Pushing the Boundaries of MR Physics
Using cutting-edge developments in magnetic resonance (MR) physics, we push our MR scanners to provide measurements of the brain that are more biologically specific. Two key areas of development are improvements in brain blood flow and brain tissues microstructure measurements.
We increase the efficiency of our MR imaging using advanced reconstruction, compressed sensing and image-readout techniques. Improved biophysical modelling helps us interpret these MR signals. We obtain extensive multi-contrast MRI signal databases using faster image-readout methods (such as spiral and radial) and integrated diffusion-relaxometry acquisitions to achieve acceleration and flexibility.
One key development builds on a recently patented technique: DIMAC – dynamic inflow magnitude contrast. This extremely fast single-slice imaging method uses inflow contrast to measure pulsatile flow in blood vessels to give measures of pulse wave velocity (i.e. arterial stiffness). Sensitivity is limited by image acquisition speed, so we develop ultra-fast imaging to go beyond standard cartesian k-space trajectories: taking advantage of the image sparsity to implement the latest MR physics techniques in radial acquisition, compressed sensing and multiband imaging to provide an ultra-fast MR sequence highly sensitive to pulsatile flow in small brain vessels.
Nowadays, Machine Learning and AI forms an integral part of neuroimaging. We used advanced AI models for data-driven feature extraction and selection, for biophysical model fitting, for image reconstruction and for real-time redesign of MR pulse sequences to account for motion in high field 7T MR systems.
We use diffusion MRI, a specialised imaging modality, to probe the microstructural properties of tissues. Diffusion MRI captures the movement of water molecules within tissue, providing unique insights into its underlying architecture. This advanced approach enables the non-invasive reconstruction of histology-like images, offering a groundbreaking alternative to traditional biopsy methods. This research has the potential to revolutionise non-invasive disease diagnosis by providing clinicians with precise, tissue-level information that is only accessible through invasive procedures.
Current efforts are concentrated on the technological development of diffusion MRI techniques and their application in cancer staging, where accurate characterisation of tumour microstructure is crucial for diagnosis, treatment planning, and monitoring therapeutic outcomes. By refining the ability to visualise and quantify tissue characteristics, BIG researchers aim to contribute significantly to the fields of oncology and precision medicine. These advancements could pave the way for earlier detection, more accurate disease assessment, and less invasive diagnostic workflows, ultimately improving patient care and outcomes.
In collaboration with colleagues from the Schools of Psychology and Welsh, we have been working on an innovative project that leverages real-time MRI to gain new insights into the Welsh language. This interdisciplinary initiative, titled Watch Your Welsh, explores the intricacies of speech production and aims to deepen our understanding of how the articulatory system works to produce the sounds of Welsh. Using cutting-edge real-time MRI technology, the project allows researchers to visualise the movements of the tongue, lips, and other speech-related structures in unprecedented detail as participants speak. By focusing on Welsh, a language with a rich phonetic and linguistic heritage, this work sheds light on how its unique features—such as mutations and long vowels—are articulated.
The findings from Watch Your Welsh have applications beyond linguistic study, including advancing speech therapy techniques, refining language learning tools, and contributing to the preservation and revitalisation of minority languages. By combining expertise in neuroimaging, linguistics, and psychology, the project highlights the power of interdisciplinary collaboration in addressing complex questions about language and cognition.
