Development of a Real-Time Computer-Aided Diagnosis System Based on Objective Indicators for Gastrointestinal Endoscopic Image Analysis

• Project/Area Number: 20H04157
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 60040:Computer system-related
• Research Institution: Hiroshima University
• Principal Investigator: Koide Tetsushi 広島大学, ナノデバイス研究所, 准教授 (30243596)
• Co-Investigator(Kenkyū-buntansha): 田中 信治 広島大学, 病院(医), 教授 (00260670) ; 玉木 徹 名古屋工業大学, 工学(系)研究科(研究院), 教授 (10333494)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Project Status: Completed (Fiscal Year 2022)
• Budget Amount: ¥17,810,000 (Direct Cost: ¥13,700,000、Indirect Cost: ¥4,110,000); Fiscal Year 2022: ¥4,810,000 (Direct Cost: ¥3,700,000、Indirect Cost: ¥1,110,000); Fiscal Year 2021: ¥4,810,000 (Direct Cost: ¥3,700,000、Indirect Cost: ¥1,110,000); Fiscal Year 2020: ¥8,190,000 (Direct Cost: ¥6,300,000、Indirect Cost: ¥1,890,000)
• Keywords: 計算機システム / 画像認識 / 内視鏡診断支援 / 医用応用のための画像診断システム / ハード・ソフト協調設計 / リアルタイム画像処理 / 機械学習・深層学習 / 転移学習 / 集積回路 / 機械学習 / 深層学習

Outline of Research at the Start

本研究では、高精細画像な消化管の拡大・非拡大の狭帯域光観察（NBI : Narrow Band Imaging）内視鏡画像を、医師の経験に基づいた客観的指標を用いて、腫瘍を分類し、医師へ提示することが可能な汎用的リアルタイム診断支援（CAD : Computer-Aided Diagnosis）システムを開発するために、ソフトウェアとハードウェアの協調設計を行うことにより、消化管内視鏡画像の定量化による客観的指標を提示することを可能とするリアルタイムCADシステムの基盤技術の構築を目指す。

Outline of Final Research Achievements

The colorectal NBI magnified endoscopy diagnosis support system developed in this study can provide not only the JNET classification but also the NICE classification, a colorectal NBI non-magnified endoscopic finding classification used in the US and EU, with high accuracy, sensitivity, specificity and other quantitative evaluation indices of more than 90%. In addition, segmentation of lesions and other areas of normal mucosa in the JNET and NICE classifications can be extracted from the entire endoscopic image. In addition, we have developed a high-speed CAD system for real-time processing, which enables the identification of a wide range of lesions on full HD video images.

Academic Significance and Societal Importance of the Research Achievements

本研究で開発したリアルタイム内視鏡診断支援システムにより、内視鏡画像の客観的指標となる数値を医師の診察時にリアルタイムで提示することができ、医師の客観的な判断を可能にし、診断の正確度を向上させることが可能となる。更に、内視鏡画像の客観的な指標となる数値を多くの医師に提示することが可能となり、見え方の異なる膨大な内視鏡の画像データに対しても、腫瘍のスクリーニングに対しても有効に利用することが期待できる。このように、リアルタイム診断支援技術の提供だけでなく、膨大な臨床データから得られる新しい知見を学習データとして蓄積することで、CADシステムの精度向上のための有効利用が可能となる。

