Nassar Arbid, Jannette; (2025) Optimising Multi-Echo Task-Based and Resting-State Functional Quantitative Susceptibility Mapping (fQSM). Doctoral thesis (Ph.D), UCL (University College London).

Text Nassar Arbid_10212325_thesis.pdf Access restricted to UCL open access staff until 1 September 2026. Download (19MB)

Abstract

Quantitative susceptibility mapping (QSM) is an MRI technique that reconstructs tissue magnetic susceptibility from MRI phase images, providing insights into tissue composition. Functional MRI (fMRI) is used to detect susceptibility-mediated changes in blood-oxygenation-level-dependent (BOLD) signal magnitude in response to stimulus-induced neural activity. In contrast, resting-state fMRI (rsfMRI) measures spontaneous brain activity in the absence of external stimulation and it has been shown that certain diseases (e.g. Alzheimer’s and Parkinson’s) affect resting-state brain networks such as the default mode network (DMN). Functional QSM (fQSM) is a more recent technique that directly detects blood susceptibility changes associated with neuronal activity, providing a complementary approach to functional brain mapping. Therefore, as changes in blood susceptibility underlie fMRI, I hypothesized that resting-state fQSM may also detect resting-state networks and provide a measure of functional connectivity complementary to rsfMRI. Here, aiming to improve upon previous single-echo fQSM studies, and building upon the advantages of multi-echo fMRI, I present optimisation of the first multi-echo high-resolution (1.3 mm isotropic) task-based fQSM, using a GRE-EPI sequence designed to provide simultaneous structural and functional QSM. I optimised acquisition and post-processing parameters including the acquisition multiband factor, denoising method, the QSM input and the smoothing kernel size. The optimised task-based fQSM was tested in healthy volunteers, compared to conventional magnitude-based fMRI, and used to investigate multi-echo vs single-echo fQSM, model-based vs data-driven functional analysis approaches and activations in veins, demonstrating that multi-echo fQSM yielded stronger activations than single-echo fQSM, and that as demonstrated in previous studies, fQSM has a higher spatial specificity than conventional fMRI. I also present the first resting-state fQSM study, where I successfully detected the DMN. This was achieved using two different approaches: seed-based connectivity analysis (with a medial prefrontal cortex seed) and independent component analysis. I found that rsfQSM was able to extract the DMN using both methods, demonstrating relatively good spatial similarity to conventional rsfMRI and the ground truth (GT) DMN from the Smith atlas. In the final chapter I present preliminary results applying the novel rsfQSM technique in a study of Parkinson’s disease patients compared to healthy controls. This thesis highlights the potential of the novel multi-echo task-based fQSM and rsfQSM as valuable tools for the study of brain function and disease in future.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item