Development of a Novel Real Time Computer Assisted Colonoscopy Diagnostic Tool for Colorectal Polyps: Lesion Diagnosis and Personalised Patient Management

What this Trial Is About

Researchers are evaluating a new artificial intelligence (AI) tool designed to better classify colorectal polyps found during colonoscopy. Accurate identification of polyp types is important to assess colorectal cancer risk and guide treatment decisions. Current methods require specialist skills and have limitations, so this international study aims to develop a five-class AI algorithm to provide reliable, personalized cancer risk assessments independent of the endoscopist's expertise. The study collects data from patients undergoing standard colonoscopy procedures for symptoms or screening. No changes or extra procedures are required; however, video recordings of the colonoscopy will be collected with patient consent. This observational study uses images and data entered securely to help train and validate the AI tool. The goal is to improve polyp diagnosis accuracy without increasing procedure time or complexity. Participants will undergo their usual colonoscopy care without additional visits. Researchers will record data such as polyp characteristics and endoscopist assessments. The main outcomes measured include the AI tool's accuracy in predicting polyp types and agreement between endoscopists. The study duration for outcome assessment is up to 36 months, with no extra interventions or follow-up visits beyond standard care.

Artificial Intelligence Development for Colorectal Polyp Diagnosis