Rina Ariga

I am interested in using computational approaches to patient data – ECG, imaging, genetics, blood biomarkers and health records – to improve mechanistic understanding of cardiovascular disease.

I currently work on phenotyping patients with Hypertrophic Cardiomyopathy, an inherited heart muscle disease with a risk of sudden cardiac death. Up to 60% are negative on gene testing and appear to be a clinically more heterogeneous group. Indeed some of these patients may be very low risk, yet they are followed up lifelong. My work aims to individualise patient care by using pathophysiological insights gained from deep learning of big data from UK Biobank and Hypertrophic Cardiomyopathy Registry.

The use of such artificial intelligence has the potential to revolutionise early disease detection and improve patient outcomes. Clinically, I champion the use of AI-enabled diagnostics in patient pathways to enable accurate, efficient and timely diagnosis particularly in Heart Failure.

My doctoral research pioneered Diffusion Tensor Cardiac Magnetic Resonance in patients with Hypertrophic Cardiomyopathy. This novel imaging method measures heart muscle disarray – a microstructural abnormality previously only seen at post mortem, which is believed to be the focus for fatal heart rhythms. In collaboration with Computational Cardiovascular Science, we applied machine learning and computational simulation to integrate ECG and imaging to detect subtle electrical changes indicative of such underlying defects. I have received several prizes and awards for my DPhil work, and the findings were reported in the media.