Non-invasive quantification of cerebral oxygenation in ischaemic stroke using MRI

Thesis

Measurements of oxygen availability can be used to distinguish regions of the brain that are at risk of permanent damage during and after ischaemic stroke. This has the potential to inform management decisions, or monitor progress after treatment.

Differences in the oxygenation of blood (quantified as oxygen extraction fraction [OEF]) cause changes in the rate of reversible transverse relaxation, as measured using asymmetric spin echo (ASE) magnetic resonance imaging (MRI). This relationship is described by the quantitative blood oxygen level dependent (qBOLD) signal model. A streamlined version of the qBOLD model has recently been used to measure OEF in healthy subjects, and to detect changes in related parameters in ischaemic stroke.

In this thesis, a Bayesian framework for inferring on a multiple-compartment qBOLD model is developed, and applied to simulated and in vivo data to estimate OEF and other parameters. This model is then used to test potential improvements to the already established streamlined qBOLD framework. The possibility of acquiring data without a fluid-attenuated inversion recovery (FLAIR) sequence is investigated, and the complications that arise when cerebrospinal fluid contributes to the qBOLD signal are described. Then, modifications to the model that account for the effect of diffusion, and of differences in blood vessel distributions, are tested using Monte Carlo simulations, and validated in healthy subjects. These are tested alongside changes to other acquisition parameters that could lead to more efficient data collection.

Finally, the qBOLD model is used to infer OEF in ischaemic stroke patients. It is shown that oxygenation differs between pathological regions, which suggests that this method could be usefully applied to stroke assessment in the clinic.

Authors: Cherukara, M

• DOI: 10.5287/ora-xk9a7zoa8
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English