Akihiro Yamauchi, Yushi Ogawa, Yasuharu Maeda, Kenichi Takeda, Katsuro Ichimasa, Hiroki Nakamura, Yusuke Yagawa, Naoya Toyoshima, Noriyuki Ogata, Toyoki Kudo, Tomokazu Hisayuki, Takemasa Hayashi, Kunihiko Wakamura, Toshiyuki Baba, Fumio Ishida, Hayato Itoh, Holger Roth, Masahiro Oda, Kensaku Mori

    GASTROENTEROLOGY   Vol. 154 ( 8 ) page: 2027 - +   2018.6

     More details

    Language:English   Publisher:W B SAUNDERS CO-ELSEVIER INC  

    DOI: 10.1053/j.gastro.2018.04.003

    Web of Science

    researchmap

  43. Application of Directional Statistics to Classification of Three-Channel Colour Images. Reviewed

    Kaori Tanji, Hayato Itoh, Atsushi Imiya

        page: 60 - 69   2018

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.3233/978-1-61499-929-4-60

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/appis/appis2018.html#TanjiII18

  44. Discrimination of Volumetric Shapes Using Orthogonal Tensor Decomposition. Reviewed

    Hayato Itoh, Atsushi Imiya

        page: 277 - 290   2018

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-030-04747-4_26

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/miccai/shapemi2018.html#ItohI18

  45. Cascade classification of endocytoscopic images of colorectal lesions for automated pathological diagnosis Reviewed

    Hayato Itoh, Yuichi Mori, Masashi Misawa, Masahiro Oda, Shin-ei Kudo, Kensaku Mori

    MEDICAL IMAGING 2018: COMPUTER-AIDED DIAGNOSIS   Vol. 10575   page: 1057516 - 1057516   2018

     More details

    Authorship:Lead author   Language:English   Publishing type:Research paper (international conference proceedings)   Publisher:SPIE-INT SOC OPTICAL ENGINEERING  

    This paper presents a new classification method for endocytoscopic images. Endocytoscopy is a new endoscope that enables us to perform conventional endoscopic observation and ultramagnified observation of cell level. This ultramagnified views (endocytoscopic images) make possible to perform pathological diagnosis only on endoscopic views of polyps during colonoscopy. However, endocytoscopic image diagnosis requires higher experiences for physicians. An automated pathological diagnosis system is required to prevent the overlooking of neoplastic lesions in endocytoscopy. For this purpose, we propose a new automated endocytoscopic image classification method that classifies neoplastic and non-neoplastic endocytoscopic images. This method consists of two classification steps. At the first step, we classify an input image by support vector machine. We forward the image to the second step if the confidence of the first classification is low. At the second step, we classify the forwarded image by convolutional neural network. We reject the input image if the confidence of the second classification is also low. We experimentally evaluate the classification performance of the proposed method. In this experiment, we use about 16,000 and 4,000 colorectal endocytoscopic images as training and test data, respectively. The results show that the proposed method achieves high sensitivity 93.4% with small rejection rate 9.3% even for difficult test data.

    DOI: 10.1117/12.2293495

    Web of Science

    researchmap

    Other Link: https://dblp.uni-trier.de/conf/micad/2018

  46. Variational Method for Multiresolution Image Registration. Reviewed

    Kento Hosoya, Ryo Sasaki, Kaori Tanji, Hayato Itoh, Atsushi Imiya

        page: 157 - 168   2018

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.3233/978-1-61499-929-4-157

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/appis/appis2018.html#HosoyaSTII18

  47. Analysis of Multilinear Subspaces Based on Geodesic Distance. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 384 - 396   2017

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-64689-3_31

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/caip/caip2017-1.html#ItohIS17a

  48. Multilinear Methods for Spatio-Temporal Image Recognition. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 148 - 159   2017

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-64689-3_12

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/caip/caip2017-1.html#ItohIS17

  49. Motion Language of Stereo Image Sequence. Reviewed

    Tomoya Kato, Hayato Itoh, Atsushi Imiya

        page: 1211 - 1218   2017

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1109/CVPRW.2017.160

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/cvpr/cvprw2017.html#KatoII17

  50. Linear Data Compression of Hyperspectral Images. Reviewed

    Kaori Tanji, Atsushi Imiya, Hayato Itoh, Hiroaki Kuze, Naohiro Manago

        page: 3001 - 3007   2017

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1109/ICCVW.2017.354

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/iccvw/iccvw2017.html#TanjiIIKM17

  51. Fast Approximate Karhunen-Loève Transform for Three-Way Array Data. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 1827 - 1834   2017

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1109/ICCVW.2017.216

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/iccvw/iccvw2017.html#ItohIS17

  52. Approximation of N-Way Principal Component Analysis for Organ Data. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 16 - 31   2016

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-54526-4_2

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/accv/accv2016-w3.html#ItohIS16

  53. Classification of Volumetric Data Using Multiway Data Analysis. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 231 - 240   2016

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-49055-7_21

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/sspr/sspr2016.html#ItohIS16a

  54. Volumetric Image Pattern Recognition Using Three-Way Principal Component Analysis. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 103 - 117   2016

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-51237-2_9

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/miccai/sesami2016.html#ItohIS16

  55. Discriminative Properties in Directional Distributions for Image Pattern Recognition. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 617 - 630   2015

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-29451-3_49

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/psivt/psivt2015.html#ItohIS15

  56. Optical Flow Computation with Locally Quadratic Assumption. Reviewed

    Tomoya Kato, Hayato Itoh, Atsushi Imiya

        page: 223 - 234   2015

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-23192-1_19

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/caip/caip2015-1.html#KatoII15

  57. Low-Dimensional Tensor Principle Component Analysis. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 715 - 726   2015

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-23192-1_60

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/caip/caip2015-1.html#ItohIS15

  58. Simultaneous Frame-rate Up-conversion of Image and Optical Flow Sequences. Reviewed

    Shun Inagaki, Hayato Itoh, Atsushi Imiya

        page: 68 - 75   2015

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.5220/0005296800680075

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/visapp/visapp2015-1.html#InagakiII15

  59. Variational Multiple Warping for Cardiac Image Analysis. Reviewed

    Shun Inagaki, Hayato Itoh, Atsushi Imiya

        page: 749 - 759   2015

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-23117-4_64

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/caip/caip2015-2.html#InagakiII15

  60. Multiple Alignment of Spatiotemporal Deformable Objects for the Average-Organ Computation. Reviewed

    Shun Inagaki, Hayato Itoh, Atsushi Imiya

        page: 353 - 366   2014

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-16220-1_25

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/eccv/eccv2014w4.html#InagakiII14

  61. Global Volumetric Image Registration Using Local Linear Property of Image Manifold. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 238 - 253   2014

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-319-16628-5_18

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/accv/accv2014-w1.html#ItohIS14

  62. Two-Dimensional Global Image Registration Using Local Linear Property of Image Manifold. Reviewed

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

        page: 3862 - 3867   2014

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1109/ICPR.2014.663

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/icpr/icpr2014.html#ItohIS14

  63. Dimension Reduction Methods for Image Pattern Recognition. Reviewed

    Hayato Itoh, Tomoya Sakai 0002, Kazuhiko Kawamoto, Atsushi Imiya

        page: 26 - 42   2013

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-642-39140-8_2

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/simbad/simbad2013.html#ItohSKI13

  64. Local Affine Optical Flow Computation. Reviewed

    Hayato Itoh, Shun Inagaki, Ming-Ying Fan, Atsushi Imiya, Kazuhiko Kawamoto, Tomoya Sakai 0002

        page: 203 - 215   2013

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-642-53926-8_19

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/psivt/psivt2013w.html#ItohIFIKS13

  65. Global Image Registration Using Random Projection and Local Linear Method. Reviewed

    Hayato Itoh, Tomoya Sakai 0002, Kazuhiko Kawamoto, Atsushi Imiya

        page: 564 - 571   2013

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-642-40261-6_68

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/caip/caip2013-1.html#ItohSKI13

  66. Edge Detection and Smoothing-Filter of Volumetric Data. Reviewed

    Masaki Narita, Atsushi Imiya, Hayato Itoh

        page: 489 - 498   2012

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-642-33191-6_48

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/isvc/isvc2012-2.html#NaritaII12

  67. Interpolation of Reference Images in Sparse Dictionary for Global Image Registration. Reviewed

    Hayato Itoh, Shuang Lu, Tomoya Sakai 0002, Atsushi Imiya

        page: 657 - 667   2012

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-642-33191-6_65

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/isvc/isvc2012-2.html#ItohLSI12

  68. Bifurcation of Segment Edge Curves in Scale Space. Reviewed

    Tomoya Sakai 0002, Haruhiko Nishiguchi, Hayato Itoh, Atsushi Imiya

        page: 302 - 313   2011

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-642-24785-9_26

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/scalespace/ssvm2011.html#SakaiNII11

