AI-Assisted Colonoscopy Linked to Higher Adenoma Detection and Lower Interval CRC

Artificial intelligence (AI)–assisted colonoscopy was associated with significantly improved adenoma detection and reduced interval colorectal cancer (CRC) incidence in a large real-world analysis, according to a retrospective cohort study using national electronic health record data presented at Digestive Disease Week 2026.

Investigators analyzed more than 3 million colonoscopy encounters from the TriNetX network, comparing outcomes between a pre-AI era (2015–2019) and an AI adoption era (2022–2025). After propensity score matching, more than 1.5 million patients were included in each cohort, with balanced demographics and clinical characteristics.

Within 30 days of colonoscopy, adenoma detection rates nearly doubled in the AI era compared with the pre-AI period (3.6% vs 1.8%; risk ratio [RR], 1.97; P<0.001). Advanced adenoma detection also improved, increasing from 0.13% to 0.19% (RR, 1.48; P<0.001).

Long-term outcomes showed a significant reduction in interval CRC. Between 6 and 36 months after colonoscopy, CRC incidence was 0.11% in the AI era compared with 0.21% in the pre-AI cohort, representing a 47% relative reduction (RR, 0.53; P<0.001).

The authors reported that “adenoma detection was significantly higher in the AI era,” and concluded that AI use was associated with “nearly 50% lower interval CRC.” These findings suggest that improved lesion detection translates into meaningful downstream clinical benefit.

Reference
Butt MA, Karna R, Almasri H, et al. Real-world impact of the AI colonoscopy era on adenoma detection and interval colorectal cancer: a national propensity-matched study. Presented at: Digestive Disease Week; May 2–5, 2026; Chicago, Illinois.