AGA provides an expert commentary on the role of artificial intelligence (AI) in the evaluation and management of colorectal polyps, particularly in the context of screening colonoscopies.
Key Insights from the Article:
AI in Polyp Detection: The article discusses the use of AI in colonoscopy, focusing on computer-aided detection (CADe) and computer-aided diagnosis (CADx). CADe assists endoscopists in detecting polyps during colonoscopy, while CADx aims to accurately predict polyp histology without needing a tissue biopsy.
Variability in Adenoma Detection Rates (ADRs): ADRs, which measure the rate at which clinicians detect adenomas during colonoscopies, vary significantly among endoscopists, ranging from 7% to 53%. AI tools have the potential to reduce this variability by aiding in polyp detection.
Concerns and Recommendations: The main concern with AI-assisted colonoscopy is the risk of false-positive alerts. The authors recommend a collaborative approach between clinicians and AI technologies when using CADe and CADx. They also suggest formalized training for endoscopists to ensure effective and enhancing use of AI tools in patient care.
Future of AI in Colonoscopy: The authors predict that AI tools will become an indispensable suite for colonoscopy, serving as a powerful adjunct to support safe and efficient clinical practice. They anticipate that AI tools improving colonoscopy quality may gain acceptance and demand from payors, administrators, and informed patients who seek the highest-quality examination of their colon.