Multi-b-valued diffusion-weighted imaging (DWI) of the breast is highly susceptible to image and fitting noise. A multi-compartment approach was developed to denoise multi-b-value breast DWI without spatial smoothing. In human subject exams (N=12), the denoising approach resulted in a significant reduction in variability of all perfusion and diffusion maps in breast tumor and normal fibroglandular tissue with minimal bias to the mean values, and increased statistical separation of diffusivity metrics between tumor and normal tissue. The denoising algorithm provides compartment fractions for tumor, tissue, and vascularity, which may improve visualization of tissue compartments in DWI.
This abstract and the presentation materials are available to members only; a login is required.