  69. Global Image Registration by Fast Random Projection. Reviewed

    Hayato Itoh, Shuang Lu, Tomoya Sakai 0002, Atsushi Imiya

        page: 23 - 32   2011

     More details

    Authorship:Lead author   Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-642-24028-7_3

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/isvc/isvc2011-1.html#ItohLSI11

  70. Multi-label Classification for Image Annotation via Sparse Similarity Voting. Reviewed

    Tomoya Sakai 0002, Hayato Itoh, Atsushi Imiya

        page: 344 - 353   2010

     More details

    Publishing type:Research paper (international conference proceedings)  

    DOI: 10.1007/978-3-642-22819-3_35

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/accv/accv2010-w2.html#SakaiII10

▼display all

MISC 66

  1. Report on MICCAI 2019

    小田昌宏, 伊東隼人, 宮内翔子, 諸岡健一, 松崎博貴, 花岡昇平, 古川亮, 増谷佳孝, 森健策, 森健策

    電子情報通信学会技術研究報告   Vol. 119 ( 399(MI2019 65-123)(Web) )   2020

  2. Preliminary Study on Handling Different Data Distributions in Multi-Institute Datasets for Automated Endocytoscopic Image Classification

    伊東隼人, 森悠一, 三澤将史, 小田昌宏, 工藤進英, 森健策, 森健策, 森健策

    日本コンピュータ外科学会誌   Vol. 21 ( 4 ) page: 332 - 333   2019.11

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  3. Report on MICCAI 2018

    小田昌宏, 大竹義人, 伊東隼人, 杉野貴明, 斉藤篤, 古川亮, 大西峻, 井宮淳, 森健策, 森健策

    電子情報通信学会技術研究報告   Vol. 118 ( 412(MI2018 59-115)(Web) )   2019

  4. 深層学習における学習データセット規模拡大に応じた分類精度向上に関する実験的検討-超拡大大腸内視鏡画像における腫瘍性病変分類に向けた特徴量抽出-

    伊東隼人, 森悠一, 三澤将史, 小田昌宏, 工藤進英, 森健策, 森健策, 森健策

    日本医用画像工学会大会予稿集(CD-ROM)   Vol. 38th   2019

  5. Training-Dataset Construction Method from Imbalanced Dataset-Towards Medical Image Classification based on Machine Learning Techniques-

    伊東隼人, 森悠一, 三澤将史, 小田昌宏, 工藤進英, 森健策, 森健策, 森健策

    日本コンピュータ外科学会誌   Vol. 20 ( 4 ) page: 261 - 262   2018.10

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  6. 教師なし深度推定を利用したRGB-D特徴抽出に基づくポリープのトリナリサイズ推定

    伊東 隼人, Roth Holger, 三澤 将史, 森 悠一, 小田 昌宏, 工藤 進英, 森 健策

    日本医用画像工学会大会予稿集   Vol. 37回   page: 432 - 435   2018.7

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    大腸内視鏡検査におけるポリープの大きさ情報(サイズ情報)は診断において重要な役割りを果す。5mm以下の小さなポリープ、10mm以上の大きなポリープ、それ以外のポリープといった3カテゴリに関するサイズ推定(トリナリ推定)は治療計画を立てる上で特に重要とされる。しかし内視鏡画像のみに基くサイズ推定は熟練の専門医にとっても難しい。そこで内視鏡画像ならびに動画像からポリープのサイズ推定を行うCADシステムが望まれている。内視鏡画像には三次元情報が欠如しており、二次元画像からの精密なサイズ推定は不良設定問題である。本研究ではこの不良設定問題をトリナリサイズ推定へと緩和し、内視鏡画像からポリープのサイズ推定に有用な特徴を深層学習によって抽出、そして分類する手法を提案する。特徴抽出においては教師無し学習による深度推定を利用する。提案手法により高精度のトリナリサイズ推定を達成した。(著者抄録)

    J-GLOBAL

    researchmap

  7. Feature-selection method based on Grassmann distance for the classification of neoplastic polyps on endocytoscopic images (医用画像)

    伊東 隼人, 森 悠一, 三澤 将史, 小田 昌宏, 工藤 進英, 森 健策

    電子情報通信学会技術研究報告 = IEICE technical report : 信学技報   Vol. 117 ( 518 ) page: 51 - 56   2018.3

     More details

    Language:English   Publisher:電子情報通信学会  

    CiNii Books

    researchmap

  8. Classification of neoplasia and non-neoplasia for colon endocytoscopic images by convolutional neural network

    伊東隼人, 森悠一, 三澤将史, 小田昌宏, 工藤進英, 森健策, 森健策

    電子情報通信学会技術研究報告   Vol. 117 ( 220(MI2017 37-46) )   2017

  9. Automated Pathological Diagnosis Using Colon Endocytoscopic Images~Analysis of Classification Accuracy of Neoplastic Lesions~

    伊東隼人, 森悠一, 三澤将史, 小田昌宏, 工藤進英, 森健策, 森健策

    日本コンピュータ外科学会誌   Vol. 19 ( 4 )   2017

  10. Tensor-Based Methods for Dimension Reduction of Volumetric Data

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告   Vol. 115 ( 517(PRMU2015 164-197) )   2016

  11. Second-Order Tensor Principal Component Analysis Meets Two-Dimensional Singular Value Decomposition

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告   Vol. 115 ( 24(PRMU2015 1-31) )   2015

  12. Eigenfunctions in Linear Scale Space

    IMIYA Atsushi, ITOH Hayato

    電子情報通信学会技術研究報告   Vol. 115 ( 388(PRMU2015 100-114) )   2015

  13. Pattern Classification by Sparse Subspace Method

    伊東隼人, 酒井智弥, 井宮淳

    情報処理学会研究報告(CD-ROM)   Vol. 2010 ( 1 )   2010

  14. Self-attention Class Balanced DenseNet_LSTM framework for Subarachnoid Hemorrhage CT image Classification on Extremely Imbalanced Brain CT Dataset

    Zhongyang LU, Masahiro ODA, Yuichiro HAYASHI, Tao HU, Hayato ITOH, Takeyuki WATADANI, Osamu ABE, Masahiro HASHIMOTO, Masahiro JINZAKI, Kensaku MORI

      Vol. 40回   page: 69 - 75   2021.10

     More details

  15. ピットパターン特徴量の解析に向けた超拡大内視鏡画像の再構成法に関する初期的検討

    伊東 隼人, 小田 昌宏, 森 悠一, 三澤 将史, 工藤 進英, 森 健策

    日本医用画像工学会大会予稿集   Vol. 40回   page: 309 - 317   2021.10

     More details

    Authorship:Lead author   Language:Japanese  

    researchmap

  16. Synthesized Perforation Detection from Endoscopy Videos Using Model Training with Synthesized Images by GAN

    Kai Jiang, Hayato Itoh, Masahiro Oda, Taishi Okumura, Yuichi Mori, Masashi Misawa, Takemasa Hayashi, Shin-Ei Kudo, Kensaku Mori

      Vol. 40回   page: 223 - 228   2021.10

     More details

    Language:English  

    researchmap

  17. [総論]AI時代を見据えた消化器外科手術 AIによる大腸T2癌リンパ節転移予測

    中原 健太, 石田 文生, 一政 克朗, 森 悠一, 三澤 将史, 澤田 成彦, 工藤 進英, Villard Ben, 伊東 隼人, 森 健策

    日本消化器外科学会総会   Vol. 75回   page: WS15 - 6   2020.12

     More details

    Language:Japanese   Publisher:(一社)日本消化器外科学会  

    researchmap

  18. SUN database 大腸ポリープ自動検出器の精度評価に向けた試験用画像

    伊東 隼人, 三澤 将史, 森 悠一, 小田 昌宏, 工藤 進英, 森 健策

    日本コンピュータ外科学会誌   Vol. 22 ( 4 ) page: 346 - 347   2020.11

     More details

    Language:Japanese   Publisher:(一社)日本コンピュータ外科学会  

    researchmap

  19. 【多元計算解剖学の診断・治療・医工学への展開】人工知能に基づく医療機器EndoBRAINの臨床導入 薬事承認取得・保険算定への挑戦

    森 悠一, 工藤 進英, 三澤 将史, 伊東 隼人, 小田 昌宏, 森 健策

    MEDICAL IMAGING TECHNOLOGY   Vol. 38 ( 5 ) page: 213 - 216   2020.11

  20. 大腸鏡画像にYOLO-v3を用いた、穿孔の検出と局在化の比率損失設計に関する予備的研究(Preliminary Study of Loss-Functions Design for Detection and Localization of Perforations with YOLO-v3 in Colonoscopic Images)

