Multi-tissue spherical deconvolution of tensor-valued diffusion MRI

Neuroimage. 2021 Dec 15;245:118717. doi: 10.1016/j.neuroimage.2021.118717. Epub 2021 Nov 11.

Abstract

Multi-tissue constrained spherical deconvolution (MT-CSD) leverages the characteristic b-value dependency of each tissue type to estimate both the apparent tissue densities and the white matter fiber orientation distribution function from diffusion MRI data. In this work, we generalize MT-CSD to tensor-valued diffusion encoding with arbitrary b-tensor shapes. This enables the use of data encoded with mixed b-tensors, rather than being limited to the subset of linear (conventional) b-tensors. Using the complete set of data, including all b-tensor shapes, provides a categorical improvement in the estimation of apparent tissue densities, fiber ODF, and resulting tractography. Furthermore, we demonstrate that including multiple b-tensor shapes in the analysis provides improved contrast between tissue types, in particular between gray matter and white matter. We also show that our approach provides high-quality apparent tissue density maps and high-quality fiber tracking from data, even with sparse sampling across b-tensors that yield whole-brain coverage at 2 mm isotropic resolution in approximately 5:15 min.

Keywords: B-tensors; Magnetic resonance imaging; Multi-tissue constrained spherical deconvolution; Multidimensional diffusion encoding; Tensor-valued diffusion encoding; Tractography.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Brain Mapping
  • Diffusion Tensor Imaging / methods*
  • Healthy Volunteers
  • Humans
  • Image Processing, Computer-Assisted
  • White Matter / diagnostic imaging*