The Library
Advancing diffusion tensor image analysis and Interpretation for clinical application
Tools
Rajan, Surya (2021) Advancing diffusion tensor image analysis and Interpretation for clinical application. PhD thesis, University of Warwick.
|
PDF
WRAP_Theses_Rajan_2021.pdf - Submitted Version - Requires a PDF viewer. Download (28Mb) | Preview |
Official URL: http://webcat.warwick.ac.uk/record=b3684494
Abstract
The white matter of the brain is increasingly understood to play a crucial role in neurodegenerative diseases including mild cognitive impairment (MCI) and Alzheimer’s disease (AD). Diffusion tensor imaging (DTI) has been developed to image the white matter in vivo, and several parameters such as fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD) and axial diffusivity (AxD) may be computed from the acquired signal to infer the integrity of the tissue microstructure. The studies carried out in this thesis aim to improve DTI analysis for clinical applications and investigate the association of DTI signal with the underlying white matter physiology. Imaging data and neurophysiological assessments used in this thesis were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI).
A qualitative comparison between three different segmentation strategies of the corpus callosum emphasised the need for a consistent analysis protocol across DTI studies. It was demonstrated that the median of pixel values in a manually delineated ROI may provide more accurate measurements compared to the more widely used atlas-based ROI. A study of corpus callosum sub-regions using this segmentation strategy revealed statistically significant (p<0.05) alterations in DTI parameters in specific regions projecting to motor-related areas of the brain in MCI and AD, compared with healthy ageing. This involvement of the corpus callosum was supported by neurophysiological assessments of subjects that showed increased motor deficits in MCI patients such as tremors and gait imbalance.
Further evidence to support these results was obtained from tractographybased analysis of the corpus callosum of these subjects where similar alterations were found in DTI parameters in motor-related regions (p < 0.003). Results obtained from NODDI-DTI (an adaptation of Neurite Orientation and Dispersion Density Imaging (NODDI) for clinically acquired images) analysis of the healthy ADNI cohort indicated a decrease in neurite density in the corpus callosum with ageing that correlated with changes in FA. Adaptation of advanced methods to clinically acquired images was thus demonstrated to provide more specific information about the white matter changes, extending what was achieved in the conventional DTI analysis.
In the final section of the thesis, numerical models of healthy white matter, acute and chronic demyelination, and neuroinflammation were simulated to investigate the associations of FA, MD, RD, and AxD with underlying physiological mechanisms. The results indicated that acute demyelination generated a larger decrease in FA and larger increases in MD, RD, and AxD compared to other disease models - the pattern that is most seen in clinical studies. The study also investigated the dependence of these metrics on the transverse relaxation time (T2) of the white matter and its compartments. The results suggested that separating out the effects of relaxation and diffusion on the acquired signal provided a more accurate estimation of DTI metrics, achievable for the white matter as a whole at typical clinical scan settings, and for its compartments at more advanced scan settings.
Data collected in this thesis using DTI, tractography, and NODDI-DTI suggested that a reduction in fibre packing density was a major factor contributing to the decrease in FA and increases in MD, RD, and AxD, as widely reported in ageing, MCI, and AD. These results were demonstrated in the corpus callosum, but are likely to hold in other white matter tracts as well. The results from white matter modelling supported these findings, where acute demyelination (modelled as a decrease in fibre packing density) was found to cause the largest alterations in FA, MD, RD, and AxD, compared to other disease cases modelled.
In summary, the collected work in this thesis presents an analysis framework enabling application of research developments in clinical DTI, supported by improved specificity gained from modelling.
Item Type: | Thesis (PhD) | ||||
---|---|---|---|---|---|
Subjects: | R Medicine > R Medicine (General) R Medicine > RC Internal medicine |
||||
Library of Congress Subject Headings (LCSH): | Diffusion tensor imaging, Diffusion tensor imaging -- Diagnostic use, White matter, Image analysis, Alzheimer's disease -- Diagnosis, Imaging systems in medicine | ||||
Official Date: | February 2021 | ||||
Dates: |
|
||||
Institution: | University of Warwick | ||||
Theses Department: | School of Engineering | ||||
Thesis Type: | PhD | ||||
Publication Status: | Unpublished | ||||
Supervisor(s)/Advisor: | Collingwood, Joanna Frances | ||||
Sponsors: | University of Warwick. School of Engineering | ||||
Format of File: | |||||
Extent: | xxiv, 252 leaves : illustrations | ||||
Language: | eng |
Request changes or add full text files to a record
Repository staff actions (login required)
View Item |
Downloads
Downloads per month over past year