    蒋 凱, 伊東 隼人, 小田 昌宏, 奥村 大志, 森 悠一, 三澤 将史, 林 武雅, 工藤 進英, 森 健策

    日本コンピュータ外科学会誌   Vol. 22 ( 4 ) page: 348 - 349   2020.11

     More details

    Language:English   Publisher:(一社)日本コンピュータ外科学会  

    researchmap

  21. 腹腔鏡動画像用オンラインアノテーションツールの開発

    屠 芸豪, 伊東 隼人, 小澤 卓也, 小田 昌宏, 竹下 修由, 伊藤 雅昭, 森 健策

    日本コンピュータ外科学会誌   Vol. 22 ( 4 ) page: 306 - 307   2020.11

     More details

    Language:Japanese   Publisher:(一社)日本コンピュータ外科学会  

    researchmap

  22. 腸閉塞およびイレウスの診断支援システムにおける距離マップの導入

    小田 紘久, 林 雄一郎, 北坂 孝幸, 玉田 雄大, 滝本 愛太朗, 檜 顕成, 内田 広夫, 鈴木 耕次郎, 伊東 隼人, 小田 昌宏, 森 健策

    日本コンピュータ外科学会誌   Vol. 22 ( 4 ) page: 282 - 283   2020.11

     More details

    Language:Japanese   Publisher:(一社)日本コンピュータ外科学会  

    researchmap

  23. 位置特徴量とマルチスケール特徴量による胃壁マイクロCT像からの胃壁の層構造及び腫瘍の抽出

    御手洗 翠, 小田 紘久, 杉野 貴明, 守谷 享泰, 伊東 隼人, 小田 昌宏, 小宮山 琢真, 古川 和宏, 宮原 良二, 藤城 光弘, 森 雅樹, 高畠 博嗣, 名取 博, 森 健策

    日本医用画像工学会大会予稿集   Vol. 39回   page: 569 - 572   2020.9

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    本稿では,Spherical K-means(SpK)により抽出された複数スケールの特徴量と位置特徴量を用いた胃壁μCT画像からの粘膜層,粘膜下層,筋層及び腫瘍の抽出手法について報告する.胃壁のμCT画像から解剖学的構造を抽出することで腫瘍の立体構造把握が可能である.従来手法では複数スケールの特徴抽出フィルタをSpKによって学習し,パッチから特徴抽出を行った.その後得られた特徴量に基づいてSVMによって分類を行っていた.しかし,SpKにより生成された特徴抽出フィルタは濃度値に基づく特徴量を抽出するため,粘膜層と筋層のように濃度値が類似した領域同士の特徴量が近しい値になる.本手法ではSpKにより得られる複数スケールの特徴量にパッチの位置特徴量を追加して分類する事によって,濃度値に基づく特徴量では分類できない領域の分類を可能にする.パッチの位置特徴量としてパッチの中心座標を用いる.本手法を胃壁μCT像に適用した結果,粘膜層,粘膜下層,筋層及び腫瘍の抽出に対するDICE係数は68.3%,63.2%,82.2%,69.1%であった.特に従来手法と比べ,腫瘍の抽出に対するDICE係数は26.1%向上した.(著者抄録)

    researchmap

  24. 大腸内視鏡のための教師なし深度画像推定法における補助タスク検討

    伊東 隼人, 小田 昌宏, 森 悠一, 三澤 将史, 工藤 進英, 堀田 欣一, 高畠 博嗣, 森 雅樹, 名取 博, 森 健策

    日本医用画像工学会大会予稿集   Vol. 39回   page: 563 - 568   2020.9

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    医用画像理解においては人体に関する3次元情報の抽出が非常に重要である.一方で,一般的な大腸内視鏡では2次元画像もしくはそれらの時系列集合しか得られない.大腸内視鏡診断におけるコンピュータ支援診断システムを構築する上で,単眼画像から大腸形状の3次元情報を抽出・推定する技術が求められている.先行研究として,多視点幾何学に基づくカメラ運動推定を視差推定の補助タスクとして加えた深度推定方法が提案されている.しかし,あくまで相対的な深度情報しか得られない上にその推定精度が不十分であると報告されている.本研究では深度推定の高精度化を目的に,拡散反射モデルを深度推定に組み込むための補助タスクを提案し,先行研究との比較実験的を行った.実験結果より提案した補助タスクが大腸腸壁のテクスチャに起因する誤推定を抑制することを示し,深度推定の高精度化を達成した.(著者抄録)

    researchmap

  25. 腹腔鏡手術動画像データベース構築に向けたリモートアノテーションツールのプロトタイプ開発

    屠 芸豪, 伊東 隼人, 小澤 卓也, 小田 昌宏, 竹下 修由, 伊藤 雅昭, 森 健策

    日本医用画像工学会大会予稿集   Vol. 39回   page: 611 - 615   2020.9

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    本稿では腹腔鏡手術動画像における術具や解剖構造をウェブブラウザを介してアノテーションするためのリモートアノテーションツールの開発について報告する.腹腔鏡手術動画像において術具や解剖構造のアノテーションデータを作成することにより,機械学習を利用した画像処理によって手技解析法や手術支援技術の開発が可能となる.先行研究として我々は動画像に向けたアノテーションソフトウェアNuVATを開発した.しかし大規模データベースの構築においては,病院へ専門的な機材やソフトウェアを導入するコストが高い.加えて,データの外部への持ち出しによる流出が懸念される.そこで,開発したソフトウェアをウェブブラウザ上で動作するように拡張することによって,病院へのシステム導入のコスト削減ならびにwebを介したアノテーション作業によって作業の効率化を図る.(著者抄録)

    researchmap

  26. 広範囲の隣接関係を考慮したグラフニューラルネットワークを用いた腹部動脈血管名自動命名の検討

    日比 裕太, 林 雄一郎, 北坂 孝幸, 伊東 隼人, 小田 昌宏, 三澤 一成, 森 健策

    日本医用画像工学会大会予稿集   Vol. 39回   page: 268 - 271   2020.9

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    本稿では,3次元腹部CT像から抽出された腹部動脈領域に対してGraph Neural Network(GNN)を用いた血管名自動命名についての検討を行ったので報告する.腹部動脈領域に対して機械学習を用いた血管名自動命名を行う手法はこれまでにもいくつか提案されてきた.また,近年ではグラフ構造に対する機械学習が盛んに行われており,その有用性が示されているため,GNNを用いた機械学習による血管名自動命名について検討した.一般的なGNNでは一つの隣接関係だけを学習することしかできないが,広い範囲の隣接関係を同時に使うことで高精度の自動命名が実現可能であると考えた.本稿では範囲の異なる複数の隣接関係を用いて学習を行うことができるMixHopと一般的なGNNであるGraph Convolutional Networkを使用し,CT像100症例に対して10分割交差検定による実験と比較を行った.実験の結果,MixHopは全症例のノードを用いた評価では85.2%,データセットごとの評価では最高で88.2%の精度で命名することができた.(著者抄録)

    researchmap

  27. 大腸内視鏡における穿孔の自動検出および位置推定に関する予備的検討

    蒋 凱, 伊東 隼人, 小田 昌宏, 奥村 大志, 森 悠一, 三澤 将史, 林 武雅, 工藤 進英, 森 健策

    日本医用画像工学会大会予稿集   Vol. 39回   page: 29 - 29   2020.9

     More details

    Language:English  

    researchmap

  28. AIによるSSA/Pの超拡大内視鏡診断

    小川 悠史, 工藤 進英, 森 悠一, 三澤 将史, 片岡 伸一, 前田 康晴, 一政 克朗, 石垣 智之, 工藤 豊樹, 若村 邦彦, 林 武雅, 馬場 俊之, 石田 文生, 伊東 隼人, 小田 昌宏, 森 健策

    日本大腸検査学会雑誌   Vol. 36 ( 2 ) page: 125 - 125   2020.5

     More details

    Language:Japanese   Publisher:(社)日本大腸検査学会  

    J-GLOBAL

    researchmap

  29. EndoBRAIN, a computer-aided diagnostic software for colonoscopy: Its clinical effectiveness and cost reduction expected from its use.

