Modelling white matter with spherical deconvolution: How and why?
NMR in Biomedicine
Since the realization that diffusion MRI can probe the microstructural organization and orientation of biological tissue in vivo and non-invasively, a multitude of diffusion imaging methods have been developed and applied to study the living human brain. Diffusion tensor imaging was the first model to be widely adopted in clinical and neuroscience research, but it was also clear from the beginning that it suffered from limitations when mapping complex configurations, such as crossing fibres. In
... this review, we highlight the main steps that have led the field of diffusion imaging to move from the tensor model to the adoption of diffusion and fibre orientation density functions as a more effective way to describe the complexity of white matter organization within each brain voxel. Among several techniques, spherical deconvolution has emerged today as one of the main approaches to model multiple fibre orientations and for tractography applications. Here we illustrate the main concepts and the reasoning behind this technique, as well as the latest developments in the field. The final part of this review provides practical guidelines and recommendations on how to set up processing and acquisition protocols suitable for spherical deconvolution. KEYWORDS diffusion imaging, diffusion tensor imaging, fiber orientation density function, fiber response, ODF, MRI, spherical deconvolution, tractography | INTRODUCTION Even before diffusion MRI was proposed as a method for tractography, 1-3 it was recognized that the diffusion tensor model was strongly affected by the presence of crossing fibres. In their influential 1996 publication, Basser and Pierpaoli noted that their proposed measures of anisotropy were highly dependent on the degree of coherence of fibre tract directions. 4 Nevertheless, the unique ability to extract orientational information non-invasively from living biological tissues made the diffusion tensor one of the main tool of modern neuroimaging. 5 Many tensor-based tractography methods have since been proposed and successfully used in clinical and neuroscience research applications. 6-9 -This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.