A Novel Simplified Endoscopic Score Using TXI, RDI, and NBI to Evaluate Ulcerative Colitis Activity and Predict Clinical Outcomes with Artificial Intelligence: the MONET Study

What this Trial Is About

Researchers are conducting an international multicenter study to develop a new simplified endoscopic score focused on vascular features of the colon to distinguish between inactive and mild inflammation in ulcerative colitis (UC). This study aims to compare the new score's ability to define disease activity and remission against existing endoscopic and histological scores, as well as to predict long-term clinical outcomes. Additionally, the study seeks to adapt artificial intelligence (AI) algorithms to enhance disease assessment and outcome prediction using various enhanced endoscopic techniques. The study will use high-definition white light endoscopy (WLE-HD) along with texture and colour enhancement imaging (TXI), red dichromatic imaging (RDI), and narrow-band imaging (NBI) modes. It includes several phases: developing the score using video analysis and expert consensus, validating it in a large group of UC patients, assessing its reproducibility among gastroenterologists, and creating AI algorithms to standardize grading and prediction. Patients undergoing colonoscopy for disease assessment or surveillance will have colonoscopy with biopsies, blood and stool samples taken to monitor inflammation. Participants will be followed up at 6 and 12 months with clinic or telephone assessments to evaluate disease activity using the Partial Mayo Score and clinical outcomes. Researchers will analyze the diagnostic performance of the new score within six months and correlate it with existing scores and AI developments over two years. The study involves collecting clinical data, imaging, biopsies, blood, and stool samples to thoroughly assess and predict UC disease activity and outcomes.

AI-driven Narrow-band Imaging Score for Disease Assessment and Outcome Prediction in Ulcerative Colitis