Prospective Students

If you are a Warwick student looking for undergraduate or masters projects with me: please see this pageLink opens in a new window.

If you are interested in applying to Warwick under my supervision: Please see our PhD Link opens in a new windowor MSc by ResearchLink opens in a new window pages for details on these courses as well as possible funding options.

I am currently only accepting students with self or scholarship funding or the potential of getting scholarship funding. Please feel free to reach out to me via email with a CV, details of research interests, grades in previous degrees (esp. in machine learning/AI, mathematics and programming related subjects) as well as possible funding options.

Research Directions

I conduct research in AI for Biology and Medicine in the following key directions. For my research interests, please visit my publications pageLink opens in a new window.

• Multimodal Foundation Models for Biomedicine
  Developing and benchmarking large-scale, transferable models that learn unified representations from tissue images, spatial transcriptomics, genomics, and clinical data. This direction focuses on pretraining and adapting multimodal foundation models that can generalise across tasks such as diagnosis, prognosis, risk stratification, and treatment response prediction across diseases and cohorts.
• Agentic AI for Scientific and Clinical Workflows
  Designing agent-based AI systems capable of planning, reasoning, and executing end-to-end biomedical analysis pipelines. These systems coordinate tools for image analysis, molecular data processing, statistical modelling, and visualisation, enabling semi-autonomous discovery, reproducible research workflows, and clinical decision support.
• Spatial and Geometric Learning in Biology and Medicine
  Building graph, geometric, and spatial learning models to capture cell–cell interactions, tissue architecture, and molecular communication. The goal is to discover spatial biomarkers, characterise tissue microenvironments, and link spatial organisation to disease mechanisms and patient outcomes.
• Interpretable, Causal, and Robust AI in Healthcare
  Developing machine learning models that are transparent, reliable, and clinically meaningful. This includes causal representation learning, confounder-aware modelling, and methods for explaining predictions and ensuring that models generalise across institutions, technologies, and patient populations.
• Translational AI for Real Clinical Problems
  Applying modern AI methods to high-impact clinical challenges such as predictive modelling of endometrial dysfunction, colorectal cancer progression, treatment response, and patient stratification. Projects are conducted in close collaboration with clinicians and focus on moving from methodological innovation to deployable tools that can support real clinical and biomedical decision making.