Background and study aims Polypectomy-related costs could potentially be reduced through optical diagnosis strategies, such as 'diagnose-and-leave' and 'resect-and-discard.' Artificial intelligence, using computer-aided diagnosis (CAD), may provide a reproducible optical diagnosis of colorectal lesions. This study aimed to assess the performance of the CAD-EYE® system in the real-time characterization of colonic polyps. Methods We conducted a cross-sectional, multicenter study evaluating the CAD-EYE® system in patients undergoing screening colonoscopies at five French centers. CAD-EYE® predictions and assessments by endoscopists (hyperplastic vs. neoplastic) were compared to histopathology results. The primary outcome was the sensitivity of CAD-EYE® for predicting neoplastic polyps, compared to the predefined threshold of 85%. The secondary outcomes were the specificity, positive predictive value (PPV), negative predictive value (NPV), endoscopists’ performance, and polyp detection rates. Results Of 398 polyps analyzed, 343 were included in the primary analysis. CAD-EYE® characterization was feasible in 96% of cases. The sensitivity was 0.80 (95% confidence interval, 0.74–0.85), which failed to achieve the predefined threshold of 85% (p = 0.064). The specificity, NPV, and PPV were 0.79, 0.64 and 0.90, respectively. Performance was higher for diminutive rectosigmoid polyps (DRSPs). Endoscopists showed higher sensitivity than CAD-EYE® (0.90 vs. 0.80, p = 0.001). CAD-EYE®-assisted colonoscopies detected more polyps per procedure (3.3 vs. 2.3, p < 0.001) than endoscopists alone. Conclusion The performance of CAD-EYE® was insufficient for the characterization of neoplastic colonic polyps. CAD-EYE® performed better for DRSPs. AI appears to be beneficial for polyp detection.
Maigné et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: