Interpretable ML model combines AI DBT features and blood immune markers to classify breast masses

Logistic regression model integrating AI-structured DBT features and blood inflammatory biomarkers classified breast masses with AUROC 0.884 (95% CI 0.807-0.950), sensitivity 0.806, specificity 0.878 in testing set.

• SHAP analysis identified suspicious calcifications, irregular mass shape, advanced age, elevated neutrophil-to-lymphocyte ratio, and larger lesion size as top malignancy predictors.
• Malignancy rate ranged from 2.5% in the low-risk group to 67.6% in the high-risk group.
• Limitation: single-center, retrospective study; prospective multicenter external validation is needed.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.