5. Causes and consequences of valvular heart disease: analyses of large-scale datasets
Valvular heart disease (VHD) refers to a group of conditions which affect an increasing number of adults worldwide.
Causes and consequences of valvular heart disease: analyses of large-scale datasets
Valvular heart disease (VHD) refers to a group of conditions which affect an increasing number of adults worldwide. Despite this, our knowledge of its burden, underlying causes and consequences is limited. Indeed, a large proportion of VHD is still considered to be 'degenerative' with no clear understanding of its causes and no established preventative strategies. Most previous research in this field has been based on small-scale mechanistic studies or cross-sectional studies with their inherent limitations. Recent accumulation of Big Data from routine health records (such as the UK Clinical Practice Research Datalink [CPRD]), registries and large-scale cohorts (such as the UK Biobank) provides an unprecedented opportunity to investigate the burden and determinants of VHD and help identify potentially modifiable risk factors. Early analyses from our group have shown highly promising results. The more recent imaging substudy of the UK Biobank, which will involve cardiac MRI and echo studies from up to 100,000 participants, will enable more in-depth analyses of the causes and consequences of VHD. For this DPhil project, the student will use very large datasets from different to investigate the epidemiology of VHD with cross-validation in different cohorts. The specific focus of the studies to be undertaken will depend the student’s skills and interests. As a member of the team, the student will become involved in the design, conduct and interpretation of related Big Data projects.
This research opportunity seeks a candidate with a quantitative background (e.g. MSc in biostatistics, epidemiology or computer science) with interest in public health and cardiovascular medicine. The project is also suitable for a student with medical background, in which case some experience with use of R and basic understanding of epidemiological concepts are essential. For clinical candidates, learning opportunities include training in advanced statistical methods, epidemiology and usage of statistical packages such as R. For non-clinical candidates, learning opportunities involve training in clinical epidemiology as well as machine learning to interrogate some of world’s largest and most complex datasets to address questions of high relevance to public health globally. This project will be part of a new interdisciplinary programme entitled ‘Big Data Science in Healthcare and Medicine’ at the Oxford Martin School. The research team provides an exceptional educational environment with expert individual supervision and support from several of experienced and enthusiastic researchers with backgrounds in clinical medicine, statistics, epidemiology and engineering. Further support in grant writing and high-impact scientific publications will be provided.