Computer-aided Detection Failed in Pragmatic Trial in Clinical Practice
Douglas K. Rex, MD, MASGE, reviewing Ladabaum U, et al. Gastroenterology 2022 Dec 16.
Computer-aided detection (CADe) programs are expensive to implement in clinical practice; therefore, endoscopy groups often want to trial the programs before purchase or lease.
In this study, a group reports its pragmatic trial results, and the results were negative. The group, which performs endoscopy in 6 centers, trialed CADe for 3 months at the largest site, with the other 5 units serving as control sites. This design controlled for a period effect independent of CADe use.
At the CADe site, the adenoma detection rate and adenomas per colonoscopy were 40.1% and 0.78, respectively, during the 3-month implementation period, compared with 41.8% and 0.89 in the preimplementation period. Similarly, there were no differences in these time periods at the control sites.
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.
Ladabaum U, Shepard J, Weng Y, Desai M, Singer SJ, Mannalithara A. Computer-aided detection of polyps does not improve colonoscopist performance in a pragmatic implementation trial. Gastroenterology 2022 Dec 16. (Epub ahead of print) (https://doi.org/10.1053/j.gastro.2022.12.004)