The clinical validity of radiomics-based prediction of molecular subtypes in breast cancer from digital mammary tomosynthesis

1 Introduction Breast cancer (BC) is the most common cancer diagnosed in women, and the second leading cause of all cancer-related deaths (1). Early diagnosis of BC, as well as predicting prognosis and treatment response, is a primary focus of clinical research. Based on specific receptor expression levels, BC subtypes include luminal, human epidermal growth factor receptor 2 (HER2)-enriched, and triple-negative (TN) (2, 3).