    森悠一, 工藤進英, 三澤将史, 武田健一, 前田康晴, 小川悠史, 一政克朗, 若村邦彦, 林武雅, 工藤豊樹, 宮地英行, 馬場俊之, 伊東隼人, 小田昌宏, 森健策

    日本大腸検査学会雑誌   Vol. 36 ( 2 ) page: 77 - 82   2020.5

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  30. 腫瘍の診断・治療 AIを用いた大腸T2癌リンパ節転移予測 全層切除に向けた取り組み

    一政 克朗, 工藤 進英, 森 悠一, 中原 健太, 神山 勇太, 三澤 将史, 島田 翔士, 竹原 雄介, 榎並 延太, 工藤 豊樹, 林 武雅, 若村 邦彦, 澤田 成彦, 馬場 俊之, Villard Ben, 伊東 隼人, 森 健策, 石田 文生

    日本大腸検査学会雑誌   Vol. 36 ( 2 ) page: 117 - 117   2020.5

     More details

    Language:Japanese   Publisher:(社)日本大腸検査学会  

    J-GLOBAL

    researchmap

  31. Electronic bowl cleansing based on CycleGAN and its application to detection of intestinal obstruction part

    西尾光平, 小田紘久, 千馬耕亮, 北坂孝幸, 林雄一郎, 伊東隼人, 小田昌宏, 檜顕成, 内田広夫, 森健策, 森健策, 森健策

    電子情報通信学会技術研究報告   Vol. 119 ( 399(MI2019 65-123)(Web) )   2020

  32. Clinical Application of EndoBRAIN, a Medical Device Adopting Artificial Intelligence: Challenges to Obtain Regulatory Approval and Insurance Reimbursement

    森悠一, 森悠一, 工藤進英, 三澤将史, 伊東隼人, 小田昌宏, 森健策

    Medical Imaging Technology (Web)   Vol. 38 ( 5 )   2020

  33. Segmentation of Layer Structure of Stomach Wall and Tumor from MicroCT volumes Using Neural Network and Spherical K-means

    御手洗翠, 小田紘久, 杉野貴明, 守谷享泰, 伊東隼人, 小田昌宏, 小宮山琢真, 古川和宏, 宮原良二, 藤城光弘, 森雅樹, 高畠博嗣, 名取博, 森健策, 森健策, 森健策

    電子情報通信学会技術研究報告(Web)   Vol. 120 ( 156(MI2020 17-32) )   2020

  34. Grad-CAMを用いた脳CT像からのくも膜下出血の出血領域可視化に関する検討(Visualization of subarachnoid hemorrhage area from brain CT images using Grad-CAM)

    魯 仲陽, 伊東 隼人, 小田 昌宏, 林 雄一郎, 渡谷 岳行, 阿部 修, 橋本 正弘, 陣崎 雅弘, 森 健策

    日本コンピュータ外科学会誌   Vol. 21 ( 4 ) page: 229 - 230   2019.11

     More details

    Language:English   Publisher:(一社)日本コンピュータ外科学会  

    researchmap

  35. Preliminary study on automated perforation detection in Endoscopic Submucosal Dissection

    大石仁美, 伊東隼人, 森悠一, 三澤将史, 林武雅, 奥村大志, 小田昌宏, 工藤進英, 森健策, 森健策, 森健策

    日本コンピュータ外科学会誌   Vol. 21 ( 4 ) page: 232 - 233   2019.11

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  36. Segmentation of Stomach Wall on Micro-CT Volumes using Multi-scale Representation Learning

    御手洗翠, 小田紘久, 杉野貴明, 守谷享泰, 伊東隼人, 小田昌宏, 小宮山琢真, 古川和宏, 宮原良二, 藤城光弘, 森雅樹, 高畠博嗣, 名取博, 森健策, 森健策, 森健策

    日本コンピュータ外科学会誌   Vol. 21 ( 4 ) page: 273 - 274   2019.11

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  37. Accuracy Improvement Automated Anatomical Labeling of Abdominal Arteries Using Graph Convolutional Neural Network by Data Augmentation

    日比裕太, 林雄一郎, 北坂孝幸, 伊東隼人, 小田昌宏, 三澤一成, 森健策, 森健策, 森健策

    日本コンピュータ外科学会誌   Vol. 21 ( 4 ) page: 226 - 227   2019.11

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  38. 小児腸閉塞患者のCT像における電子洗浄手法の評価

    西尾 光平, 小田 紘久, 千馬 耕亮, 北坂 孝幸, 林 雄一郎, 伊東 隼人, 小田 昌宏, 檜 顕成, 内田 広夫, 森 健策

    日本コンピュータ外科学会誌   Vol. 21 ( 4 ) page: 321 - 321   2019.11

     More details

    Language:Japanese   Publisher:(一社)日本コンピュータ外科学会  

    J-GLOBAL

    researchmap

  39. Forceps region segmentation on laparoscopic movies using data augmentation based on 3D models of forceps

    小澤卓也, 小田紘久, 伊東隼人, 北坂孝幸, 林雄一郎, 小田昌宏, 三澤一成, 竹下修由, 伊藤雅昭, 森健策, 森健策, 森健策

    日本コンピュータ外科学会誌   Vol. 21 ( 4 ) page: 284 - 285   2019.11

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  40. SSA/Pの人工知能支援下超拡大内視鏡診断

    小川 悠史, 工藤 進英, 森 悠一, 三澤 将史, 武田 健一, 片岡 伸一, 前田 康晴, 一政 克朗, 工藤 豊樹, 若村 邦彦, 林 武雅, 馬場 俊之, 石田 文生, 伊東 隼人, 小田 昌宏, 森 健策

    Gastroenterological Endoscopy   Vol. 61 ( Suppl.2 ) page: 2167 - 2167   2019.10

     More details

    Language:Japanese   Publisher:(一社)日本消化器内視鏡学会  

    researchmap

  41. グラフ畳み込みニューラルネットワークを用いた腹部動脈血管名自動命名の初期検討

    日比 裕太, 林 雄一郎, 北坂 孝幸, 伊東 隼人, 小田 昌宏, 三澤 一成, 森 健策

    日本医用画像工学会大会予稿集   Vol. 38回   page: 615 - 617   2019.7

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    本稿では,3次元腹部CT像から抽出された腹部動脈領域に対してグラフ畳み込みニューラルネットワークを用いた血管名自動命名についての検討を行ったので報告する.血管は構造が複雑で個人差が大きく,その構造の把握は困難である.血管名を自動命名することにより医師が外科手術の際に患者の血管構造を把握する助けとなり,医師の負担を軽減することができる.そのため,これまで腹部動脈領域に対して機械学習を用いた血管名自動命名を行う手法がいくつか提案されてきた.また,近年ではグラフ構造に対する機械学習が盛んに行われており,その有用性が示されている.そこで本稿では,血管構造をグラフ構造と捉え,血管が持つ太さや長さ,腹部臓器との位置関係などを特徴量としてグラフ畳み込みニューラルネットワークを用いた機械学習による血管名自動命名を行った.100症例のCT像に対して血管名自動命名を行った結果,平均精度は85.2%であった.(著者抄録)

    J-GLOBAL

    researchmap

  42. 小児腸閉塞患者のCT像におけるCycleGANを用いた電子洗浄手法の検討

    西尾 光平, 小田 紘久, 千馬 耕亮, 北坂 孝幸, 伊東 隼人, 小田 昌宏, 檜 顕成, 内田 広夫, 森 健策

    日本医用画像工学会大会予稿集   Vol. 38回   page: 426 - 429   2019.7

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    本研究では,小児腸閉塞患者のCT像における,CycleGANを用いた電子洗浄手法を提案する.これまでの腸管閉塞部位検出手法は造影されていない残渣を含んだ腸管を対象としており,残渣と濃度値の類似した腹水などを誤抽出する場合があった.そこで,腸管内の濃度値を空気と同程度に変換し,残渣が含まれていないCT像の生成(電子洗浄)を目指す.多くの電子洗浄手法は残渣が造影されたCT像に対する手法であり,血液の造影のみである小児腸閉塞患者のCT像に適用することは困難である.そこで,本研究ではCycleGANを用いた電子洗浄手法について検討する.残渣の多い小児腸閉塞患者のCT像と残渣の少ない大腸CT検査画像を学習データとして,CycleGANによる双方向それぞれの画像生成モデルを学習する.その後,残渣の多い画像から残渣の少ない画像への生成モデルを小児腸閉塞患者のCT像に適用する.実験の結果,小児腸閉塞患者のCT像から残渣の少ないCT像を得ることができた.(著者抄録)

