Novel Diagnostics in Colorectal Tumors Using Artificial Intelligence and the Genetic Nature of Rapidly Developing Cancer

2022 Fiscal Year Final Research Report

• Project/Area Number: 21K20881
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 0902:General internal medicine and related fields
• Research Institution: Tokyo Medical University
• Principal Investigator: Yamaguchi Hayato 東京医科大学, 医学部, 助教 (90617721)
• Project Period (FY): 2021-08-30 – 2023-03-31
• Keywords: Interval cancer / 大腸癌 / 人工知能 / AI

Outline of Final Research Achievements

The average ADR, PDR, and SDR of the endoscopists were 31.3%, 68.3%, and 12.5%, respectively, for colonoscopy using automatic lesion detection support software based on artificial intelligence. Among patients with interval cancer, the >T1 cancers group had significantly more non-polypoid growth (NPG)-type cases than PG-type colorectal cancer cases, and some interval cancers grew rapidly from SSA/P in the right colon. This study suggests that if AI can identify true precursor lesions that should be endoscopically resected, it will provide new insights into rapidly growing cancer and have a ripple effect beyond the treatment of colorectal cancer.

Free Research Field

消化器内科学分野

Academic Significance and Societal Importance of the Research Achievements

国際的に内視鏡検査精度が向上する一方で、短期間での検査間隔における急速発育癌の発生に関わる分子学的背景は未だ不透明である。人工知能（AI）を用いた大腸内視鏡診断支援システムは高精度かつリアルタイムでの病理診断予測を可能とした革新的な新規ツールの位置づけにあるが、遺伝子分野への応用例はない。本研究では、前癌病変に対するAI診断の基礎的研究を完成し、急速発育癌のバイオマーカー、遺伝子解析の全貌を解明する。この成果で、AIを基軸とした内視鏡診断、遺伝子診断を融合させた新規診断学を確立し、さらなる大腸癌減少への新たな波及効果を得たい。