Medical image synthesis using generative machine learning models for testing public health hypotheses related to ischemic stroke risk (CSC ONLY)

Generative AI is increasingly being applied in medical imaging to enhance the quality of collected images or to extend imaging studies to include modalities that would otherwise be impossible to obtain. This can be especially useful when dealing with medical imaging modalities that are invasive and involve radiation exposure, and thus are not suitable for large-scale data collection in the general population.<br /> <br /> Your PhD project will develop new generative AI models for the cross-modality synthesis of brain of magnetic resonance angiograms in large population imaging databases. You will develop methods specifically designed to capture the details of the small, tortuous blood vessels that may have some significance in ischemic stroke risk. By applying your methods in a large population imaging database, you will conduct a study to explore the vascular risk factors behind ischemic stroke in the UK population.

<p>Stroke is the fourth single leading cause of death in the UK and the single largest cause of complex disability. Stroke is responsible for 34,000 deaths and 100,000 stroke victims annually in the UK. Ischemic stroke, i.e. blockage of the brain blood vessels due to clots, is the most common form of stroke. Some risk factors of ischemic stroke have been identified in clinical studies: high blood pressure, diabetes, atherosclerosis, atrial fibrillation, obesity, etc., but further studies are needed to identify risk factors so that mitigation can be attempted earlier. The UK Biobank (UKBB) is a longitudinal population study that collects multi-modal imaging, genetic, and clinical biomarkers to enable large-scale association studies linking genetic and lifestyle risk factors with the incidence of various diseases (including stroke). Some studies have indicated that the lack of certain collateral vascular pathways in the brain is strongly associated with ischemic stroke risk. However, direct study of the brain vasculature and its dozens of possible configurations cannot be performed in the UKBB as it does not contain brain vascular images.</p> <p>You will develop novel generative deep learning models to synthetise missing vascular brain images in the UKBB. Cross-modality image translation by deep neural networks is an emerging technology that is increasingly being used to extend clinical imaging studies to include different types of images. It uses source images from one modality (in this case, structural MR images) and learns to reproduce target images in another modality (in this case, angiographic MR or CT images) using deep neural networks, for example generative adversarial networks (GANs), variational autoencoders (VAEs), or diffusion probabilistic models (DPMs). You will develop novel generative models specifically targeting small-scale anatomical features (i.e. blood vessels) that need to be accurately recreated to characterise the circle of Willis, a redundant network of blood vessels that supplies blood to the brain. Since many existing image generative models fail to accurately reproduce small details of images or may hallucinate features, this project further develops anatomy-guidance and attention mechanisms to improve the continuity and anatomical fidelity of the generated vasculature.</p> <p>After developing the generative model, you will perform a validation study to demonstrate the reliability of the synthetic MRA/CT images and develop algorithms for the automatic segmentation and classification of circle of Willis phenotypes at scale. Then, using data from the UK Biobank brain MR imaging study, we will generate and classify synthetic vascular images a large population cohort. These synthetic data will be used to perform a cross-sectional proof-of-concept population imaging study to test the hypothesis that CoW phenotype identification can improve prediction accuracy for ischemic stroke risk when considered together with lifestyle and genetic risk factors.</p>

<p>Formal applications for research degree study should be made online through the <a href="https://www.leeds.ac.uk/research-applying/doc/applying-research-degrees">University's website</a>. Please state clearly in the Planned Course of Study section that you are applying for <em><strong>PHP Computer Science FT</strong></em> and in the research information section that the research degree you wish to be considered for is <em><strong>Medical image synthesis using generative machine learning models for testing public health hypotheses related to ischemic stroke risk</strong></em> as well as <a href="https://eps.leeds.ac.uk/computing/staff/1542/dr-toni-lassila"><strong>Dr Toni Lassila</strong></a> as your proposed supervisor. <em><strong>Please state clearly that you are applying for <a href="https://phd.leeds.ac.uk/funding/48-china-scholarship-council-university-of-leeds-scholarships-2026">China Scholarship Council-University of Leeds Scholarship 2026/27</a>.</strong></em></p> <p>Applications will be considered after the closing date.  Potential applicants are strongly encouraged to contact the supervisors for an informal discussion before making a formal application. We also advise that you apply at the earliest opportunity as the application and selection process may close early, should we receive a sufficient number of applications or that a suitable candidate is appointed.</p> <p>Please note that you must provide the following documents in support of your application by the closing date of Wednesday 7 January 2026:</p> <ul> <li>Full Transcripts of all degree study or if in final year of study, full transcripts to date including the grading scheme</li> <li>Personal Statement outlining your interest in the project</li> <li>CV</li> <li>Evidence of meeting English Language requirements (certificate must be provided by Wednesday 7 January 2026)</li> </ul> <p><em>As an international research-intensive university, we welcome students from all walks of life and from across the world. We foster an inclusive environment where all can flourish and prosper, and we are proud of our strong commitment to student education. Across all Faculties we are dedicated to diversifying our community and we welcome the unique contributions that individuals can bring, and particularly encourage applications from, but not limited to Black, Asian, people who belong to a minority ethnic community, people who identify as LGBT+ and people with disabilities. Applicants will always be selected based on merit and ability.</em></p>

Applicants to research degree programmes should normally have at least a first class or an upper second class British Bachelors Honours degree (or equivalent) in an appropriate discipline. The criteria for entry for some research degrees may be higher, for example, several faculties, also require a Masters degree. Applicants are advised to check with the relevant School prior to making an application. Applicants who are uncertain about the requirements for a particular research degree are advised to contact the School or Graduate School prior to making an application.

The minimum English language entry requirement for research postgraduate research study is an IELTS of 6.5 overall with at least 6.5 in writing and 6.0 in reading, listening and speaking or equivalent. The test must be dated within two years of the start date of the course in order to be valid.

<p>Highly Competitive full-time PhD funding available jointly from the China Scholarship Council and University of Leeds for 3 years. This opportunity is open to Chinese nationals only who fulfil the criteria of the China Scholarship Council. All eligible candidates will be evaluated based on academic merit only. The University of Leeds will provide full academic fees for 3 years and the China Scholarship Council will provide maintenance to cover living costs, flights and visa costs. Please note that you must complete your academic application for PhD study first to obtain your Student ID Number, then complete a separate application for the Scholarship. It is recommended that you read the <a href="https://phd.leeds.ac.uk/funding/48-china-scholarship-council-university-of-leeds-scholarships-2026">China Scholarship Council-University of Leeds Scholarship 2026/27</a> information, prior to applying.</p> <p>You must be able to start on 1 October 2026 as the start date is fixed under the terms and conditions of the scholarship.</p>

<p>For further information about your application, please contact PGR Admissions by email to <a href="mailto:phd@engineering.leeds.ac.uk">phd@engineering.leeds.ac.uk</a></p> <p>For further information about this project, please contact Dr Toni Lassila by email to  <a href="mailto:T.Lassila@leeds.ac.uk">T.Lassila@leeds.ac.uk</a></p>