Randomized Trial Comparing AI Systems and Standard Colonoscopy for Better Detection and Diagnosis of Colon Polyps in Adults Aged 50 to 90

What this Trial Is About

This research aims to evaluate whether artificial intelligence AI can improve the detection and classification of polyps during colonoscopy. The study compares different AI systems to see if they differ in their ability to find polyps and whether missed polyps are benign or malignant. This randomized controlled clinical trial is conducted at several hospital sites with patients undergoing elective colonoscopy. Participants are randomly assigned to receive either a colonoscopy using one of three AI systemsEndoAid, CADEYE, or GI-Geniusor a conventional colonoscopy without AI assistance. During AI-assisted colonoscopy, the AI system is activated during instrument withdrawal to identify and alert the endoscopist about polyps. Two AI systems also provide interpretations regarding the benign or malignant nature of the polyps. After the procedure, questionnaires are completed documenting the findings. During the study, data collected includes patient demographics, medical history, examination details such as polyp number and classification, bowel preparation quality, procedure times, and any complications. Researchers will compare adenoma detection rates and the accuracy of AI polyp characterization against conventional histology. Costs associated with AI-assisted versus conventional colonoscopy will also be assessed. Participation involves one colonoscopy procedure with data recorded during and after the examination.

Comparative Analysis of AI Software for Enhanced Polyp Detection and Diagnosis