This category contains the results data for UK Biobank project reference 338, and associated publication Suitability of UK BIOBANK Retinal Images for Automatic Analysis (PLOS One, 2015) This project is intended to analyse 2,000 colour fundus images from the UKBiobank with VAMPIRE (Vascular Assessment and Measurement Platform for Images of the REtina) . Even though our software involves computerised analysis, it is semi-automatic and thus requires a small degree of user interaction - i.e. 10-15 minutes per image. Hence, we intend to use a small group (5-10) of trained operators to measure the 2,000 images over a 3 month period. Not only will this generate valuable retinal vascular measurements for inclusion in the UKBiobank and for subsequent cross-validation studies with other software, it will also allow to assess the feasibility of training non-ophthalmic specialists in retinal image assessment using VAMPIRE.
Using the Vascular Assessment and Measurement Platform for Images of the Retina (VAMPIRE) to derive retinal outcomes
The retina is one of the few places in the human body allowing easy, non-invasive observation of blood vessels, and there is mounting evidence that features associated with retinal vessels are early indicators of disease such as diabetes, hypertension, stroke and cardiovascular disease. Furthermore, blood vessels of the retina are thought to predict neurovascular disease in particular because they are part of the brain's vascular system, so they share anatomical features and respond similarly to stress and disease. Retinal imaging might provide an additional means to stratify risk and help identify people who would benefit from early lifestyle changes and preventative therapies. By applying computational analysis to fundus images it is possible identify subtle vessel abnormalities which in turn may act as biomarkers of disease. We propose to assess the feasibility of using trained individuals to analyse fundus images via computerised software (VAMPIRE) on a subset of 2,000 fundus images (1,000 with myocardial infarction and 1,000 controls). This project will also generate retinal vascular measurement for use by other researchers. This project requires baseline data on eye images and data on cardiovascular risk factors (e.g., age, sex, smoking, obesity, family history, etc.) in a subset of UK Biobank participants.
|Lead investigator:||Dr Thomas MacGillivray|
|Lead institution:||University of Edinburgh|
|165||Suitability of UK BIOBANK Retinal Images for Automatic Analysis of morphometric properties of the vasculature||MacGillivray et al||2015||PLOS One 2015|