Local Assessment and Small Bowel Crohn’s Disease Severity Scoring using AI

We present a machine learning and computer vision approach for a localized, automated, and standardized scoring of Crohn’s disease (CD) severity in the small bowel, overcoming the current limitations of manual measurements CT enterography (CTE) imaging and qualitative assessments, while also considering the complex anatomy and distribution of the disease. Two radiologists introduced a severity score and evaluated disease severity at 7.5 mm intervals along the curved planar reconstruction of the distal and terminal ileum using 236 CTE scans. A hybrid model, combining deep-learning, 3-D CNN, and Random Forest model, was developed to classify disease severity at each mini-segment. Precision, sensitivity, weighted Cohen’s score, and accuracy were evaluated on a 20% hold-out test set. The hybrid model achieved precision and sensitivity ranging from 42.4% to 84.1% for various severity categories (normal, mild, moderate, and severe) on the test set. The model’s Cohen’s score (κ = 0.83) and accuracy (70.7%) were comparable to the inter-observer agreement between experienced radiologists (κ = 0.87, accuracy = 76.3%). The model accurately predicted disease length, correlated with radiologist-reported disease length (r = 0.83), and accurately identified the portion of total ileum containing moderate-to-severe disease with an accuracy of 91.51%. The proposed automated hybrid model offers a standardized, reproducible, and quantitative local assessment of small bowel CD severity and demonstrates its value in CD severity assessment. •Crohn’s disease (CD) severity scores were assessed by two radiologists at 7.5 mm intervals (mini-segments) along the curved planar reconstruction (CPR) of the ileum, originating from the ileocecal valve, using 236 CTE unique scans of patients diagnosed with CD.•An AI model, combining clinically inspired and non-traditional features through a designed 3-D CNN and Random Forest model, achieved desirable disease severity classification at each mini-segment. It demonstrated an accuracy of 91.51% for moderate-to-severe disease.•The model’s performance, as measured by weighted Cohen’s score (κ = 0.83, accuracy 70.7%), was comparable to inter-observer agreement between experienced radiologists (κ = 0.87, accuracy = 76.3%) on the test set.