Artificial Intelligence Boosts Colonoscopy Cancer Detection

FRIDAY, Oct. 1, 2021 (HealthDay News) — Computer-assisted colonoscopies reduce rates of missed lesions, according to a study published online Sept. 13 in Clinical Gastroenterology and Hepatology.

Jeremy R. Glissen Brown, M.D., from Beth Israel Deaconess Medical Center in Boston, and colleagues randomly assigned 223 patients presenting for colorectal cancer screening or surveillance to either artificial intelligence-based computer-aided polyp detection (CADe) colonoscopy first or high-definition white light (HDWL) colonoscopy first, followed immediately by the other procedure in tandem fashion by the same endoscopist.

The researchers found that the adenoma miss rate was lower in the CADe-first group versus HDWL-first group (20.12 versus 31.25 percent; odds ratio, 1.8048). The sessile serrated lesion miss rate also was lower in the CADe-first group (7.14 versus 42.11 percent). Furthermore, the CADe-first group had a higher rate of first-pass adenomas per colonoscopy (1.19 versus 0.90) and first-pass adenoma detection rate (50.44 versus 43.64 percent).

“Our study demonstrates that computer-aided polyp detection has the potential to decrease variability in colonoscopy quality among providers by reducing the miss rate even for experienced physicians,” a coauthor said in a statement.

Two authors disclosed financial ties to the medical imaging technology industries.

Abstract/Full Text