Development of automated diagnostic system for colonoscopy by the collaborative research between medical and engineering team.

2016 Fiscal Year Final Research Report

• Project/Area Number: 15K19351
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Gastroenterology
• Research Institution: Showa University
• Principal Investigator: Mori Yuichi 昭和大学, 医学部, 助教 (20459209)
• Research Collaborator: Mori Kensaku 名古屋大学, 情報連携統括本部 ; Kudo Shin-ei 昭和大学, 医学部, 教授 ; Misawa Masashi 昭和大学, 医学部, 助教 ; Takeda Kenichi 昭和大学, 医学部, 助教
• Project Period (FY): 2015-04-01 – 2017-03-31
• Keywords: 人工知能 / 大腸ポリープ / 大腸癌 / 機械学習

Outline of Final Research Achievements

We had investigated on automated diagnostic system for colorectal lesions by using endocytoscopy (=ultra-magnifying endoscopy) in the previous study to realize objective diagnosis. In the current research, we aimed to achieve excellent diagnostic performance by initiating the collaboration with the engineers who specialized in information science. We evaluated the developed system by conducting the international web test. The system provided the accuracy of 89% for differentiation of adenomas while the experts showed the accuracy of 90%. (Mori Y, et al. Endoscopy 2016 48: 1110-1118)

Free Research Field

消化器内科