    J-GLOBAL

    researchmap

  43. 表現学習とSVMによる胃壁マイクロCT像の半教師ありセグメンテーション手法

    御手洗 翠, 小田 紘久, 杉野 貴明, 守谷 享泰, 伊東 隼人, 小田 昌宏, 小宮山 琢真, 森 雅樹, 高畠 博嗣, 名取 博, 森 健策

    日本医用画像工学会大会予稿集   Vol. 38回   page: 246 - 248   2019.7

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    本稿では,Spherical K-means(SpK)による表現学習とSVMを用いた胃壁μCT像から粘膜層,粘膜下層,筋層及び腫瘍を半教師ありで抽出する手法について報告する.μCT画像はμmオーダーで標本を3次元的に観察可能であり,胃壁μCT像から腫瘍及び層構造を抽出することで腫瘍の立体的構造把握が可能となる.しかし,豊富なラベルデータを作成するのは容易ではないため,教師ありの抽出手法を用いるのは難しい.また,胃壁μCT像はコントラストが低いことから,教師なしの抽出手法で精度良く抽出することは困難である.そこで本手法では,対象画像とごく少量のラベルデータを利用する半教師ありの抽出手法により問題の解決,抽出精度の向上を図った.本手法は(1)SpKによる表現学習,(2)特徴抽出,(3)SVMを用いたラベルの割り当ての3段階から成る.本手法を胃壁μCT像に適用した結果,粘膜層,粘膜下層,筋層及び腫瘍の抽出のF値の平均がそれぞれ59.6%,41.9%,70%,32.3%であった.(著者抄録)

    researchmap

  44. 少量のラベルデータを用いた学習によるイレウス症例CT像における拡張腸管の自動抽出

    小田 紘久, 西尾 光平, 北坂 孝幸, 天野 日出, 千馬 耕亮, 内田 広夫, 鈴木 耕次郎, 伊東 隼人, 小田 昌宏, 森 健策

    日本医用画像工学会大会予稿集   Vol. 38回   page: 143 - 146   2019.7

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    本稿では,Fullyconvolutionalnetwork(FCN)を用いたイレウス患者のCT像における腸管領域の抽出において,手塗りされた教師データが少量であっても精度よく抽出を行う手法を提案する.腸閉塞をはじめとするイレウス症例の緊急診断支援のため,腸管を抽出してその走行を提示するシステムの開発が必要である.一般にFCNの学習を行う場合には大量の学習データが必要であるが,小腸は複雑に入り組んでいるほか非常に長く,手動でのラベル作成は容易でない.本稿では症例ごと7枚のAxialスライスのみにラベルが作成された教師データを効率的に使用するためのデータ拡張として,回転・非剛体変形などの一般的な画像処理のほか,事前に教師データごとにラベルを他のスライスへ伝播する処理を行うことで,ラベルが手作業で作成されていないスライスも学習に使用可能とする.ネットワークは3DU-netをもとに入出力サイズを変更を施したSuppressed3DU-netを用いた.実験は他のスライスへの伝播処理の有効性を評価するため,伝播処理あり・なしの比較を行った.伝播あり・なしの抽出精度を表すDice係数はそれぞれ0.744,0.782であり,伝播処理を行わないほうが高い抽出精度が得られることが知られた.(著者抄録)

    J-GLOBAL

    researchmap

  45. 「大腸画像強調内視鏡の現状と未来」 人工知能(AI)に基づく大腸内視鏡検査によるリアルタイム病変検出支援システム

    趙 智成, 工藤 進英, 三澤 将史, 前田 康晴, 武田 健一, 一政 克朗, 中村 大樹, 矢川 裕介, 豊嶋 直也, 森 悠一, 小形 典之, 工藤 豊樹, 久行 友和, 林 武雅, 若村 邦彦, 馬場 俊之, 石田 文生, 伊東 隼人, 小田 昌宏, 森 健策

    日本大腸検査学会雑誌   Vol. 35 ( 2 ) page: 112 - 112   2019.4

     More details

    Language:Japanese   Publisher:(社)日本大腸検査学会  

    J-GLOBAL

    researchmap

  46. 大腸内視鏡とAI 超拡大内視鏡Endocytoを用いた診断支援システム

    森 悠一, 工藤 進英, 三澤 将史, 伊東 隼人, 森 健策

    BIO Clinica   Vol. 34 ( 3 ) page: 313 - 316   2019.3

     More details

    Language:Japanese   Publisher:(株)北隆館  

    大腸内視鏡は、大腸癌の発見・治療にとどまらず、大腸癌死亡を抑制する点においても疫学的に有効性が高い。しかし、ヒューマンエラーにより、有効な癌抑制効果が均一に得られていないのも事実である。このアンメットニーズを人工知能の力で解決しようとする、内視鏡診断支援システム(CAD)の研究開発がここ10年程、急速に脚光を浴びている。本稿では先日、薬事承認された内視鏡CAD、EndoBRAINをはじめとして、内視鏡CADの状況について概説する。(著者抄録)

    J-GLOBAL

    researchmap

  47. AIを用いた大腸内視鏡診療(病変の自動検出・病理診断予測)は実用化できるか

    森悠一, 工藤進英, 三澤将史, 伊東隼人, 森健策

    先進内視鏡治療研究会   Vol. 13th   2019

  48. SSA/Pの人工知能支援下超拡大内視鏡診断

    小川悠史, 工藤進英, 森悠一, 三澤将史, 武田健一, 片岡伸一, 前田康晴, 一政克朗, 工藤豊樹, 若村邦彦, 林武雅, 馬場俊之, 石田文生, 伊東隼人, 小田昌宏, 森健策

    Gastroenterological Endoscopy (Web)   Vol. 61 ( Supplement2 )   2019

  49. A Study on Automated Lymph Node Matching Method Considering Abdominal Organs Deformation in Follow-up CT Volumes

    舘高基, 小田昌宏, 林雄一郎, 伊東隼人, 中村嘉彦, 北坂孝幸, 三澤一成, 森健策

    電子情報通信学会技術研究報告   Vol. 118 ( 412(MI2018 59-115)(Web) )   2019

  50. Development of Tool for Correcting Detection Failure of Intestinal Obstructions on Ileus Diagnosis Assistant System

    西尾光平, 小田紘久, 千馬耕亮, 北坂孝幸, ROTH Holger R., 伊東隼人, 林雄一郎, 小田昌宏, 檜顕成, 内田広夫, 森健策, 森健策, 森健策

    日本コンピュータ外科学会誌   Vol. 20 ( 4 ) page: 348 - 349   2018.10

     More details

    Language:Japanese  

    J-GLOBAL

    researchmap

  51. Fast Marching Algorithmに基づく小児CT像からの腸管閉塞部位検出手法

    西尾 光平, 小田 紘久, 千馬 耕亮, 北坂 孝幸, Roth Holger R., 伊東 隼人, 小田 昌宏, 檜 顕成, 内田 広夫, 森 健策

    日本医用画像工学会大会予稿集   Vol. 37回   page: 90 - 93   2018.7

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    本稿では、血管造影された小児イレウス(腸閉塞)患者のCT像から腸管閉塞部位を検出する手法を提案する。イレウス患者の腸管内に含まれている食物・便は、腹水等と濃度値の差が小さいことから、これまでの抽出手法では精度よく腸管内領域を抽出できない。そこで本研究では一般的に血管造影CT像が使われることに着目し、Fast Marching Algorithmに基づいた手法により腸を抽出することで腸管閉塞部位検出を行う。手動で指定した初期の抽出開始点から食物・便である液体領域と空気領域を交互に抽出することで、閉塞部位までの腸管内領域を辿り閉塞部位を検出する。このとき、液体領域では血管造影により造影された腸壁の強調処理とFast Marching Algorithmを組み合わせた手法、空気領域では領域拡張法を用いる。提案手法をイレウスによる腸管閉塞部位が存在する血管造影された小児CT像5症例に適用し、辿ることのできた腸管の割合により評価を行った。辿ることのできた腸管の割合の平均は75.4%であった。(著者抄録)

    J-GLOBAL

    researchmap

  52. 機械学習による内視鏡動画インスタンスセグメンテーションのための手動アノテーションツールの開発

    小澤 卓也, 小田 紘久, 伊東 隼人, 北坂 孝幸, Roth Holger R., 小田 昌宏, 林 雄一郎, 三澤 一成, 伊藤 雅昭, 竹下 修由, 森 健策

    日本医用画像工学会大会予稿集   Vol. 37回   page: 94 - 97   2018.7

     More details

    Language:Japanese   Publisher:日本医用画像工学会  

    本論文では、内視鏡動画像において術具や解剖領域を機械学習によりインスタンスセグメンテーションするために必要とされるアノテーションデータを効率的に生成可能な内視鏡動画像アノテーションツール開発について報告する。内視鏡動画像に対して詳細に術具や解剖領域がアノテーションされたデータは、手術支援機器における画像からの手術状況の自動識別技術の実現や内視鏡手術手技の評価技術の実現など利用価値が高い。従ってこれらのデータベースを効率的に生成するための基盤構築が求められている。そこで本研究では、動画像に対してセグメンテーション作業を行う為のツールNuVATを開発した。このツールにより、セグメンテーション作業をお絵かきツールを扱う感覚で簡単に行うことが可能になった。実際に国立がん研究センター東病院において、腹腔鏡動画像における術具・解剖領域のアノテーション作業に利用された。(著者抄録)

    J-GLOBAL

    researchmap

  53. 人工知能に基づく大腸内視鏡のポリープ自動検出ソフトウェア

    三澤 将史, 工藤 進英, 森 悠一, 片岡 伸一, 中村 大樹, 武田 健一, 矢川 裕介, 一政 克朗, 石垣 智之, 豊嶋 直也, 小形 典之, 工藤 豊樹, 久行 友和, 林 武雅, 若村 邦彦, 馬場 俊之, 石田 文生, 伊東 隼人, 小田 昌宏, 森 健策

