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

Development and clinical evaluation of fast quantitative susceptibility mapping for the liver using deep learning and compressed sensing

Research Project

Project/Area Number 20K16755
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 52040:Radiological sciences-related
Research InstitutionHamamatsu University School of Medicine (2022)
University of Yamanashi (2020-2021)

Principal Investigator

Funayama Satoshi  浜松医科大学, 医学部附属病院, 診療助教 (40790449)

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: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
Fiscal Year 2021: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2020: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
KeywordsMRI / 肝臓 / 深層学習 / 画像再構成 / 定量的磁化率マッピング / QSM / 人工知能 / 機械学習
Outline of Research at the Start

定量的磁化率マッピングは組織内の磁化率分布をMRIによって定量的に画像化する全く新しいMRIコントラストである。頭部領域で応用が進んでおり、これまでMRIでは鑑別の難しかった石灰化と血腫の鑑別等に成功している。腹部領域でも少数の報告がみられるが、頭部とは異なり、磁化率測定の大きな交絡因子である脂肪信号を分離する必要性から撮像時間や空間分解能に課題を抱えており、肝での応用を阻む要因の1つとなっている。そこで本研究では、深層学習と圧縮センシングを融合させた超高速撮像を用いれば肝磁化率マッピングの撮像時間や空間分解能を向上させ、肝腫瘍の診断能向上やびまん性肝疾患の病態解析を行うことを目指す。

Outline of Final Research Achievements

Clinically obtained liver MRI images were collected retrospectively to develop a fast MRI image reconstruction method that combines deep learning and compressed sensing. Basic image quality evaluation of the images reconstructed by the developed method and image quality evaluation by a diagnostic radiologist were performed. The images reconstructed by the developed method showed superior image quality compared to the conventional method. Liver quantitative susceptibility mapping images were acquired by the developed method, but the clinical usefulness of the liver susceptibility images could not be found within the study period.

Academic Significance and Societal Importance of the Research Achievements

深層学習と圧縮センシングを融合した画像再構成手法は従来法に比較して少ない情報量から良好な画質を示すことが明らかとなった。より短時間の呼吸停止による腹部MRIが実現できる可能性がある。

Report

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

    (5 results)

All 2023 2022 2021 2020

All Journal Article (1 results) (of which Peer Reviewed: 1 results,  Open Access: 1 results) Presentation (4 results) (of which Int'l Joint Research: 1 results,  Invited: 2 results)

  • [Journal Article] Model-based Deep Learning Reconstruction Using a Folded Image Training Strategy for Abdominal 3D T1-weighted Imaging2023

    • Author(s)
      Funayama S, Motosugi U, Ichikawa S, Morisaka H, Omiya Y, Onishi H
    • Journal Title

      Magnetic Resonance in Medical Sciences

      Volume: 22 Issue: 4 Pages: 515-526

    • DOI

      10.2463/mrms.mp.2021-0103

    • ISSN
      1347-3182, 1880-2206
    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Presentation] Meet the teacher13: AI Hands-on Let's try your AI in the MR image processing2022

    • Author(s)
      Satoshi Funayama
    • Organizer
      The 50th annual meeting of the japanese society for magnetic resonance in medicine
    • Related Report
      2022 Annual Research Report
    • Invited
  • [Presentation] AI Zoo in Diagnostic Radiology2022

    • Author(s)
      Satoshi Funayama
    • Organizer
      FCA webinar in 東海
    • Related Report
      2022 Annual Research Report
    • Invited
  • [Presentation] Model-based deep learning reconstruction using folded image training strategy (FITS) for abdominal 3D T1-weighted images2021

    • Author(s)
      Satoshi Funayama, Utaroh Motosugi, Shintaro Ichikawa, Hiroyuki Morisaka, Yoshie Omiya, Hiroshi Onishi
    • Organizer
      第49回日本磁気共鳴医学会大会
    • Related Report
      2021 Research-status Report
  • [Presentation] FITs-CNN: A Very Deep Cascaded Convolutional Neural Networks Using Folded Image Training Strategy for Abdominal MRI Reconstruction2020

    • Author(s)
      Satoshi Funayama, Tetsuya Wakayama, Hiroshi Onishi, and Utaroh Motosugi
    • Organizer
      ISMRM2020
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research

URL: 

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

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi