When you have screening colonoscopy should you worry about the endoscopist’s skills? Elizabeth Tracey reports
Update: 2025-10-20
Description
AI assisted colonoscopy, where a computer helped interpret images seen during the procedure, resulted in endoscopists being less adept at recognizing precancerous lesions known as polyps, a recent study finds. Kimmel Cancer Center director William Nelson at Johns Hopkins comments.
Nelson: They've got pretty used to using it in this group, and so what the study was was they looked at a cohort of about 1400 subjects who underwent colonoscopy that wasn't computer aided, both before and after the introduction of the AI tool, tried to look at what the polyp detection rate was. What they saw was that it fell from 38.4% to 22.4%. Before they were using the AI tool they detected polyps more readily than after. :30
Nelson says he isn’t concerned about this outcome since when AI assistance is available it is utilized to achieve the best result, and if it isn’t available endoscopists won’t learn to rely on it, and notes that technologies are increasingly ubiquitous and largely beneficial. At Johns Hopkins, I’m Elizabeth Tracey.
Nelson: They've got pretty used to using it in this group, and so what the study was was they looked at a cohort of about 1400 subjects who underwent colonoscopy that wasn't computer aided, both before and after the introduction of the AI tool, tried to look at what the polyp detection rate was. What they saw was that it fell from 38.4% to 22.4%. Before they were using the AI tool they detected polyps more readily than after. :30
Nelson says he isn’t concerned about this outcome since when AI assistance is available it is utilized to achieve the best result, and if it isn’t available endoscopists won’t learn to rely on it, and notes that technologies are increasingly ubiquitous and largely beneficial. At Johns Hopkins, I’m Elizabeth Tracey.
Comments
In Channel
Download from Google Play
Download from App Store
United States00:00
00:00
x