    Gastroenterological Endoscopy   Vol. 60 ( Suppl.1 ) page: 704 - 704   2018.4

     More details

    Language:Japanese   Publisher:(一社)日本消化器内視鏡学会  

    researchmap

  54. U-Netを用いた腹腔鏡動画像における出血領域検出に関する検討

    小澤卓也, 小田紘久, 伊東隼人, 北坂孝幸, ROTH Holger R., 小田昌宏, 林雄一郎, 三澤一成, 伊藤雅昭, 竹下修由, 森健策, 森健策, 森健策

    日本コンピュータ外科学会誌   Vol. 20 ( 4 )   2018

  55. 人工知能に基づく大腸内視鏡のポリープ自動検出ソフトウェア

    三澤将史, 工藤進英, 森悠一, 片岡伸一, 中村大樹, 武田健一, 矢川裕介, 一政克朗, 石垣智之, 豊嶋直也, 小形典之, 工藤豊樹, 久行友和, 林武雅, 若村邦彦, 馬場俊之, 石田文生, 伊東隼人, 小田昌宏, 森健策

    Gastroenterological Endoscopy (Web)   Vol. 60 ( Supplement1 )   2018

  56. Artificial Intelligence in Super-magnified endoscopic images

    森健策, 伊東隼人, 三澤将史, 森悠一, 工藤進英

    Optics & Photonics Japan講演予稿集(CD-ROM)   Vol. 2018   2018

  57. 大腸・小腸疾患に対する診断の進歩 人工知能による、リアルタイム大腸内視鏡診断への挑戦

    森 悠一, 工藤 進英, 三澤 将史, 武田 健一, 一政 克朗, 前田 康晴, 石垣 智之, 若村 邦彦, 林 武雅, 小田 昌宏, 伊東 隼人, 森 健策

    日本大腸肛門病学会雑誌   Vol. 70 ( 抄録号 ) page: A71 - A71   2017.9

     More details

    Language:Japanese   Publisher:(一社)日本大腸肛門病学会  

    researchmap

  58. Statistical properties of Colour HoG

    TANJI Kaori, IMIYA Atsushi, ITOH Hayato

    電子情報通信学会技術研究報告   Vol. 116 ( 411(PRMU2016 127-151) )   2017

  59. 人工知能による,リアルタイム大腸内視鏡診断への挑戦

    森悠一, 工藤進英, 三澤将史, 武田健一, 一政克朗, 前田康晴, 石垣智之, 若村邦彦, 林武雅, 小田昌宏, 伊東隼人, 森健策

    日本大腸肛門病学会雑誌(Web)   Vol. 70   2017

  60. Mathematical Properties of the Gradient-Based Discriminative Methods

      Vol. 45 ( 4 ) page: 311 - 315   2015.6

     More details

    Language:English  

    CiNii Books

    researchmap

  61. 3D Global Image Registration Using Local Linear Property in Sparse Dictionary

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告   Vol. 113 ( 402(PRMU2013 91-120) )   2014

  62. Explicit Local Linear Method for 2D Affine Image Registration

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告   Vol. 113 ( 346(PRMU2013 68-90) )   2013

  63. Validation of Dimension Reduction Methods for Two-Dimensional Image Pattern Classification

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告   Vol. 112 ( 495(PRMU2012 180-226) )   2013

  64. Effects of dimension reduction on appearance-based pattern classification

    Itoh Hayato, Sakai Tomoya, Imiya Atsushi

    Technical report of IEICE. PRMU   Vol. 112 ( 357 ) page: 25 - 30   2012.12

     More details

    Language:English   Publisher:The Institute of Electronics, Information and Communication Engineers  

    In this paper, we experimentally evaluate the validities of dimension reduction for image pattern recognition. Biometric such as face recognition, iris recognition, and finger print recognition are achieved as image pattern recognition. Image pattern recognition uses pattern recognition techniques for classification of image data. For the numerical achievement of image pattern recognition techniques, images are sampled using the array of pixels. This sampling procedure derives vectors in higher dimensional metric space from image patterns. For accurate achievement of pattern recognition techniques, the dimension reduction of data vectors are essential methodology, since time and space complexities of data processing depends on the dimension of data. However, dimension reduction causes information loss of geometrical and topological features of image patterns. Desired dimension reduction selects appropriate low-dimensional subspace preserving the information for classification. We experimentally evaluate effects of the dimension reduction techniques which are used as preprocessing of pattern recognition of image data.

    CiNii Books

    researchmap

  65. NN-based Local Subspace Method for Image Registration with Sparse Dictionary

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告   Vol. 112 ( 197(PRMU2012 30-50) )   2012

  66. Efficient Global Image Registration using Local Linearity of Transformed Image

    伊東隼人, LU Shuang, 酒井智弥, 井宮淳

    電子情報通信学会技術研究報告   Vol. 111 ( 331(MI2011 63-76) )   2011

▼display all

Presentations 51

  1. Unsupervised colonoscopic depth estimation by domain translations with a Lambertian-reflection keeping auxiliary task

    Hayato Itoh, Masahiro Oda, Yuichi Mori, Masashi Misawa, Shin-Ei Kudo, Kenichiro Imai, Sayo Ito, Kinichi Hotta, Hirotsugu Takabatake, Masaki Mori, Hiroshi Natori, Kensaku Mori

    International Journal of Computer Assisted Radiology and Surgery  2021.5  SPRINGER HEIDELBERG

     More details

    Event date: 2021.5

    Language:English  

    Purpose A three-dimensional (3D) structure extraction technique viewed from a two-dimensional image is essential for the development of a computer-aided diagnosis (CAD) system for colonoscopy. However, a straightforward application of existing depth-estimation methods to colonoscopic images is impossible or inappropriate due to several limitations of colonoscopes. In particular, the absence of ground-truth depth for colonoscopic images hinders the application of supervised machine learning methods. To circumvent these difficulties, we developed an unsupervised and accurate depth-estimation method. Method We propose a novel unsupervised depth-estimation method by introducing a Lambertian-reflection model as an auxiliary task to domain translation between real and virtual colonoscopic images. This auxiliary task contributes to accurate depth estimation by maintaining the Lambertian-reflection assumption. In our experiments, we qualitatively evaluate the proposed method by comparing it with state-of-the-art unsupervised methods. Furthermore, we present two quantitative evaluations of the proposed method using a measuring device, as well as a new 3D reconstruction technique and measured polyp sizes. Results Our proposed method achieved accurate depth estimation with an average estimation error of less than 1 mm for regions close to the colonoscope in both of two types of quantitative evaluations. Qualitative evaluation showed that the introduced auxiliary task reduces the effects of specular reflections and colon wall textures on depth estimation and our proposed method achieved smooth depth estimation without noise, thus validating the proposed method. Conclusions We developed an accurate depth-estimation method with a new type of unsupervised domain translation with the auxiliary task. This method is useful for analysis of colonoscopic images and for the development of a CAD system since it can extract accurate 3D information.

    researchmap

  2. Report on MICCAI 2019

    小田昌宏, 伊東隼人, 宮内翔子, 諸岡健一, 松崎博貴, 花岡昇平, 古川亮, 増谷佳孝, 森健策, 森健策

    電子情報通信学会技術研究報告  2020 

     More details

    Event date: 2020

    researchmap

  3. Robust endocytoscopic image classification based on higher-order symmetric tensor analysis and multi-scale topological statistics.

    Hayato Itoh, Yukitaka Nimura, Yuichi Mori, Masashi Misawa, Shin-ei Kudo, Kinichi Hotta, Kazuo Ohtsuka, Shoichi Saito, Yutaka Saito, Hiroaki Ikematsu, Yuichiro Hayashi, Masahiro Oda, Kensaku Mori

    Int. J. Comput. Assist. Radiol. Surg.  2020 

  4. Preliminary Study on Handling Different Data Distributions in Multi-Institute Datasets for Automated Endocytoscopic Image Classification

    伊東隼人, 森悠一, 三澤将史, 小田昌宏, 工藤進英, 森健策, 森健策, 森健策

    日本コンピュータ外科学会誌  2019.11 

     More details

    Event date: 2019.11

    Language:Japanese  

    researchmap

  5. Multilinear Subspace Method Based on Geodesic Distance for Volumetric Object Classification.

    Hayato Itoh, Atsushi Imiya

    2019 

  6. 深層学習における学習データセット規模拡大に応じた分類精度向上に関する実験的検討-超拡大大腸内視鏡画像における腫瘍性病変分類に向けた特徴量抽出-

    伊東隼人, 森悠一, 三澤将史, 小田昌宏, 工藤進英, 森健策, 森健策, 森健策

    日本医用画像工学会大会予稿集(CD-ROM)  2019 

     More details

    Event date: 2019

    researchmap

  7. Training-Dataset Construction Method from Imbalanced Dataset-Towards Medical Image Classification based on Machine Learning Techniques-

