Researchers have developed and evaluated a machine learning approach of using patient core needle biopsy data to identify the risk that atypical ductal hyperplasia (ADH) breast lesions may upgrade to cancer. This knowledge can potentially help clinicians and low-risk patients decide whether active surveillance and hormonal therapy is a reasonable management approach. Using the method could spare patients with benign lesions from invasive surgeries while maintaining high sensitivity for predicting malignant lesions.