Statistical characterisation of brain tissue microstructure with diffusion magnetic resonance imaging
This research project is in competition for funding with one or more projects available across the EPSRC Doctoral Training Partnership (DTP). Usually the projects which receive the best applicants will be awarded the funding. Find out more information about the DTP and how to apply.
In this project, you will introduce a novel representation of tissue based on statistical descriptors.
This approach has been used for a long time in engineering for describing microstructure of materials but never applied to brain tissue. This framework has the advantage of capturing the relevant moments for characterising tissue from a statistical perspective, and to evaluate changes due to degeneration.
Diffusion Magnetic Resonance Imaging (dMRI) is a medical imaging technique that allows to characterise brain structures in the microscale in vivo and non-invasively. The capabilities of dMRI to describe tissue at the cellular level depend on signal models providing relevant tissue parameters from MR acquisitions.
Most of the existing models assume simplified tissue characteristics (eg spherical/cylindrical cells), imposing a hard constraint to the validity of the results and the knowledge that can be extracted from them. Moreover, these signal representations do not take into account the statistical variability existing within a measuring unit (~1mm), limiting the scopes of the technique.
Project aims and methods
The research plan has four stages to be completed in 3.5 years:
Literature search, training in brain tissue characteristics and diffusion MRI
Analysis of statistical descriptors for representing brain tissue
Study of the relation between statistical descriptors and the corresponding dMRI measurements
Practical demonstrations, reporting, and viva
Those with an existing background in dMRI will be able to move into stage 2 sooner. Experiments will be performed numerically and in the Connectom scanner, a unique facility available in Cardiff University Brain Research Imaging Centre (CUBRIC) for characterising tissue microstructure based on diffusion experiments.
This project is particularly suited to students with an interest in MR-physics, statistical modelling, and numerical analysis with neuroimaging applications. You are expected to be knowledgeable in Matlab/Python and to hold a degree in Physics, Engineering, or similar.