Artificial Intelligence Program Increases Adenoma Detection Rate in Randomized Controlled Trial by 14%

Artificial Intelligence Program Increases Adenoma Detection Rate in Randomized Controlled Trial by 14%

Douglas K. Rex, MD, FASGE, reviewing Repici A, et al. Gastroenterology 2020 May 1.

In a randomized controlled trial conducted at 3 Italian centers with 6 experienced endoscopists and 685 patients with a mix of screening, surveillance, and fecal immunochemical test-positive indications, an artificial intelligence program increased the adenoma detection rate (ADR) for nonpolypoid adenomas from 18.4% to 26.6% and for polypoid lesions from 26.5% to 37.3%. There were gains in detection of both diminutive (1-5 mm) and small (6-9 mm) lesions. There was an increase in the ADR for both the proximal and distal colon and an increase in adenomas per colonoscopy. The detection of sessile serrated lesions was not increased.

Douglas K. Rex, MD, FASGE

COMMENT

Several real-time detection trials have now been positive for an increase in the ADR. This study demonstrated substantial detection gains in a high-prevalence population and included high-detecting endoscopists. Three detection programs have now received the CE mark in Europe, but none have been approved yet by the U.S. Food and Drug Administration. We enthusiastically look forward to the availability of these programs, though the cost remains uncertain.

Note to readers: At the time we reviewed this paper, its publisher noted that it was not in final form and that subsequent changes might be made.

CITATION(S)

Repici A, Badalamenti M, Maselli R, et al. Efficacy of real-time computer-aided detection of colorectal neoplasia in a randomized trial. Gastroenterology 2020 Apr 22. (Epub ahead of print) (https://doi.org/10.1053/j.gastro.2020.04.062)

Nach oben scrollen