Single and Multi Diffusion-Tensor Based Kernels for Anisotropic Filtering of Brain DW-MR Images

Abstract

Diffusion Weighted Magnetic Resonance Imaging is widely used to study the structure of the fiber pathways of brain white matter. Though, the recovered axon orientations could be prone to error because of the low signal to noise ratio. Spatial regularization can improve the estimations but it must be done carefully such that real information is not removed and false orientations are not introduced. In this work we investigate the advantages to apply an anisotropic filtering based on single and multiple axon bundle orientation kernels. To this aim, we compute local diffusion kernels based on Diffusion Tensor and multi Diffusion Tensor models. We show the benefits of our approach on three different types of DW-MRI Data: synthetic, in vivo human data, and acquired from a diffusion phantom.