Colorectal cancer (CRC) remains a leading cause of cancer-related morbidity and mortality worldwide, with incidence steadily increasing across the Middle East. Quality colonoscopy metrics, including adenoma detection rate (ADR), are key determinants of effective CRC prevention, yet regional data remain limited. This multicenter randomized controlled trial aimed to assess whether computer-aided detection (CADe) technology enhances colonoscopy quality indicators—particularly ADR—compared with conventional white-light colonoscopy in Middle Eastern populations.

This prospective multicenter RCT was conducted between January 2022 and December 2024 across seven tertiary centers in Egypt, Bahrain, Iran, Jordan, Saudi Arabia, and Kuwait. Expert endoscopists performed screening, surveillance, or diagnostic colonoscopies in adults aged 40–80 years using either conventional white-light colonoscopy or AI-assisted colonoscopy equipped with CADe (CAD EYE, Fujifilm). Patients were randomized 1:1 to each arm. The primary outcome was ADR. Secondary outcomes included adenomas per colonoscopy (APC), polyp detection rate (PDR), and polyps per colonoscopy (PPC). Statistical significance was set at p < 0.05.

A total of 441 patients (mean ± SD age 55 ± 10 years; 46% female) were included. The CADe-assisted group achieved a similar ADR to the conventional group (34.3% vs 30.3%; p = 0.394). However, CADe significantly improved overall PDR (60.4% vs 48.2%; p = 0.010) and detected a greater proportion of diminutive polyps (≤ 5 mm) compared with white-light colonoscopy (51.6% vs 40.6%; p = 0.013). The mean APC and PPC were also higher in the CADe group. 

CADe-assisted colonoscopy significantly improved overall polyp detection, particularly for diminutive lesions, demonstrating its value as both a clinical aid and a training tool to enhance endoscopist performance. Although the ADR increase was modest, it aligns with several real-world and nonrandomized studies, reinforcing the consistency of CADe’s benefit across diverse settings. As the first multicenter RCT in the Middle East, this study underscores the potential of AI-assisted systems to standardize detection quality and improve diagnostic accuracy in regional populations. Future large-scale studies should focus on optimizing algorithm performance and evaluating its role in endoscopic education and practice integration.