Physical Sciences of Imaging in the Biomedical Sciences CDT
Thesis project - "Investigation of Brain Microenvironments Using Advanced Diffusion MRI"
Professor Andrew Peet, Cancer Sciences
Dr Jan Novak, Institute of Cancer and Genomic Sciences
Dr Hamid Dehghani, School of Computer Science
Molecular diffusion, the random motion of molecules suspended in a fluid resulting from collisions, is well understood in simple homogeneous systems. In complex multi-component systems such as brain tissue, the diffusion is non-linear and requires the acquisition and analysis of multi-exponential data leading to significant challenges in data acquisition and interpretation. Diffusion of water can be measured by magnetic resonance methods such as nuclear magnetic resonance (NMR) and magnetic resonance imaging (MRI). This is achieved by employing magnetic field gradients at varying strengths, referred to simply as b-values.
Diffusion data at low and high b-values has been found to show multi-exponential behaviour, indicative of the complex system which is being probed. This observation provides a challenge in terms of producing a realistic model system, which is required for standardisation of methods, biophysical interpretation of results and further MRI method development. In order to mimic biological tissue, a multi-compartment model is required to incorporate several different water diffusion environments. At present no such model system exists for use in MR, but tissue substitutes have been developed for studying other biophysical properties.
Despite the lack of through understanding of the link between tissue water diffusion and MR diffusion metrics, diffusion weighted (DW) magnetic resonance imaging is widely used throughout the medical community for the assessment of a variety of neurological conditions. In particular,the contrast in DW-MRI is known to be linked to the tissue cellularity and viability. However, in current clinical environments, diffusion imaging is only used on a qualitative basis partially due to the poor understanding of the relationship between water diffusion in tissue and MRI diffusion metrics. Improving this understanding has the potential to provide quantitative measures of the tissue microenvironment, such as microvascular density, tissue perfusion and tissue microstructure providing the basis for this investigation.
This PhD project will concentrate on developing methods to probe complex diffusion within biomedical systems using MRI. This will require the design and construction of model systems, which exhibit well-defined multi-component diffusion properties in conjunction with MRI methodology development. The methods for data analysis will be developed, which accommodate the complex non-linear diffusion processes found in biomedical systems. The ability of MRI to probe multi-component diffusion will allow the techniques to be integrated into clinical scanners to allow for radiological assessment. The methods will be implemented on a cohort of children with brain tumours to assess formally their value.
The main area of research in this PhD is based on MRI, which has a strong background in the physical sciences, in particular quantum mechanics and structural organic chemistry. Understanding will be gained in the theoretical aspects of MR and this knowledge will be consolidated through practical experience in operating a clinical scanner. This will provide the basis for MRI technique development, leading to an implementation of a method for advanced diffusion imaging.