OMOM SmartScan provides noninferior assessment of Lewis score and CECDAI in Crohn's disease vs full-video review.

BACKGROUND: Small bowel capsule endoscopy (SBCE) is central to Crohn's Disease (CD) care but limited by lengthy review and reader variability. We tested whether an artificial-intelligence (AI)-triaged, human-reviewed workflow - SmartScan, SS - can serve as a first-line alternative to full-video (FV) review, reducing reading time while maintaining safety and score fidelity. METHODS: Single-center, randomized, assessor-blinded, dual-reader, 2 × 2 crossover non-inferiority reading study of OMOM® HD SBCE videos in adults with established CD. Each study underwent paired FV and SS reads. The primary endpoint was non-inferiority of the agreement on Lewis Score (LS) activity category plus stricture status. Secondary endpoints included LS category agreement, equivalence of continuous LS/CECDAI, within-tolerance agreement, stricture sensitivity, performance with short transit and low Brotz quality, and reading-time superiority. RESULTS: We analyzed 153 paired studies. SS and FV agreed on both LS category and stricture in 96.1 % (147/153), surpassing the non-inferiority (NI) margin (p = 0.005). LS category exact match was 96.7 %. Continuous scores were equivalent, and stricture sensitivity 100 % (11/11). Within-tolerance agreement was 91.5 % (140/153) but failed to meet the prespecified NI criteria. Results were similar with short transit and low Brotz quality. Mean reading time fell from 41.5 to 12.8 min (p < 0.001). CONCLUSIONS: SS achieved NI for the primary composite endpoint and preserved equivalence of continuous LS/CECDAI, while markedly reduced reading time. These data support an AI-triaged, human-reviewed workflow with selective escalation to FV when uncertainty arises.