Use this URL to cite or link to this record in EThOS:
Title: An insight into the brain of patients with type-2 diabetes mellitus and impaired glucose tolerance using multi-modal magnetic resonance image processing
Author: Dounavi, Maria Eleni
ISNI:       0000 0004 7970 2290
Awarding Body: University of Sheffield
Current Institution: University of Sheffield
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
The purpose of this thesis was to investigate brain anatomy and physiology of subjects with impaired glucose tolerance (IGT - 12 subjects), type-2 diabetes (T2DM - 17 subjects) and normoglycemia (16 subjects) using multi-modal magnetic resonance imaging (MRI) at 3T. Perfusion imaging using quantitative STAR labeling of arterial regions (QUASAR) arterial spin labeling (ASL) was the core dataset. Optimization of the post-processing methodology for this sequence was performed and the outcome was used for hemodynamic analysis of the cohort. Typical perfusion-related parameters, along with novel hemodynamic features were quantified. High-resolution structural, angiographic and carotid flow scans were also acquired and processed. Functional acquisitions were repeated following a vasodilating stimulus. Differences between the groups were examined using statistical analysis and a machine-learning framework. Hemodynamic parameters differing between the groups emerged from both baseline and post-stimulus scans for T2DM and mainly from the post-stimulus scan for IGT. It was demonstrated that quantification of not-typically determined hemodynamic features could lead to optimal group-separation. Such features captured the pattern of delayed delivery of the blood to the arterial and tissue compartments of the hyperglycemic groups. Alterations in gray and white matter, cerebral vasculature and carotid blood flow were detected for the T2DM group. The IGT cohort was structurally similar to the healthy cohort but demonstrated functional similarities to T2DM. When combining all extracted MRI metrics, features driving optimal separation between different glycemic conditions emerged mainly from the QUASAR scan. The only highly discriminant non-QUASAR feature, when comparing T2DM to healthy subjects, emerged from the cerebral angiogram. In this thesis, it was demonstrated that MRI-derived features could lead to potentially optimal differentiation between normoglycemia and hyperglycemia. More importantly, it was shown that an impaired cerebral hemodynamic pattern exists in both IGT and T2DM and that the IGT group exhibits functional alterations similar to the T2DM group.
Supervisor: Wilkinson, Iain D. ; Martin, Christopher ; Armitage, Paul Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available