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

2019 Fiscal Year Final Research Report

Development of Radiomics for molecular diagnosis of gliomas

Research Project

  • PDF
Project/Area Number 16K10778
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Research Field Neurosurgery
Research InstitutionOsaka University (2018-2019)
Osaka International Cancer Institute (2016-2017)

Principal Investigator

Kinoshita Manabu  大阪大学, 医学系研究科, 講師 (40448064)

Co-Investigator(Kenkyū-buntansha) 吉岡 芳親  大阪大学, 生命機能研究科, 特任教授(常勤) (00174897)
有田 英之  大阪大学, 医学系研究科, 招へい教員 (60570570)
金村 米博  独立行政法人国立病院機構大阪医療センター(臨床研究センター), その他部局等, 研究員 (80344175)
橋本 直哉  京都府立医科大学, 医学(系)研究科(研究院), 教授 (90315945)
Project Period (FY) 2016-04-01 – 2020-03-31
Keywords神経膠腫 / MRI / Radiomics
Outline of Final Research Achievements

We were able to develop a radiomics analyzing system for gliomas. This system allows the user to perform a radiomics analysis of gliomas quickly. Using the method developed by ourselves, we performed radiomics for 169 cases of lower-grade gliomas. We proved that the IDH mutation status of the tumor was predicted with a diagnostic accuracy of higher than 80%. Diagnostic accuracy of discriminating lower-grade gliomas into IDH mutant, IDH mutant with 1p19q codeletion and IDH wildtype was still as low as 56%. When deep learning algorism was introduced into radiomics analysis, diagnostic accuracy for identifying three subgroups of lower-grade glioma improved by 10%. Finally, we performed radiomics for 162 glioblastoma cases and identified imaging-based prognostic biomarkers independent of MGMT promoter methylation status.

Free Research Field

脳神経外科

Academic Significance and Societal Importance of the Research Achievements

本研究テーマを遂行することにより、①社会実装可能な脳腫瘍に対するAI支援下非侵襲分子診断技術が開発され、さらに②大規模臨床試験など臨床医学における大規模データを有効に活用することにより、脳腫瘍に限らず他の体幹部がん疾患でも放射線画像と遺伝子情報を中心としたビッグデータによる汎用性の高い実用的なAI支援下radiogenomics診断技術開発が進むことが期待される。

URL: 

Published: 2021-02-19  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi