• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

High-performance diagnostic and prognostic model using artificial intelligence by integration of functional imaging and clinical information

Research Project

Project/Area Number 20K07990
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 52040:Radiological sciences-related
Research InstitutionKanazawa University

Principal Investigator

Nakajima Kenichi  金沢大学, 先進予防医学研究科, 特任教授 (00167545)

Project Period (FY) 2020-04-01 – 2023-03-31
Project Status Completed (Fiscal Year 2022)
Budget Amount *help
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2022: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2021: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2020: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Keywords人工知能 / 画像データベース / 心不全 / リスクモデル / 多施設研究 / 脳受容体画像 / 神経変性疾患
Outline of Research at the Start

機能画像と臨床情報の統合データベースを用いて機械学習あるいはディープラーニングにより診断および予後を推定するシステムを作成する。画像データと臨床情報ともに本施設および多施設での研究を組織しデータベースの収集を行う。このデータを元に、画像データあるいはそれに由来する機能指標を用いて機械学習を行い、可能性のある診断を確率的に表示する。また、予後については重症化や死亡率などの短期長期予後を推定するような機械学習のトレーニングと検証も行う。臨床情報を含めた統合的な情報によりさらに診断率、あるいは予後予測の改善を図る。最終的には、臨床画像と情報を入力して診断を行うコンピュータ支援診断方法を確立する。

Outline of Final Research Achievements

The aim of this study was to create artificial intelligence models for diagnosis and estimation of prognosis by integrating medical images and clinical information. In cardiology, databases of more than 1000 patients were generated for chronic heart failure from Japan and Europe. A risk model for predicting modes of cardiac death, namely heart-failure death and sudden cardiac death/arrhythmic death, was created by using machine learning in patients with chronic heart failure. A three-dimensional quantification method with cardiac sympathetic nerve imaging was also made by deep learning and 123I-MIBG SPECT. In neurological patients with parkinsonism and Lewy-body diseases, a machine learning based diagnostic method for abnormality of brain dopamine transporter function was created. The integrated method of functional images and clinical information was proved to be useful for diagnosis and prognostic analysis.

Academic Significance and Societal Importance of the Research Achievements

診断と予後の推定は医療において重要な課題であるが、従来の統計的手法のみでは十分に達成できなかった。特に機能画像を用いた診断においては、単に疾患名の診断だけではなく予後診断も求められており、患者の背景も利用した統合的なデータベースとそれに基づいた解析が必要で人工知能の利用が期待される。本研究では、機械学習や深層学習を用いることにより機能画像を解析して臓器を抽出し、特定の機能指標を算出し、それを臨床情報と統合する方法の妥当性が明らかになった。心不全での死亡原因を推定すること、交感神経機能を3次元的に定量すること、神経疾患の診断分類をすることなど、さまざまな領域での適用が可能であることを示した。

Report

(4 results)
  • 2022 Annual Research Report   Final Research Report ( PDF )
  • 2021 Research-status Report
  • 2020 Research-status Report
  • Research Products

    (20 results)

All 2022 2021 2020

All Journal Article (7 results) (of which Int'l Joint Research: 3 results,  Peer Reviewed: 7 results,  Open Access: 7 results) Presentation (12 results) (of which Int'l Joint Research: 11 results,  Invited: 7 results) Patent(Industrial Property Rights) (1 results)

  • [Journal Article] Nuclear Cardiology Data Analyzed Using Machine Learning2022

    • Author(s)
      Nakajima K, Maruyama K
    • Journal Title

      Annals of Nuclear Cardiology

      Volume: 8 Issue: 1 Pages: 80-85

    • DOI

      10.17996/anc.22-00164

    • ISSN
      2189-3926, 2424-1741
    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Diagnosis of Parkinson syndrome and Lewy-body disease using 123I-ioflupane images and a model with image features based on machine learning2022

    • Author(s)
      Nakajima K, Saito S, Chen Z, Komatsu J, Maruyama K, Shirasaki N, Watanabe S, Inaki A, Ono K, Kinuya S
    • Journal Title

      Ann Nucl Med

      Volume: 36 Issue: 8 Pages: 765-776

    • DOI

      10.1007/s12149-022-01759-z

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Methods of calculating 123I-methyl-P-iodophenyl-pentadecanoic acid washout rates in triglyceride deposit cardiomyovasculopathy2022

    • Author(s)
      Chen Z, Nakajima K, Hirano K, Kamiya T, Yoshida S, Saito S, Kinuya S
    • Journal Title

      Ann Nucl Med

      Volume: 36 Issue: 11 Pages: 986-997

    • DOI

      10.1007/s12149-022-01787-9

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Prediction of multivessel coronary artery disease and candidates for stress-only imaging using multivariable models with myocardial perfusion imaging2022

    • Author(s)
      Kunita Y, Nakajima K, Nakata T, Kudo T, Kinuya S
    • Journal Title

      Ann Nucl Med

      Volume: 36 Issue: 7 Pages: 674-683

    • DOI

      10.1007/s12149-022-01751-7

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Convolutional neural network-based automatic heart segmentation and quantitation in 123I-metaiodobenzylguanidine SPECT imaging2021

