Cure8 research brief
An AI-derived score that combines inflammation severity and extent may better predict relapse risk in ulcerative colitis than severity alone. That could help identify patients in endoscopic remission who still face higher relapse risk and might need closer follow-up.
Gastroenterologists, IBD clinicians, researchers working on AI or biomarkers in IBD, and patients with ulcerative colitis interested in relapse risk prediction and endoscopy-based monitoring.
The research team trained QUAD on >84,000 colonoscopy images from nearly 1,000 patients and applied it to still images and full-length videos. QUAD scores range 0–12 (sum of quadrant scores); patients with QUAD ≥4 had a higher relapse rate than those with QUAD <4.
Automated video analysis suggested that measuring distal inflammatory burden (distal 10% segment) produced stronger relapse prediction than whole-colon or still-image assessments.
How people might use this: This work suggests AI tools that measure both how bad inflammation is and how much of the colon is affected could add prognostic information beyond conventional severity-only indices.
In the future, such tools might help clinicians identify patients in endoscopic remission who remain at higher risk of relapse and who could benefit from closer monitoring. Study scope and limits: This brief is grounded in the article abstract and reported prospective cohort data (24-month follow-up).
The abstract reports predictive performance metrics (AUCs) but does not provide full implementation details, external validation across broader clinical settings, or direct clinical outcome changes from using QUAD.
This summary is based on the article abstract (prospective cohort, 24-month follow-up). Results are predictive-model findings; the abstract does not replace peer-reviewed guidelines or establish that using QUAD in routine care improves outcomes. External validation and implementation studies would be needed.
Review the original publication for the complete reporting, methods, and context.
Funding disclosed by the source: JSPS KAKENHI - JP23K09537; the Takeda Science Foundation, and The Japanese Foundation for Research and Promotion of Endoscopy
This Cure8 brief is based on source text from the linked article. Cure8 is informational only and is not a substitute for professional medical advice, diagnosis, or treatment.
Automated, consistent scoring of the UCEIS could reduce subjectivity and variability in endoscopic assessment of ulcerative colitis, which matters for clinical trials, central reading, and potentially treatment decisions...
Measuring both how severe inflammation is and how much of the colon is affected could give a fuller picture of disease burden than current focal endoscopic scores. For patients with UC, that may affect risk assessment...
A more detailed, reproducible measure of endoscopic inflammation could better reflect total disease burden and reduce variability from human readers. That may matter for monitoring disease activity, judging treatment...
