Sign In
International
Media release
From:
1. AI-assisted colonoscopy increases polyp and adenoma detection in routine screening
A systematic review of randomized clinical trials (RCTs) comparing computer-aided detection (CADe) system-enhanced colonoscopy and conventional colonoscopy found that CADe (also known as artificial Intelligence- or AI-assisted) colonoscopies may increase overall detection of colonic polyps and adenomas, or precancerous growths, with a small increase in procedure time. Equivocal results were found regarding detection of advanced colonic neoplasia (ACN), with a small increase in ACN detection rate but no difference in ACN detected per colonoscopy. The findings are published in Annals of Internal Medicine.
Researchers from Yale University and Mass General Brigham, Harvard School of Medicine comprehensively searched several large scientific research databases for RCTs comparing colonic lesion detection with standard colonoscopy versus AI-assisted colonoscopy with polyp detection (CADe) systems. The authors compared average adenoma per colonoscopy (APC) and ACN per colonoscopy for both screening methods. Secondarily, they compared adenoma detection rate (ADR), adenoma miss rate (AMR), and ACN detection rate between the two colonoscopy types. They found that AI-assisted colonoscopy found more polyps and precancerous growths in the colon than conventional colonoscopy. However, AI-assisted colonoscopy detected marginally more serious growths (ACNs) than conventional colonoscopy but was no better than the conventional method at finding ACNs per colonoscopy.
The researchers note that there are no clear differences in benefit for detecting adenomas across different AI systems for CADe, and that there was an increase in benefit for providers with lower adenoma detection rate or those without a prior fecal immunochemical test result. They suggest that future studies focus on interval post colonoscopy colorectal cancer and may consider a study design that randomizes colonoscopists rather than patients.
