Creation of a Video/Image Library of Annotated Full-length Endoscopy Procedures for Developing AI-powered Endoscopy Quality Reporting and Educational Tools

What this Trial Is About

Researchers are creating a video and image library of complete or partial endoscopy procedures for patients with rectal cancer or inflammatory bowel disease. The goal is to develop and validate new artificial intelligence (AI) tools that can automatically detect and report endoscopy quality and provide educational resources to improve endoscopy skills. This observational study aims to use a diverse video dataset to build robust AI models that offer real-time quality feedback during endoscopy. The study involves recording upper and lower gastrointestinal endoscopy procedures using high-definition equipment and advanced imaging techniques. Board-certified gastroenterologists or supervised trainees perform these endoscopies, which are video recorded with precise timestamps and stored securely. The procedures include diagnostic and therapeutic steps such as polyp detection, optical diagnosis, and polyp removal. All biopsy and polyp specimens are sent for pathology assessment. Participants will have their endoscopy procedures documented from enrollment through 14 days after the procedure. Research staff will capture relevant annotations during the procedures. The main measurements include documented data from esophagogastroduodenoscopies and colonoscopies. This study is observational, so participants will continue with usual care, and data will be collected to support AI development and educational tool creation.

Video/Image Library of Endoscopy Procedures for the Development of AI-empowered Endoscopy Quality Reporting and Educational Modules