    伊東隼人, 森悠一, 三澤将史, 小田昌宏, 工藤進英, 森健策, 森健策, 森健策

    日本コンピュータ外科学会誌  2018.10 

     More details

    Event date: 2018.10

    Language:Japanese  

    researchmap

  8. 教師なし深度推定を利用したRGB-D特徴抽出に基づくポリープのトリナリサイズ推定

    伊東 隼人, Roth Holger, 三澤 将史, 森 悠一, 小田 昌宏, 工藤 進英, 森 健策

    日本医用画像工学会大会予稿集  2018.7  日本医用画像工学会

     More details

    Event date: 2018.7

    Language:Japanese  

    大腸内視鏡検査におけるポリープの大きさ情報(サイズ情報)は診断において重要な役割りを果す。5mm以下の小さなポリープ、10mm以上の大きなポリープ、それ以外のポリープといった3カテゴリに関するサイズ推定(トリナリ推定)は治療計画を立てる上で特に重要とされる。しかし内視鏡画像のみに基くサイズ推定は熟練の専門医にとっても難しい。そこで内視鏡画像ならびに動画像からポリープのサイズ推定を行うCADシステムが望まれている。内視鏡画像には三次元情報が欠如しており、二次元画像からの精密なサイズ推定は不良設定問題である。本研究ではこの不良設定問題をトリナリサイズ推定へと緩和し、内視鏡画像からポリープのサイズ推定に有用な特徴を深層学習によって抽出、そして分類する手法を提案する。特徴抽出においては教師無し学習による深度推定を利用する。提案手法により高精度のトリナリサイズ推定を達成した。(著者抄録)

    researchmap

  9. Feature-selection method based on Grassmann distance for the classification of neoplastic polyps on endocytoscopic images (医用画像)

    伊東 隼人, 森 悠一, 三澤 将史, 小田 昌宏, 工藤 進英, 森 健策

    電子情報通信学会技術研究報告 = IEICE technical report : 信学技報  2018.3.19  電子情報通信学会

     More details

    Event date: 2018.3

    Language:English  

    researchmap

  10. Discriminative Feature Selection by Optimal Manifold Search for Neoplastic Image Recognition.

    Hayato Itoh, Yuichi Mori, Masashi Misawa, Masahiro Oda, Shin-ei Kudo, Kensaku Mori

    2018 

  11. Towards Automated Colonoscopy Diagnosis: Binary Polyp Size Estimation via Unsupervised Depth Learning.

    Hayato Itoh, Holger R. Roth, Le Lu 0001, Masahiro Oda, Masashi Misawa, Yuichi Mori, Shin-ei Kudo, Kensaku Mori

    2018 

  12. Distances Between Tensor Subspaces.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2018 

  13. Classification of neoplasia and non-neoplasia for colon endocytoscopic images by convolutional neural network

    伊東隼人, 森悠一, 三澤将史, 小田昌宏, 工藤進英, 森健策, 森健策

    電子情報通信学会技術研究報告  2017 

     More details

    Event date: 2017

    researchmap

  14. Automated Pathological Diagnosis Using Colon Endocytoscopic Images~Analysis of Classification Accuracy of Neoplastic Lesions~

    伊東隼人, 森悠一, 三澤将史, 小田昌宏, 工藤進英, 森健策, 森健策

    日本コンピュータ外科学会誌  2017 

     More details

    Event date: 2017

    researchmap

  15. Mathematical Aspects of Tensor Subspace Method.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2016 

  16. Tensor-Based Methods for Dimension Reduction of Volumetric Data

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告  2016 

     More details

    Event date: 2016

    researchmap

  17. Second-Order Tensor Principal Component Analysis Meets Two-Dimensional Singular Value Decomposition

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告  2015 

     More details

    Event date: 2015

    researchmap

  18. Topology-Preserving Dimension-Reduction Methods for Image Pattern Recognition.

    Hayato Itoh, Tomoya Sakai 0002, Kazuhiko Kawamoto, Atsushi Imiya

    2013 

  19. Pattern Classification by Sparse Subspace Method

    伊東隼人, 酒井智弥, 井宮淳

    情報処理学会研究報告(CD-ROM)  2010 

     More details

    Event date: 2010

    researchmap

  20. Binary Polyp-Size Classification based on Deep-Learning Estimation of Spatial Information

    Hayato Itoh, Masahiro Oda, Kai Jiang, Yuichi Mori, Masashi Misawa, Shin-Ei Kudo, Kenichiro Imai, Sayo Ito, Kinichi Hotta, Kensaku Mori

    Computer Assisted Radiology and Surgery (CARS) 2021  2021.6.21 

     More details

    Event date: 2021.6

    Language:English  

    researchmap

  21. SUN database 大腸ポリープ自動検出器の精度評価に向けた試験用画像

    伊東 隼人, 三澤 将史, 森 悠一, 小田 昌宏, 工藤 進英, 森 健策

    日本コンピュータ外科学会誌  2020.11  (一社)日本コンピュータ外科学会

     More details

    Event date: 2020.11

    Language:Japanese  

    researchmap

  22. 腹腔鏡動画像用オンラインアノテーションツールの開発

    屠 芸豪, 伊東 隼人, 小澤 卓也, 小田 昌宏, 竹下 修由, 伊藤 雅昭, 森 健策

    日本コンピュータ外科学会誌  2020.11  (一社)日本コンピュータ外科学会

     More details

    Event date: 2020.11

    Language:Japanese  

    researchmap

  23. 大腸内視鏡のための教師なし深度画像推定法における補助タスク検討

    伊東 隼人, 小田 昌宏, 森 悠一, 三澤 将史, 工藤 進英, 堀田 欣一, 高畠 博嗣, 森 雅樹, 名取 博, 森 健策

    日本医用画像工学会大会予稿集  2020.9  日本医用画像工学会

     More details

    Event date: 2020.9

    Language:Japanese  

    医用画像理解においては人体に関する3次元情報の抽出が非常に重要である.一方で,一般的な大腸内視鏡では2次元画像もしくはそれらの時系列集合しか得られない.大腸内視鏡診断におけるコンピュータ支援診断システムを構築する上で,単眼画像から大腸形状の3次元情報を抽出・推定する技術が求められている.先行研究として,多視点幾何学に基づくカメラ運動推定を視差推定の補助タスクとして加えた深度推定方法が提案されている.しかし,あくまで相対的な深度情報しか得られない上にその推定精度が不十分であると報告されている.本研究では深度推定の高精度化を目的に,拡散反射モデルを深度推定に組み込むための補助タスクを提案し,先行研究との比較実験的を行った.実験結果より提案した補助タスクが大腸腸壁のテクスチャに起因する誤推定を抑制することを示し,深度推定の高精度化を達成した.(著者抄録)

    researchmap

  24. Visualising decision-reasoning regions in computer-aided pathological pattern diagnosis of endoscytoscopic images based on CNN weights analysis

    Hayato Itoh, Zhongyang Lu, Yuichi Mori, Masashi Misawa, Masahiro Oda, Shin-ei Kudo, Kensaku Mori

    MEDICAL IMAGING 2020: COMPUTER-AIDED DIAGNOSIS  2020  SPIE-INT SOC OPTICAL ENGINEERING

     More details

    Event date: 2020

    Language:English  

    Purpose of this paper is to present a method for visualising decision-reasoning regions in computer-aided pathological pattern diagnosis of endocytoscopic images. Endocytoscope enables us to perform direct observation of cells and their nuclei on the colon wall at maximum 500-times ultramagnification. For this new modality, computer-aided pathological diagnosis system is strongly required for the support of non-expert physicians. To develop a CAD system, we adopt convolutional neural network (CNN) as the classifier of endocytoscopic images. In addition to this classification function, based on CNN weights analysis, we develop a filter function that visualises decision-reasoning regions on classified images. This visualisation function helps novice endocytoscopists to develop their understanding of pathological pattern on endocytoscopic images for accurate endocytoscopic diagnosis. In numerical experiment, our CNN model achieved 90 % classification accuracy. Furthermore, experimental results show that decision-reasoning regions suggested by our filter function contain characteristic pit patterns in real endocytoscopic diagnosis.

    researchmap

  25. Stable polyp-scene classification via subsampling and residual learning from an imbalanced large dataset.

    Hayato Itoh, Holger Roth, Masahiro Oda, Masashi Misawa, Yuichi Mori, Shin-Ei Kudo, Kensaku Mori

