Tract Thickness Measurements Using Diffusion MRI

Abstract

Alzheimer disease (AD) is the most common cause of cognitive decline, resulting in progressive declines in memory, language, and visuospatial abilities. The quantitative characterisation of atrophy can provide useful biomarkers for assessing the evolution of such neurological conditions. While current literature has mainly focused on measuring the atrophy of whole brain, global gray and/or white matter (WM), specific lobes or gray matter structures (e.g. the hippocampus), it is likely that atrophy caused by such conditions also affects WM tracts via degenerative processes. If specific tract systems are more prone to atrophy than others, then tractography-guided atrophy measurements may be more sensitive than less targeted methods.With this aim in mind, this thesis presents a set of novel methods that form a framework for quantifying the width of WM tracts which may be used to differentiate between normal and patient groups or in the assessment of changes in the rate of disease progression due to pharmaceutical intervention. This framework overcomes numerous shortcomings of previously described methods in the literature. The findings of this thesis demonstrate that our framework is capable of identifying the effect of progression of AD on tractography derived pathways. Additionally, the results of the studies conducted as part of this thesis indicate that our method is sensitive to both localised cross-sectional and longitudinal changes and variations in tract width, and increases in the rates of atrophy, which can be caused by neurodegenerative processes, and therefore, can have future use as a potential biomarker in clinical trials, assessing the performance of therapeutic intervention on the rate of atrophy. The novel methodological contributions of this work include: 1) an automated method for the identification of the optimum study specific probabilistic tractography threshold; 2) an unbiased mesh subdivision technique; 3) an implementation of an algorithm for the curve-skeletonisation of WM tracts; 4) a geodesic-based sampling strategy for the quantification of tract half-width; 5) a method for establishing spatial correspondences between WM tracts, termed automated correspondence establishment (ACE); and 6) a method for performing automated statistical tests amongst corresponding WM tracts, termed tract specific statistical template (TEST).

Date of Award	5 Apr 2011
Original language	English
Awarding Institution	The University of Manchester