Diffusion MRI has been used for tumor grading due to its sensitivity to alterations at the tissue microstructural level. Recognizing the limitations of analysis methods based on region-of-interest (ROI) in which the parameter values are averaged over the tumor ROIs, histogram-based approaches have been proposed for differentiating tumor grades. One challenge in this approach is to determine an optimal percentile over the ROI to be used in the analysis. In this study, we systematically and statistically determined an optimal percentile cut-off for calculating the mean parameters obtained from a non-Gaussian diffusion model based on continuous-time random-walk (CTRW) theory for differentiation among glioma grades.
This abstract and the presentation materials are available to members only; a login is required.