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

2019 Fiscal Year Final Research Report

Next-generation computer-aided diagnosis for early detection of dementia using person's genotype and radiomic features

Research Project

  • PDF
Project/Area Number 17K09067
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Research Field Medical Physics and Radiological Technology
Research InstitutionKumamoto University

Principal Investigator

Yoshikazu Uchiyama  熊本大学, 大学院生命科学研究部(保), 准教授 (50325172)

Project Period (FY) 2017-04-01 – 2020-03-31
Keywordsアルツハイマー型認知症 / 遺伝子 / MR画像 / コンピュータ支援診断 / Radiogenomics
Outline of Final Research Achievements

We constructed next-generation computer-aided diagnosis by the integration analysis of genetic and image tests. We found that the process of disease formation is different in mild cognitive impairment and Alzheimer’s diseases (AD) according to the APOE genotype. Therefore, the early detection of AD would be possible by the image interpretation considering patient’s genotype. We also developed a method for distinguishing normal patients from ADs on the eigenspace by creating an eigenspace from normal MR images of 60, 70 and 80 ages. The proposed method made it possible to quantitatively evaluate the degree of cerebral atrophy considering the effects of normal aging. Since this method represents cerebral atrophy in a low-dimensional eigenspace, it has the advantage of avoiding the multiple test problem without setting a region of interest at a specific part of the brain.

Free Research Field

医用画像処理・認識,コンピュータ支援診断,Radiogenomics

Academic Significance and Societal Importance of the Research Achievements

遺伝子検査と画像検査の統合解析による次世代型のコンピュータ支援診断の新しい概念の研究を構築した点で学術的意義は大きい.この新しい概念は,認知症以外の他疾患にも適用可能であり,関連分野の進展に貢献できる.遺伝学的検査によって個人の遺伝型が特定できれば,認知症になりやすい患者群を特定することができる.そのような患者群は,発症を早期に発見するために,定期的な検査が行われるであろう.定期検査の際に,その患者の遺伝型に関係する画像特徴の変化(脳萎縮)が特定できれば,その変化に注目して読影を行うことができるため,現在よりも早期に病気を検出できる可能性が高く,社会的意義も大きい.

URL: 

Published: 2021-02-19  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi