Using Artificial Intelligence to Improve Safety in Complex Endoscopic Procedures Detecting Blood Vessels During Treatment of Achalasia Cardia and Tumors

What this Trial Is About

Researchers are evaluating a new artificial intelligence AI tool designed to assist during third space endoscopic procedures, such as Endoscopic Submucosal Dissection ESD and Peroral Endoscopic Myotomy POEM. These procedures treat gastrointestinal tumors and motility disorders but carry risks of bleeding and perforation if blood vessels under the mucosa are not properly identified and managed. This study aims to see if the AI can help endoscopists detect and outline blood vessels more accurately and quickly during live procedures, building on previous retrospective studies. The study uses an AI system integrated into the Olympus EVIS X1 endoscopy equipment to provide real-time visual overlays that highlight submucosal blood vessels. Participants will be randomly assigned to two groups one where the endoscopist sees the AI-generated vessel segmentation mask during the procedure, and another where the endoscopist performs the procedure without AI assistance. The performance of vessel detection and timing will be recorded and compared with expert reviews after the procedure. Participants diagnosed with Achalasia Cardia or tumors will take part in this prospective trial. Researchers will evaluate how well the AI improves the Vessel Detection Rate VDR and Vessel Detection Time VDT during the first three months. Safety monitoring will focus on procedural outcomes related to vessel identification. The total participation time and detailed follow-up procedures are aligned with the study design to ensure comprehensive evaluation of the AI tool during these complex endoscopic procedures.

Artificial Intelligence-Guided Detection of Blood Vessels to Enhance Safety in Third-Space Endoscopic Procedures