The Impact of Computer-aided Detection Technology in Adenoma Detection Rate Among Experienced Endoscopists in the Community Setting.

Abstract

Adenoma detection rate (ADR) is a key quality metric in screening colonoscopies. An adenoma detection rate of greater than 30% reduces the incidence of colorectal carcinoma (CRC). Furthermore, studies have demonstrated an inverse relationship between ADR and the incidence of CRC. Computer aided detection (CAD) can improve ADR, but these studies have largely been in major medical centers. In this retrospective single center observational study, screening colonoscopies in average-risk patients were compared among 5 experienced endoscopists in the year before and the year after implementation of the CAD (GI Genius). Training for GI Genius was completed in December 2021 and the technology was implemented the beginning of January 2022. We evaluated the adenoma detection rate (ADR) for 1838 screening colonoscopies in 2021 (before CAD incorporation) and 2629 screening colonoscopies in 2022 (after CAD incorporation) to assess efficacy of AI-assisted colonoscopy. Our study demonstrates that the incorporation of CAD technology in a group of experienced endoscopists in a community setting significantly improved ADR. The ADR of the entire group increased significantly (p < 0.05) following the implementation of CAD technology. The improvement in ADR was attributed to an increased detection of small (<6 mm) polyps. The clinical significance of improved detection of small polyps is uncertain, and further investigation should be done on the economical benefit of incorporating an AI model in the community setting.