Multi-parametric diffusion-weighted imaging (DWI) has shown potential in characterizing breast cancer. In this study, we for the first time evaluated the four common diffusion models in various breast lesions. Results showed that (1) optimal DWI signal attenuation behaviors differ between benign and malignant breast lesions (kurtosis model suited for benign and stretched-exponential model suited for malignant); (2) for predicting the malignancy of breast lesions, or the invasive nature of breast cancer, mono-exponential ADC is still the most preferred parameter; and (3) parameters based on non-mono-exponential DWI models may be more related with prognostic factors in invasive breast cancer.
This abstract and the presentation materials are available to members only; a login is required.