Motivation: Accurately differentiating BI-RADS 4 breast lesions is critical in breast cancer diagnostics, yet conventional imaging methods often result in unclear distinctions, leading to unnecessary biopsies. Goal(s): This study aimed to develop a multi-parameter MRI radiomics model to improve diagnostic accuracy in distinguishing between benign and malignant BI-RADS 4 lesions. Approach: The model incorporated DWI, ADC, and DCE-MRI sequences, with radiomic features extracted and analyzed using machine learning classifiers, including RF, SVM, and LR. Results: The model, particularly with the LR classifier, showed high diagnostic accuracy and sensitivity, effectively distinguishing benign from malignant lesions and offering clinical support to reduce unnecessary biopsies. Impact: This study's multi-parameter MRI radiomics model enhances diagnostic accuracy for BI-RADS 4 breast lesions, offering radiologists a reliable tool for distinguishing benign from malignant cases, reducing unnecessary biopsies, and improving patient management in breast cancer diagnostics.
li et al. (Tue,) studied this question.