    Healthcare technology letters  2019.12 

     More details

    Event date: 2019.12

    Language:English  

    This Letter presents a stable polyp-scene classification method with low false positive (FP) detection. Precise automated polyp detection during colonoscopies is essential for preventing colon-cancer deaths. There is, therefore, a demand for a computer-assisted diagnosis (CAD) system for colonoscopies to assist colonoscopists. A high-performance CAD system with spatiotemporal feature extraction via a three-dimensional convolutional neural network (3D CNN) with a limited dataset achieved about 80% detection accuracy in actual colonoscopic videos. Consequently, further improvement of a 3D CNN with larger training data is feasible. However, the ratio between polyp and non-polyp scenes is quite imbalanced in a large colonoscopic video dataset. This imbalance leads to unstable polyp detection. To circumvent this, the authors propose an efficient and balanced learning technique for deep residual learning. The authors' method randomly selects a subset of non-polyp scenes whose number is the same number of still images of polyp scenes at the beginning of each epoch of learning. Furthermore, they introduce post-processing for stable polyp-scene classification. This post-processing reduces the FPs that occur in the practical application of polyp-scene classification. They evaluate several residual networks with a large polyp-detection dataset consisting of 1027 colonoscopic videos. In the scene-level evaluation, their proposed method achieves stable polyp-scene classification with 0.86 sensitivity and 0.97 specificity.

    researchmap

  26. Polyp-Size Classification with RGB-D features for Colonoscopy

    Hayato Itoh, Holger R. Roth, Yuichi Mori, Masashi Misawa, Masahiro Oda, Shin-ei Kudo, Kensaku Mori

    MEDICAL IMAGING 2019: COMPUTER-AIDED DIAGNOSIS  2019  SPIE-INT SOC OPTICAL ENGINEERING

     More details

    Event date: 2019

    Language:English  

    Measurement of a polyp size is an essential task in colon cancer screening, since the polyp-size information has critical roles for decision on colonoscopy. However, an estimation of a polyp size from a single view of colonoscope without a measurement device is quite difficult even for expert physicians. To overcome this difficulty, automated size estimation techniques would be desirable for clinical scenes. This paper presents polyp-size classification method with a single colonoscopic image for colonoscopy. Our proposed method estimates depth information from a single colonoscopic image with trained model and utilises the estimated information for the classification. In our method, the model for depth information is obtained by deep learning with colonoscopic videos. Experimental results show the achievement of binary and trinary polyp-size classification with 79% and 74% accuracy from a single still image of a colonoscopic movie.

    researchmap

    Other Link: https://dblp.uni-trier.de/db/conf/micad/micad2019.html#ItohRMMOKM19

  27. Cascade classification of endocytoscopic images of colorectal lesions for automated pathological diagnosis

    Hayato Itoh, Yuichi Mori, Masashi Misawa, Masahiro Oda, Shin-ei Kudo, Kensaku Mori

    MEDICAL IMAGING 2018: COMPUTER-AIDED DIAGNOSIS  2018  SPIE-INT SOC OPTICAL ENGINEERING

     More details

    Event date: 2018

    Language:English  

    This paper presents a new classification method for endocytoscopic images. Endocytoscopy is a new endoscope that enables us to perform conventional endoscopic observation and ultramagnified observation of cell level. This ultramagnified views (endocytoscopic images) make possible to perform pathological diagnosis only on endoscopic views of polyps during colonoscopy. However, endocytoscopic image diagnosis requires higher experiences for physicians. An automated pathological diagnosis system is required to prevent the overlooking of neoplastic lesions in endocytoscopy. For this purpose, we propose a new automated endocytoscopic image classification method that classifies neoplastic and non-neoplastic endocytoscopic images. This method consists of two classification steps. At the first step, we classify an input image by support vector machine. We forward the image to the second step if the confidence of the first classification is low. At the second step, we classify the forwarded image by convolutional neural network. We reject the input image if the confidence of the second classification is also low. We experimentally evaluate the classification performance of the proposed method. In this experiment, we use about 16,000 and 4,000 colorectal endocytoscopic images as training and test data, respectively. The results show that the proposed method achieves high sensitivity 93.4% with small rejection rate 9.3% even for difficult test data.

    researchmap

    Other Link: https://dblp.uni-trier.de/conf/micad/2018

  28. Discrimination of Volumetric Shapes Using Orthogonal Tensor Decomposition.

    Hayato Itoh, Atsushi Imiya

    2018 

  29. 大腸・小腸疾患に対する診断の進歩 人工知能による、リアルタイム大腸内視鏡診断への挑戦

    森 悠一, 工藤 進英, 三澤 将史, 武田 健一, 一政 克朗, 前田 康晴, 石垣 智之, 若村 邦彦, 林 武雅, 小田 昌宏, 伊東 隼人, 森 健策

    日本大腸肛門病学会雑誌  2017.9  (一社)日本大腸肛門病学会

     More details

    Event date: 2017.9

    Language:Japanese  

    researchmap

  30. Analysis of Multilinear Subspaces Based on Geodesic Distance.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2017 

  31. Fast Approximate Karhunen-Loève Transform for Three-Way Array Data.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2017 

  32. Multilinear Methods for Spatio-Temporal Image Recognition.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2017 

  33. Approximation of N-Way Principal Component Analysis for Organ Data.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2016 

  34. Classification of Volumetric Data Using Multiway Data Analysis.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2016 

  35. Volumetric Image Pattern Recognition Using Three-Way Principal Component Analysis.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2016 

  36. Discriminative Properties in Directional Distributions for Image Pattern Recognition.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2015 

  37. Low-Dimensional Tensor Principle Component Analysis.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2015 

  38. Global Volumetric Image Registration Using Local Linear Property of Image Manifold.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2014 

  39. 3D Global Image Registration Using Local Linear Property in Sparse Dictionary

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告  2014 

     More details

    Event date: 2014

    researchmap

  40. Two-Dimensional Global Image Registration Using Local Linear Property of Image Manifold.

    Hayato Itoh, Atsushi Imiya, Tomoya Sakai 0002

    2014 

  41. Validation of Dimension Reduction Methods for Two-Dimensional Image Pattern Classification

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告  2013 

     More details

    Event date: 2013

    researchmap

  42. Explicit Local Linear Method for 2D Affine Image Registration

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告  2013 

     More details

    Event date: 2013

    researchmap

  43. Dimension Reduction Methods for Image Pattern Recognition.

    Hayato Itoh, Tomoya Sakai 0002, Kazuhiko Kawamoto, Atsushi Imiya

    2013 

  44. Local Affine Optical Flow Computation.

    Hayato Itoh, Shun Inagaki, Ming-Ying Fan, Atsushi Imiya, Kazuhiko Kawamoto, Tomoya Sakai 0002

    2013 

  45. Global Image Registration Using Random Projection and Local Linear Method.

    Hayato Itoh, Tomoya Sakai 0002, Kazuhiko Kawamoto, Atsushi Imiya

    2013 

  46. Edge Detection and Smoothing-Filter of Volumetric Data.

    Masaki Narita, Atsushi Imiya, Hayato Itoh

    2012 

  47. NN-based Local Subspace Method for Image Registration with Sparse Dictionary

    ITOH Hayato, IMIYA Atsushi, SAKAI Tomoya

    電子情報通信学会技術研究報告  2012 

     More details

    Event date: 2012

    researchmap

  48. Interpolation of Reference Images in Sparse Dictionary for Global Image Registration.

    Hayato Itoh, Shuang Lu, Tomoya Sakai 0002, Atsushi Imiya

    2012 

  49. Global Image Registration by Fast Random Projection.

    Hayato Itoh, Shuang Lu, Tomoya Sakai 0002, Atsushi Imiya

    2011 

  50. Efficient Global Image Registration using Local Linearity of Transformed Image

    伊東隼人, LU Shuang, 酒井智弥, 井宮淳

    電子情報通信学会技術研究報告  2011 

     More details

    Event date: 2011

    researchmap

  51. コロナ禍における数理工学研究とその応用 Invited

    伊東 隼人

    第1回名古屋大学イニシアティブウェビナー「生命と人工知能」  2021.7.21 

     More details

▼display all

Works 1

  1. SUN Colonoscopy Video Database

    Hayato Itoh, Masashi Misawa, Yuichi Mori, Masahiro Oda, Shin-Ei Kudo, Kensaku Mori

    2020.10

     More details

    Work type:Database science  

KAKENHI (Grants-in-Aid for Scientific Research) 1

  1. Automated diagnosis for colonoscopy using deep learning

    Grant number:17K15971  2017.4 - 2019.3

    MISAWA MASASHI

      More details

    Authorship:Other 

    The aim of this research was to develop computer-aided detection(CADe) and computer-aided diagnosis(CADx) system for colonoscopy and evaluate its diagnostic performance.1.We developed CADx system based on ultra-magnifying endoscopy for differentiating colonic neoplasms and non-neoplasms. Support vector machine which is a traditional machine learning method, was applied for the CADx and achieved 97.4% accuracy. 2.The CADe system that works on conventional endoscopy, was developed using 3-dimensional convolution neural network. We prepared fully annotated 1.8 million frame of colonoscopy videos for machine learning.The CADe system achieved 90% sensitivity for colorectal lesion based on video based analysis.(Misawa M, et al. Gastroenterology 2018)