Diffusion Weighted Imaging (DWI) MRI detects prostate cancers but is very sensitive to motion artifacts. There has been little quantitative evaluation of variability to guide clinical use of DWI. We found very high variability between individual acquisitions used for averaging at high b-values (% ranges of 74.08% - 115.56% in cancers and 53.53% - 159.91% in normal prostate tissue), primarily due to motion during diffusion-sensitizing gradients. High signals in cancer voxels appear in some acquisitions but not others. Therefore, standard averaging can obscure cancers. We propose alternative methods for combining information from individual images at each b-value to maximize diagnostic accuracy.
This abstract and the presentation materials are available to members only; a login is required.