Research Products

• [Journal Article] Classification with CNN features and SVM on Embedded DSP Core for Colorectal Magnified NBI Endoscopic Video Image (2022)
  • Author(s): Masayuki Odagawa, Takumi Okamoto, Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Hiroshi Mieno, Shinji Tanaka
  • Journal Title: IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences Volume: E105.A Issue: 1 Pages: 25-34
  • DOI: 10.1587/transfun.2021EAP1036
  • NAID: 130008138760
  • ISSN: 0916-8508, 1745-1337
  • Year and Date: 2022-01-01
  • Peer Reviewed
• [Journal Article] Feasibility Study for Computer-Aided Diagnosis System with Navigation Function of Clear Region for Real-Time Endoscopic Video Image on Customizable Embedded DSP Cores (2022)
  • Author(s): Masayuki Odagawa, Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Hiroshi Mieno, Shinji Tanaka
  • Journal Title: IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences Volume: E105.A Issue: 1 Pages: 58-62
  • DOI: 10.1587/transfun.2021EAL2044
  • NAID: 130008138735
  • ISSN: 0916-8508, 1745-1337
  • Year and Date: 2022-01-01
  • Peer Reviewed
• [Journal Article] Development of multi-class computer-aided diagnostic systems using the NICE/JNET classifications for colorectal lesions (2022)
  • Author(s): Yuki Okamoto, Shigeto Yoshida, Seiji Izakura, Daisuke Katayama, Ryuichi Michida, Tetsushi Koide, Toru Tamaki, Yuki Kamigaichi, Hirosato Tamari, Yasutsugu Shimohara, Tomoyuki Nishimura, Katsuaki Inagaki, Hidenori Tanaka, Ken Yamashita, Kyoku Sumimoto, Shiro Oka, Shinji Tanaka
  • Journal Title: Journal of Gastroenterology and Hepatology Volume: 37 Issue: 1 Pages: 104-110
  • DOI: 10.1111/jgh.15682
  • Peer Reviewed / Open Access
• [Journal Article] A Hardware Implementation on Customizable Embedded DSP Core for Colorectal Tumor Classification with Endoscopic Video toward Real-Time Computer-Aided Diagnosais System (2021)
  • Author(s): Masayuki Odagawa, Takumi Okamoto, Tetsushi Koide, Toru Tamaki, Bisser Raytchev, Kazufumi Kaneda, Shigeto Yoshida, Hiroshi Mieno, Shinji Tanaka, Takayuki Sugawara, Hiroshi Toishi, Masayuki Tsuji, Nobuo Tamba
  • Journal Title: IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences Volume: E104.A Issue: 4 Pages: 691-701
  • DOI: 10.1587/transfun.2020EAP1069
  • NAID: 130008013924
  • ISSN: 0916-8508, 1745-1337
  • Year and Date: 2021-04-01
  • Peer Reviewed / Open Access
• [Presentation] 大腸NBI内視鏡におけるコンピュータ支援診断のためのFully Convolutional Networkを用いた病変部位のセグメンテーション法 (2022)
  • Author(s): 呉 泳飛, 片山 大輔, 道田 竜一, 小出 哲士, 玉木 徹, 吉田 成人, 岡本 由貴, 岡 志郎, 田中 信治
  • Organizer: 第28回画像センシングシンポジウム(SSII2022)
• [Presentation] 深層学習を用いたNICE/JNET分類に基づく大腸内視鏡画像診断支援の一手法 (2022)
  • Author(s): 片山 大輔, 呉 泳飛, 道田 竜一, 小出 哲士, 玉木 徹, 吉田 成人, 岡本 由貴, 岡 志郎, 田中 信治
  • Organizer: 第28回画像センシングシンポジウム(SSII2022)
• [Presentation] A U-Net Based Lesion Segmentation Method for Computer-Aided Diagnosis in Colorectal NBI Endoscopy (2022)
  • Author(s): Yongfei Wu, Daisuke Katayama, Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Yuki Okamoto, Shiro Oka, Shinji Tanaka
  • Organizer: The 36th International Technical Conference on Circuits/Systems, Computers, and Communications 2022(ITC-CSCC2022)
  • Int'l Joint Research
• [Presentation] An Endoscopic image Classifier Using Deep Learning Considering Progression of Colorectal Cancer (2022)
  • Author(s): Daisuke Katayama, Yongfei Wu, Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Yuki Okamoto, Shiro Oka, Shinji Tanaka
  • Organizer: The 36th International Technical Conference on Circuits/Systems, Computers, and Communications 2022(ITC-CSCC2022)
  • Int'l Joint Research
• [Presentation] Implementation of parallel AES processing with CAM-based massive-parallel SIMD matrix core (2022)
  • Author(s): Kyosuke Kageyama, Sota Arai, Hajime Hamano, Xiangbo Kong, Takeshi Kumaki and Tetsushi Koide
