Automated grading system for the retinal arteriolosclerosis in a cardiology risk screening

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K10662
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 58030:Hygiene and public health-related: excluding laboratory approach
• Research Institution: Osaka University
• Principal Investigator: Kawasaki Ryo 大阪大学, 医学系研究科, 特任教授(常勤) (70301067)
• Co-Investigator(Kenkyū-buntansha): 大久保 孝義 帝京大学, 医学部, 教授 (60344652) ; 中島 悠太 大阪大学, データビリティフロンティア機構, 准教授 (70633551)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 眼底写真 / 人工知能 / 健康診断 / 動脈硬化

Outline of Final Research Achievements

"The eye is the window to the entire body," and fundus findings have long been used in medical examinations to assess the risk of cardiovascular disease. The evaluation method for "retinal arteriolosclerosis findings" (Scheie: S findings) has relied on visual inspection. In this study, we developed a system to evaluate retinal arteriolosclerosis findings. Using a pipeline consisting of multiple deep learning models, we obtained a precision of 98.1% and a recall of 89.7% for the extraction of crossings through the extraction of blood vessels from fundus images. The estimated severity of retinal artery crossing phenomenon was 0.61 in Kappa value and 0.77 in agreement. Through this study, we were able to construct an automatic determination system while reproducing the process of a physician's determination of fundus photographs.

Free Research Field

疫学

Academic Significance and Societal Importance of the Research Achievements

特定健康診査において循環器リスク評価のための眼底検査の潜在的な対象人数は約2600万人にも上る。このような対象者に、迅速かつ再現性の高い眼底評価方法で循環器病リスク評価に貢献する眼底指標を提供することは我が国の循環器病予防に貢献する可能性がある。