Vascular Territory Map
Vascular territories above the circle of Willis
Dynamic angiography above the circle of Willis
University Research Lecturer
- Royal Academy of Engineering Research Fellow
- Head of Neurovascular Imaging
My research focusses on the development of novel non-invasive MRI methods which visualise blood flowing through the arteries that feed the brain and the resulting perfusion of the brain tissue. In particular, I am interested in techniques which allow blood from individual feeding arteries to be followed through the vascular tree. In addition to providing information about the structural and functional status of each artery, these methods allows the assessment of "collateral blood flow". This is when the main feeding artery to a certain brain region becomes blocked or significantly narrowed, but the flow of blood from secondary arteries maintains perfusion in this brain region, preventing a significant stroke. The presence or absence of collateral flow can be important in deciding between potential treatment options in patients with arterial disease. These vessel-selective strategies also have applications in diseases where the arterial source of blood flow is important, such as tumours and arteriovenous malformations.
My work is mainly based on a method known as "vessel-encoded arterial spin labelling". This allows blood in various combinations of arteries to be magnetically labelled. Acquiring a series of images with different combinations of tagged arteries allows the signal arising from each artery to be separated in post-processing whilst maintaining a high signal-to-noise ratio. I have applied this method to perform vascular territory mapping, in which the perfusion territories of the main feeding arteries to the brain are visualised. In addition, I have extended this approach to dynamically visualise blood flowing through the arteries feeding the brain, allowing assessment of arterial disease directly. I have also developed post-processing methods to extract quantitative blood flow rates from these images, which will enable a more objective assessment of each patient to be made.
I am currently trialling these new methods in collaboration with clinical colleagues in a range of patient groups, including those with chronic arterial disease, acute stroke and arteriovenous malformation. I hope that this will show the potential utility of these techniques for understanding the progression of these diseases, and in the longer term help with diagnosis, prognosis and therapeutic planning in these patients.
I have recently been awarded a five year research fellowship from the Royal Academy of Engineering, which provides me with an excellent opportunity to take forward this research. In this project, I aim to address the key downside to these imaging techniques, which is that obtaining 3D time-resolved images of the arteries as well as maps of tissue perfusion is time-consuming, and therefore difficult to apply in a clinical setting. I plan to design a single scan which can track the flow of blood through the arteries, all the way into the tissue, thereby providing both sets of information at the same time. I will use recently developed methods to accelerate this process, allowing images to be acquired in a fraction of the time normally required. I will also adapt mathematical models to allow a range of quantitative physiological information to be extracted from the images, before testing this new technique in a range of different patient groups.
A Frequency-Domain Machine Learning Method for Dual-Calibrated fMRI Mapping of Oxygen Extraction Fraction (OEF) and Cerebral Metabolic Rate of Oxygen Consumption (CMRO2)
Germuska M. et al, (2020), Frontiers in Artificial Intelligence, 3
Highly accelerated vessel-selective arterial spin labeling angiography using sparsity and smoothness constraints.
Schauman SS. et al, (2020), Magn Reson Med, 83, 892 - 905
Impact of Cardiovascular Risk Factors on Cerebrovascular Structure and Perfusion in Young Adults
Williamson W. et al, (2017), CIRCULATION, 136