  • Organizer: World Symposium on Communication Engineering (WSCE2022)
  • Int'l Joint Research
• [Presentation] An Endoscopic lesion image Classifier Using Deep Learning Considering Progression of Colorectal Cancer (2022)
  • Author(s): Daisuke Katayama, Yongfei Wu, Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Yuki Okamoto, Shiro Oka, Shin Morimoto, Shinji Tanaka, Masayuki Odagawa, Toshihiko Sugihara
  • Organizer: The 7th International Symposium on Biomedical Engineering
  • Int'l Joint Research
• [Presentation] A Lesion Segmentation Method for Computer-Aided Diagnosis System in Colorectal NBI Endoscopy Based on U-Net (2022)
  • Author(s): Yongfei Wu, Daisuke Katayama, Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Yuki Okamoto, Shiro Oka, Shin Morimoto, Shinji Tanaka, Masayuki Odagawa, Toshihiko Sugihara
  • Organizer: The 7th International Symposium on Biomedical Engineering
  • Int'l Joint Research
• [Presentation] Practical Improvement to the Real-Time CAD (Computer-Aided Diagnosis) System with Navigation Function (2022)
  • Author(s): Masayuki Odagawa,Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Yuki Okamoto, Shin Morimoto, Shiro Oka, Shinji Tanaka
  • Organizer: The 7th International Symposium on Biomedical Engineering
  • Int'l Joint Research
• [Presentation] A Lesion Classification Method Using Deep Learning Based on NICE Classification for Computer-Aided Diagnosis System in Colorectal NBI Endoscopy (2021)
  • Author(s): Daisuke Katayama, Ryuichi Michida, Seiji Izakura, Yongfei Wu, Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Hiroshi Mieno, Yuki Okamoto, Shinji Tanaka
  • Organizer: The 36th International Technical Conference on Circuits/Systems, Computers, and Communications 2021(ITC-CSCC2021)
  • Int'l Joint Research
• [Presentation] A Lesion Classification Method Using Deep Learning Based on JNET Classification for Computer-Aided Diagnosis System in Colorectal Magnified NBI Endoscopy, (2021)
  • Author(s): Ryuichi Michida, Daisuke Katayama, Seiji Izakura, Yongfei Wu, Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Hiroshi Mieno, Yuki Okamoto, Shinji Tanaka
  • Organizer: The 36th International Technical Conference on Circuits/Systems, Computers, and Communications 2021(ITC-CSCC2021)
  • Int'l Joint Research
• [Presentation] Floating-point arithmetic of content addressable memory-based massive-parallel SIMD matrix core (2021)
  • Author(s): Kyosuke Kageyama, Akimitsu Hamai, Kensuke Watanabe, Tetsushi Koide, Takeshi Kumaki
  • Organizer: RISP International workshop on Nonlinear Circuit, computer and Signal Processing (NCSP2021)
  • Int'l Joint Research
• [Presentation] Development of multi-class computer-aided diagnosis support system for diagnosis of NICE/JNET classification in colorectal lesions (2021)
  • Author(s): Yuki Okamoto, Shigeto Yoshida, Seiji Izakura, Daisuke Katayama, Ryuichi Michida, Tetsushi Koide, Toru Tamaki, Yuki Kamigaichi, Hirosato Tamari, Yasutsugu Shimohara, Tomoyuki Nishimura, Katsuaki Inagaki, Hidenori Tanaka, Ken Yamashita, Kyoku Sumimoto, Yuki Ninomiya1, Shiro Oka, Shinji Tanaka
  • Organizer: UEG Week - United European Gastroenterology
  • Int'l Joint Research
• [Presentation] Multiplication of Baugh-Wooley arithmetic processing by content addressable memory-based massive-parallel SIMD matrix core (2021)
  • Author(s): Kageyama Kyosuke, Arai Sota, Hamano Hajime, Hamai Akimitsu, Kong Xiangbo, Koide Tetsushi, Kumaki Takeshi
  • Organizer: The 6th International Symposium on Biomedical Engineering
  • Int'l Joint Research
• [Presentation] A Denoising Autoencoder on Anormaly Detection via Cutmix for Computer-Aided Diagnosis System in Colorectal NBI Endoscopy (2021)
  • Author(s): Yongfei Wu, Daisuke Katayama, Ryuichi Michida, Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Yuki Okamoto, Oka Shiro, Shinji Tanaka, Masayuki Odagawa, Toshihiko Sugihara
