Enhancing radiologist's detection: an imaging-based grading system for differentiating Crohn's disease from...

Area of Science:

• Gastroenterology
• Radiology
• Medical Imaging

Background:

• Delayed diagnosis of inflammatory bowel disease (IBD) is a significant clinical challenge.
• Current imaging modalities lack effective methods to differentiate between Crohn's Disease (CD) and Ulcerative Colitis (UC).

Purpose of the Study:

• To develop and validate an imaging report and data system for IBD (IBD-RADS).
• To establish reliable imaging indicators for distinguishing CD from UC patients.

Main Methods:

• A multicenter retrospective study analyzed CT imaging signs in IBD patients.
• Visceral fat metrics, including visceral to subcutaneous fat ratio (VSR), were extracted.
• Binary logistic regression and ROC analysis identified key indicators for the IBD-RADS system.

Main Results:

• Asymmetric enhancement (AE), perianal fistula, and VSR were independent predictors of CD.
• The IBD-RADS system, incorporating AE and VSR (visceral fat predominance > 0.97), demonstrated high diagnostic efficiency (AUC: 0.929).
• IBD-RADS accurately classified CD patients, achieving 96.5% in the derivation cohort and 98.0% in the validation cohort.

Conclusions:

• The developed IBD-RADS system aids radiologists in differentiating CD and UC in suspected IBD cases.
• IBD-RADS offers improved diagnostic accuracy for distinguishing between these two major forms of IBD.