Seunarine, KK; Alexander, DC; (2009) Multiple fibers: Beyond the diffusion tensor. In: UNSPECIFIED (55 - 72).
Full text not available from this repository.
This chapter describes the techniques for resolving multiple fiber populations in each voxel, as well as methods to exploit the information they recover. It begins by describing the limitations of DTI and the problems that, complex white-matter configurations such as crossing fiber-populations and bending fibers present. The technique has become popular because it provides two unique insights into tissue microstructure: it quantifies diffusion anisotropy, which is a useful index of white matter integrity, and provides an estimate of the principal direction of axon fibers, which enables tractography. Powerful though it is, DTI has several limitations. One key limitation is that it can only recover a single fiber orientation in each voxel and fails at fiber crossings. This limitation is a major obstacle for tractography and connectivity mapping. The multi-tensor model is a simple generalization of DTI, which replaces the Gaussian model for p with a mixture of n Gaussian densities. The model assumes the voxel contains n distinct groups or "populations" of fibers and that diffusing molecules stay within only one population. It describes multiple-tensor models, diffusion spectrum imaging, QBall, spherical deconvolution, and persistent angular structure (PAS) MRI, along with the pros and cons of each method. One class of algorithms, including QBall, DOT, and PASMRI, estimates features of the particle displacement density, p, that are spherical functions with peaks that provide fiber-orientation estimates. The aim is usually to recover the distribution of fiber orientations or fODF. However, the relationship between p and the fODF is complex and unclear. Spherical deconvolution methods estimate the fODF more directly, but rely on overly simple modeling assumptions. © 2009 Copyright © 2009 Elsevier Inc. All rights reserved.
|Title:||Multiple fibers: Beyond the diffusion tensor|
|UCL classification:||UCL > School of BEAMS > Faculty of Engineering Science > Computer Science|
Archive Staff Only: edit this record