Implementation of Artificial Intelligence in Colonoscopy Practice in Japan

Area of Science:

• Medical Technology
• Gastroenterology
• Artificial Intelligence

Background:

• Artificial intelligence (AI) is increasingly integrated into medical procedures, including colonoscopy.
• AI tools like computer-aided detection (CADe) and computer-aided characterization (CADx) analyze colonoscopy videos in real-time.
• These technologies aim to improve clinical outcomes, such as the adenoma detection rate.

Purpose of the Study:

• To review the implementation, processes, and challenges of AI in colonoscopy.
• To highlight the collaborative efforts between medical and computer science researchers.
• To discuss regulatory approval and reimbursement as critical factors for AI adoption.

Main Methods:

• Review of AI implementation in colonoscopy procedures.
• Analysis of the roles of CADe and CADx in real-time video analysis.
• Case study of obtaining regulatory approval for an AI tool (EndoBRAIN) in Japan.

Main Results:

• Regulatory approval is mandatory for AI tools in healthcare due to potential risks.
• Gaining regulatory approval involves establishing examination criteria and performance levels.
• Reimbursement, particularly for CADe tools in Japan from 2024, is expected to accelerate AI implementation.

Conclusions:

• Widespread adoption of AI in colonoscopy depends on regulatory approval and reimbursement.
• Concerns remain regarding the balance of benefits and harms, impact on cancer prevention, and effectiveness across diverse populations.
• Further research and clinical guidelines are necessary for rigorous assessment and optimal adoption of AI in colonoscopy practice to improve patient care.