Radiomics analysis of contrast-free synthetic MRI and ADC for early predicting neoadjuvant chemotherapy response in breast cancer

Motivation: Contrast-free sequences are receiving increasing attention. As a novel technology, radiomics analysis of synthetic MRI (SyMRI) in breast treatment has not been widely explored. Goal(s): To investigate the radiomics features extracted from the contrast-free SyMRI and apparent diffusion coefficient (ADC) in predicting pathological complete response in breast cancer. Approach: The support vector machine-based recursive feature elimination (SVM-RFE) was used to build models. The model performance was assessed by receiver operating characteristic curve and compared by Delong test. Results: Radiomics model based on delta-SyT2 and 1st-ADC achieved the optimal performance after one cycle treatment. After adding receptor status, the clinical-radiomics achieved the highest performance. Impact: Multiparametric SyMRI and ADC, as noncontrast sequences, have the potential to early predict NAC response in radiomics analysis, thus reducing the need for repeated contrast agent injections during treatment.