The predictive value of preoperative Multiparametric MRI Radiomics Model for axillary lymph node metastasis in breast cancer

Motivation: Axillary lymph node (ALN) involvement plays a pivotal role in the prognosis of breast cancer（BRCA）. Goal(s): This study aims to develop a radiomics model that integrates features from multiparametric MRI to preoperatively diagnose ALN metastasis in BRCA . Approach: Radiomics features were extracted from the multiparametric MRI of 227 patients. Patients were classified into three groups based on the metastatic ALN load .Models were developed using machine learning, with performance evaluated using the area under the curve (AUC). Results: The models achieved AUCs of 0.854, 0.687, 0.825 in the training cohort, and 0.702, 0.522, 0.753 in the validation cohort for three groups, respectively. Impact: These radiomics models serve as a noninvasive tool to predict ALN metastasis in BRCA preoperatively.