AI-Assisted Colonoscopy to Predict Polyp Histology and Reduce Costs in Colorectal Cancer Screening

What this Trial Is About

Researchers are evaluating the use of artificial intelligence AI to improve colonoscopy screening for colorectal polyps, aiming to distinguish between cancerous neoplastic and non-cancerous non-neoplastic polyps. This trial, conducted at multiple centers with three parallel groups, will also assess cost-saving strategies by using AI-assisted optical biopsy and develop new AI tools to better identify cancer risk regardless of the physicians skill. The study includes a cost-effectiveness analysis using real-world data and advanced trial emulation methods. Participants will undergo high-definition colonoscopy with one of three approaches standard colonoscopy assisted by AI detection CADe, colonoscopy assisted by both AI detection and diagnosis CADeCADx with all polyps removed, or colonoscopy assisted by AI detection and diagnosis where certain small polyps in specific locations may be left in place if judged to be harmless. All removed polyps will be examined by pathology. The trial compares these approaches to find the best balance between accurate diagnosis and cost savings. During the study, participants will be monitored over four years to measure outcomes such as the adenoma detection rate, the accuracy of ruling out colorectal neoplasia, the agreement between surveillance plans based on polyp removal versus leaving some polyps in place, and the impact on costs related to polyp removal and pathology. This long-term follow-up includes assessments at regular intervals to evaluate effectiveness and safety, contributing to improved personalized colorectal cancer screening.

Artificial Intelligence to Implement Cost-saving Strategies for Colonoscopy Screening Based on in Vivo Prediction of Polyp Histology