DEEP LEARNING COMPUTER-AIDED POLYP DETECTION REDUCES ADENOMA MISS RATE: A U.S. MULTI-CENTER RANDOMIZED TANDEM COLONOSCOPY STUDY (CADeT-CS Trial).

BACKGROUND AND AIMS Artificial intelligence-based computer aided polyp detection (CADe) systems are intended to address the issue of missed polyps during colonoscopy. The effect of CADe during screening and surveillance colonoscopy has not previously been studied in a United States population. METHODS We conducted a prospective, multi-center, single-blind randomized tandem colonoscopy study to evaluate a deep-learning based CADe system (EndoScreener, Shanghai Wision AI, China). Patients were enrolled across four U.S. academic medical centers from 2019 through 2020. Patients presenting for colorectal cancer screening or surveillance were randomized to 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 primary outcome was adenoma miss rate (AMR), and secondary outcomes included sessile serrated lesion (SSL) miss rate and adenomas per colonoscopy (APC). RESULTS 2 32 patients entered the study, with 116 patients randomized to the AI-assisted colonoscopy first and 116 patients to HDWL colonoscopy first. After the exclusion of 9 patients, the study cohort was 223. AMR was lower in the CADe-first group compared to the HDWL-first group: 20.12% (34/169) versus 31.25% (45/144) (P=0.0247), (OR) 1.8048 (95% CI 1.0780-3.0217). SSL miss rate was lower in the CADe-first group, 7.14% (1/14) versus the HDWL-first group, 42.11% (8/19) (P=0.0482). First-pass APC was higher in the CADe-first group (1.19 (SD 2.03) versus 0.90 (SD 1.55), (P=0.0323)). First-pass ADR was 50.44% in the CADe-first group and 43.64 % in the HDWL-first group (P=0.3091). CONCLUSION In this U.S. multicenter tandem colonoscopy RCT, we demonstrate a decrease in adenoma miss rate and SSL miss rate and an increase in first-pass APC with the use of a CADe-system when compared to HDWL colonoscopy.