  • Organizer: The 6th International Symposium on Biomedical Engineering
  • Int'l Joint Research
• [Presentation] A Lesion Classification Method UsiDeep Learning for Computer-Aided Diagnosis System in Colorectal NBI Endoscopy (2021)
  • Author(s): Daisuke Katayama, Yongfei Wu, Ryuichi Michida, Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Yuki Okamoto, Oka Shiro, Shinji Tanaka, Masayuki Odagawa, Toshihiko Sugihara
  • Organizer: The 6th International Symposium on Biomedical Engineering
  • Int'l Joint Research
• [Presentation] 深層学習を用いたJNET分類に基づく大腸内視鏡画像診断支援の一手法 (2021)
  • Author(s): 道田 竜一，片山 大輔，井櫻 星二，呉 泳飛，小出 哲士，玉木 徹，吉田 成人，三重 野寛，岡本 由貴，田中 信治
  • Organizer: 第27回画像センシングシンポジウム(SSII2021)
• [Presentation] 深層学習を用いたNICE分類に基づく大腸内視鏡画像診断支援の一手法 (2021)
  • Author(s): 片山 大輔，道田 竜一，井櫻 星二，呉 泳飛，小出 哲士，玉木 徹，吉田 成人，三重 野寛，岡本 由貴，田中 信治
  • Organizer: 第27回画像センシングシンポジウム(SSII2021)
• [Presentation] Proposal of content addressable memory-based massive-parallel SIMD matrix core (2020)
  • Author(s): Kyosuke Kageyama, Akira Sekino, Kensuke Watanabe, Akimitsu Hamai, Tetsushi Koide, and Takeshi Kumaki
  • Organizer: Proc. of RISP International Workshop on Nonlinear Circuit, Computer and Signal Processing (NCSP2020)
  • Int'l Joint Research
• [Presentation] Acceleration of arithmetic processing with CAM-based massive-parallel SIMD matrix core (2020)
  • Author(s): Kyosuke Kageyama, Kensuke Watanabe, Akimitsu Hamai, Takeshi Kumaki and Tetsushi Koide
  • Organizer: Proc. of IEEE International MidWest Symposium on Circuits And Systems (MWSCAS2020)
  • Int'l Joint Research
• [Presentation] Multistep Reactions by Aligned Tablet Reagents for Long Term Monitoring of Plant Culture Solution (2020)
  • Author(s): Yoko Azuma, Toshihiro Kasama, Yoshishige Endo, Tetsushi Koide, Chiharu Sone, Masashi Komine, Atsushi Ogawa, and Ryo Miyake
  • Organizer: Proc. of the 24th International Conference on Miniaturized Systems for Chemistry and Life Sciences (MicroTAS 2019)
  • Int'l Joint Research
• [Presentation] Classification Method with CNN features and SVM for Computer-Aided Diagnosis System in Colorectal Magnified NBI Endoscopy (2020)
  • Author(s): Masayuki Odagawa, Tetsushi Koide, Takumi Okamoto, Toru Tamaki, Bisser Raytchev, Kazufumi Kaneda, Shigeto Yoshida, Hiroshi Mieno, Shinji Tanaka
  • Organizer: Proc. of 2020 IEEE Region 10 Conference (TENCON2020)
  • Int'l Joint Research
• [Presentation] 機械学習を用いたリアルタイム大腸がん内視鏡診断支援システム～ ソフトウェアとハードウェアの協調設計によるリアルタイム処理の実現 ～ (2020)
  • Author(s): 小出哲士，玉木徹， 吉田成人，三重野寛，田中信治
  • Organizer: 医療機器開発のみらいをつくるシーズ，東京都医工連携HUB機構，オンラインセミナー
  • Invited
• [Book] 内視鏡画像AI（医療AIとディープラーニングシリーズ） (2022)
  • Author(s): 玉木徹, 小出哲士, 吉田成人
  • Total Pages: 250
  • Publisher: コロナ社
  • ISBN: 9784274225642
• [Book] Biomedical Engineering, Chapter 16, A Hierarchical Type Segmentation Hardware for Colorectal Endoscopic Images with Narrow Band Imaging Magnification (2021)
  • Author(s): Tetsushi Koide, Toru Tamaki, Shigeto Yoshida, Shinji Tanaka
  • Total Pages: 380
  • Publisher: Jenny Stanford Publishing Pte Ltd.
  • ISBN: 9814877638
• [Remarks] 生体医歯工学共同研究拠点
  • URL: http://www.tmd.ac.jp/ibbc/
• [Remarks] 広島大学ナノデバイス研究所
  • URL: https://www.rnbs.hiroshima-u.ac.jp/
• [Remarks] 地域オープンイノベーション拠点選抜制度（J-Innovation HUB；Jイノベ）
  • URL: https://j-innovation.meti.go.jp/universities/hiroshima-u
• [Remarks] Google Scholar: Tetsushi Koide
  • URL: https://bit.ly/3uzRnPZ
• [Remarks] 広島大学研究者総覧
  • URL: https://seeds.office.hiroshima-u.ac.jp/profile/ja.3b0f364e58607230520e17560c007669.html
• [Remarks] ResearchGate: Tetsushi Koide
  • URL: https://www.researchgate.net/profile/Tetsushi-Koide
• [Remarks] 広島大学ナノデバイス・バイオ融合科学研究所論文刊行物
  • URL: http://www.rnbs.hiroshima-u.ac.jp/papconf.html
• [Remarks] 広島大学ナノデバイス・バイオ融合科学研究所学会発表リスト
  • URL: http://www.rnbs.hiroshima-u.ac.jp/conference.html
• [Remarks] Google Scholor : Koide Tetsushi
  • URL: https://bit.ly/3uzRnPZ