DBT radiomics nomogram accurately classifies BI-RADS 4 breast masses

A nomogram combining DBT-based intralesional and perilesional radiomics with patient age and BI-RADS 4 subclassification achieved an AUC of 0.942 in classifying benign vs malignant breast masses in the validation set, with SHAP analysis showing feature contributions.

• Retrospective study of 312 patients with BI-RADS 4 masses assessed by DBT; combined intralesional and perilesional radiomics was superior to single-region models.
• Validation AUC of final nomogram (radiomics score, age, subclassification) was 0.942 (training AUC 0.960).
• Key limitation is single-center retrospective design without external validation; may not generalize to other screening populations or DBT platforms.

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