Automated retinal microvascular quantification as a predictor of cardiovascular disease risk in UK Biobank
Kingston’s School of Computer Science and Mathematics and colleagues from St George’s University of London Community Health Sciences and University College London Institute of Ophthalmology have secured funding to understand how the shape and size of the retinal vessels are related to risk of cardiovascular disease in later life. The work builds on previous funding received from the Medical Research Council and it is hoped the results will help predict individuals at high risk of disease. The grant will fund post-doctoral researchers at Kingston and St George’s for two years to analyse the UK Biobank dataset. The UK Biobank is a major national health resource which includes probably the world’s largest retinal image repository (nearly 136,000 images) in a middle-aged population-based cohort study. Dr Roshan Welikala and Dr Sarah Barman lead on the Kingston image analysis team.
Contributors to the bid:
- Prof Christopher Owen, Dr Alicja Rudnicka (PI), Professor Peter Whincup, Professor David Strachan, St George’s University of London
- Dr Roshan Welikala, Dr Sarah Barman Kingston University
- Professor Paul Foster, University College London Institute of Ophthalmology
Funding body: British Heart Foundation
Morphological changes in the retinal vascular network are associated with future risk of many systemic and vascular diseases. However, uncertainty over the presence and nature of some of these associations exists. The extraction of quantitative measures from vessel morphology in large datasets are needed to help resolve these uncertainties, and this has been prohibitive to date given the considerable amount of manual operator involvement required. Advances in computational methods are providing opportunities to analyse retinal vessels with speed and precision. QUARTZ (QUantitative Analysis of Retinal vessel Topology and siZe) is a software system developed at Kingston University that provides an automated framework for extracting quantitative measures of vessel morphology from large numbers of retinal images. The UK Biobank is a major national health resource which includes probably the world’s largest retinal image repository (nearly 136,000 images) in a middle-aged population-based cohort study. The recent research award from the British Heart Foundation will enable us to evaluate and then implement our software on the entire UK Biobank retinal image dataset. Once processed, the data will be used in epidemiological studies by our collaborators at St George’s University of London, examining the association of vessel morphology to disease risk and outcomes.