AI fails to improve colonoscopic detection of advanced neoplasias

Colonoscopy assisted by artificial intelligence (AI) does not seem to enhance the detection of advanced colorectal neoplasias, according to a study.

This multicentre, parallel, randomized controlled trial examined the contribution of computer-aided detection to colonoscopic detection of advanced colorectal neoplasias, as well as adenomas, serrated polyps, and nonpolypoid and right-sided lesions. A total of 3,213 individuals with a positive foecal immunochemical test were randomized to colonoscopy with or without AI detection.

Advanced colorectal neoplasia referred to advanced adenoma and/or advanced serrated polyp, according to the investigators.

No significant between-group difference was seen in the rate of advanced colorectal neoplasia detection (34.8 percent with intervention vs 34.6 percent for controls; adjusted risk ratio [aRR], 1.01, 95 percent confidence interval [CI], 0.92‒1.10) or the mean number of advanced colorectal neoplasias detected per colonoscopy (0.54 vs 0.52; aRR, 1.04, 99.9 percent CI, 0.88‒1.22).

Likewise, no difference was noted in the rate of adenoma detection (64.2 percent vs 62.0 percent; aRR, 1.06, 99.9 percent CI, 0.81‒1.23).

On the other hand, computer-aided detection increased the mean number of nonpolypoid lesions (0.56 vs 0.47; aRR, 1.19, 99.9 percent CI, 1.01‒1.41), proximal adenomas (0.94 vs 0.81; aRR, 1.17, 99.9 percent CI, 1.03‒1.33), and lesions of 5 mm or smaller (polyps in general and adenomas and serrated lesions in particular) detected per colonoscopy.

“The high adenoma detection rate in the control group may limit the generalizability of the findings to endoscopists with low detection rates,” the investigators said.