Logistic regression aids BI-RADS 4–5 microcalcification risk using mammography and DCE-MRI

A logistic regression model combining mammographic descriptors, age, and DCE-MRI enhancement classified BI-RADS 4/5 microcalcifications with 90.9% accuracy (F1 0.889) in preliminary internal testing of 53 biopsy-proven cases. Findings require multicenter validation before practi…

• Single-center retrospective study of 53 biopsy-confirmed women with BI-RADS 4/5 microcalcifications; no external validation cohort was used.
• AdaBoost achieved recall of 1.000, but logistic regression was the most balanced classifier overall (AUC not explicitly stated in source).
• Key limitation: very small sample size (n=53) and no independent external or prospective validation, raising high risk of overfitting.

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