Cardiovascular disease (CVD) is one of the leading causes of death in the UK. The statistics are stark: 7.6 million British people are living with CVD, 160,000 die every year from the condition, and probably half of us will develop CVD in our lifetime (British Heart Foundation, 2023).

The good news is that it is preventable. