Cardiovascular Event Prediction Using Retinal Images & Patient Physiological Information
Approved Research ID: 77343
Approval date: October 11th 2021
The aim of this project is to build a tool to identify people at high risk of heart attack and stroke using images of their retina.
The retina is a layer of tissue at the back of the eye which provides a convenient window into vascular health, as it is easy to image, and clearly shows small blood vessels. Previous research using UK Biobank data has shown that it is possible to use retinal images to predict an individual's likelihood of suffering a heart attack or stroke. We will build on this research to make a product that informs people of their risk when they go for routine eye checks at optometrists.
This has the potential to have a massive impact on public health. Hundreds of thousands of heart attacks and strokes occur in the UK every year, making them the second most common cause of death and costing the NHS £7 billion annually. The majority of these are preventable with early treatment and lifestyle changes, yet millions of people are unaware of their risk. Deployment in an optometry setting would allow cheap, fast screening of millions of people every year, enabling earlier intervention and help stop thousands of deaths and disabilities annually while reducing the burden on the NHS.
The project will take place over 36 months, and involve evaluating the efficacy of the screening method on UK Biobank as well as external datasets to ensure suitability for deployment in an optometry context.