Combination of time‑dependent diffusion MRI and Intravoxel Incoherent Motion for Predicting NPI and Molecular Subtypes of Breast Cancer

Motivation: Conventional diffusion-weighted MRI with apparent diffusion coefficient in clinical breast imaging may not accurately reflect the authentic diffusion. Goal(s): To quantitatively compare the diagnostic utility of the time-dependent diffusion MRI (td-dMRI) and intravoxel incoherent motion (IVIM) to predict Nottingham Prognostic Index and molecular subtypes of breast cancer. Approach: The Mann-Whitney U-test and ROC analysis were used to evaluate the diagnostic efficacy of each parameter and model in predicting NPI and molecular subtypes. Results: Quantitative parameters derived from td-dMRI have great promise as imaging indicators for the characterization of breast cancer. Additionally, the integration of td-dMRI and IVIM parameters achieves the superior diagnostic performance. Impact: The integration of td-dMRI and IVIM provides a new perspective for characterizing breast cancer by comprehensively taking the cellularity, vascularity, and microstructure of tissues into account, thus having potential clinical utility in the prognostic estimation and personalized treatment strategies.