Deep learning of multimorbidity trajectories in cardiovascular disease: capturing cross-talk between diseases with AI and big data

Cardiovascular disease remains the leading cause of death globally. Emerging evidence increasingly reveals the complex cross-talk between cardiac pathology and dysfunctions in other organs, highlighting diverse, overlapping multimorbidity trajectories. This growing complexity in disease progression and risk prediction calls for novel, data-driven approaches to aid the understanding of disease mechanism. 

Advances in artificial intelligence (AI) and the availability of large-scale population data, such the UK Biobank imaging cohort, provide powerful lens and materials to uncover the systemic nature of cardiovascular disease, but also capture the individual disease trajectories for precise and personalised interventions. 

This project aims to develop novel AI machine learning approach on UK Biobank imaging cohort, to:

• Capture image evidence of multi-organ associations in cardiovascular disease using machine learning for medical imaging;
• Study the associations of myocardial injuries identified by novel AI-enhanced imaging, such as virtual native enhancement of cardiac MRI, with conditions of multiple organs using advanced machine learning models.
• Develop personalised risk prediction and recommendation models based on the new findings.

You will benefit from an interdisciplinary core supervision team of machine learning scientists, biomedical statisticians and cardiologists, and develop and validate deep generative AI models at population scale. Skills training includes machine learning (deep learning), cardiac imaging and data analysis for large population studies. 

The project is in partnership with Oxford Centre for Clinical Magnetic Resonance Research (OCMR) which supports expertise in cardiac imaging and cardiovascular medicine.