Development of a Personalised Foundation Model for High-Risk Pregnancy using Real World Evidence (Prof Lucy Mackillop & Dr Huiqui Yvonne Lu)

PROJECT TITLE

Development of a Personalised Foundation Model for High-Risk Pregnancy using Real World Evidence.

SUPERVISORS

• Prof Lucy Mackillop

• Dr Huiqi Yvonne Lu

DESCRIPTION OF PROJECT

Electronic health records, physiological measurements, imaging, laboratory results, and clinical text provide rich longitudinal information for understanding high-risk pregnancies. This project will use real-world evidence from the Thames Valley and Surrey Secure Data Environment Maternity Mart to develop a personalised foundation model for predicting adverse pregnancy outcomes. The project will adapt pre-trained foundation models using federated learning, enabling models to learn across distributed maternity datasets without transferring identifiable patient-level data. The resulting model will integrate longitudinal maternal, fetal and clinical information to estimate individual risks of pregnancy complications, characterise personalised risk trajectories and support clinical risk stratification and decision-making.

TRAINING OPPORTUNITIES

• Co supervision from Department of Women’s and Reproductive Health and Department of Engineering.
• Opportunity to learn state-of-the-art artificial intelligence and machine learning (AI/ML) by engaging with cutting-edge AI/ML techniques, including advanced deep learning models and computer vision applications.
• Opportunity to work on a novel RWE maternity dataset.
• Opportunity to attend relevant courses offered at the University of Oxford.

Funding Information

The position is not currently funded, and therefore the candidate will need to secure funding. Supervisors are happy to support internal and external scholarship opportunities.

HOW TO APPLY

To apply for this research degree, please click here.