    • Author(s)
      Saito S, Nakajima K, Edenbrandt L, Enqvist O, Ulen J, Kinuya S
    • Journal Title

      EJNMMI Res

      Volume: 11 Issue: 1

    • DOI

      10.1186/s13550-021-00847-x

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] The relation between cardiac 123I-mIBG scintigraphy and functional response 1 year after CRT implantation2021

    • Author(s)
      Verschure DO, Poel E, De Vincentis G, Frantellizzi V, Nakajima K, Gheysens O, de Groot JR, Verberne HJ.
    • Journal Title

      Eur Heart J Cardiovasc Imaging

      Volume: 22 Issue: 1 Pages: 49-57

    • DOI

      10.1093/ehjci/jeaa045

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Machine learning-based risk model using 123I-metaiodobenzylguanidine to differentially predict modes of cardiac death in heart failure2020

    • Author(s)
      Nakajima Kenichi、Nakata Tomoaki、Doi Takahiro、Tada Hayato、Maruyama Koji
    • Journal Title

      Journal of Nuclear Cardiology

      Volume: - Issue: 1 Pages: 190-201

    • DOI

      10.1007/s12350-020-02173-6

    • Related Report
      2022 Annual Research Report 2021 Research-status Report 2020 Research-status Report
    • Peer Reviewed / Open Access
  • [Presentation] Progress of Nuclear Cardiology in Japan: 2022 Updates2022

    • Author(s)
      Nakajima, Kenichi
    • Organizer
      World Federation of Nuclear Medicine and Biology
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] Standardized MIBG in clinical practice of cardiology and neurology2022

    • Author(s)
      Nakajima , Kenichi
    • Organizer
      Joint Symposium Nuclear Cardiology and Neurology, Taiwan
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] Creation of machine learning based classifiers for interpretation of I-123 Ioflupane images2021

    • Author(s)
      Nakajima K, Saito S, Chen ZC, Komatsu J, Inaki A, Watanabe S, Kinuya S
    • Organizer
      Annual Meeting of Society of Nuclear Medicine and Molecular Imaging 2021
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Creation of mortality risk calculator using a I-123 mIBG-based machine learning model: differential prediction of arrhythmic death and heart-failure death2021

    • Author(s)
      Nakajima K, Nakata T, Doi T, Tada H, Saito S, Maruyama
    • Organizer
      International Conference of Nuclear Cardiology and Cardiac CT
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Machine learning: Can we predict and prevent sudden cardiac death in heart failure?2021

    • Author(s)
      Nakajima K
    • Organizer
      European Society of Cardiology Congress 2021
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research / Invited
  • [Presentation] Differential prediction of sudden cardiac death and heart failure death using machine learning and nuclear cardiology2021

    • Author(s)
      Nakajima K
    • Organizer
      European Society of Cardiology Congress 2021
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research / Invited
  • [Presentation] Improvement of myocardial perfusion images with data-driven corrections of motion and respiration: Comparison between retrospective and prospective gating methods2021

    • Author(s)
      Nakajima K, Sibutani T, Massanes F, Vija AH, Shimizu T, Yoshida S, Yoneyama H, Onoguchi M, Kinuya S
    • Organizer
      34th Annual Congress of the European Association of Nuclear Medicine
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] Cardiac SPECT imaging of cardiomyopathy2021

    • Author(s)
      Nakajima K
    • Organizer
      IAEA Nuclear Cardiology workshop
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research / Invited
  • [Presentation] Nuclear Cardiology Beyond Perfusion Imaging: Imaging innervation with 123I-MIBG2020

    • Author(s)
      Nakajima K
    • Organizer
      Annual Meeting of European Association of Nuclear Medicine
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research / Invited
  • [Presentation] Application of convolutional neural network to123I-MIBG SPECT imaging: automatic quantitation vs. manual measurements2020

    • Author(s)
      Saito S, Nakajima K, Edenbrandt L, et al.
    • Organizer
      Annual Meeting of Society of Nuclear Medicine and Molecular Imaging 2020
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Let us use the same measures and diagnostic thresholds for 123I-MIBG imaging2020

    • Author(s)
      Nakajima K
    • Organizer
      International Conference on Clinical and Functional Image for Neurodegenerative Disorders of Society of Nuclear Medicine
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research / Invited
  • [Presentation] Japanese cardiac sarcoidosis prognostic study (J-CASP) registry: Comparison between clinical and histological groups2020

    • Author(s)
      Nakajima K, Nakata T, Naya M, et al.
    • Organizer
      Annual Meeting of Society of Nuclear Medicine and Molecular Imaging 2020
    • Related Report
      2020 Research-status Report
  • [Patent(Industrial Property Rights)] プログラム、情報処理装置及び情報処理方法2021

    • Inventor(s)
      中嶋憲一、北村千枝美
    • Industrial Property Rights Holder
      中嶋憲一、北村千枝美
    • Industrial Property Rights Type
      特許
    • Industrial Property Number
      2021-080533
    • Filing Date
      2021
    • Related Report
      2021 Research-status Report

URL: 

Published: 2020-04-28   Modified: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi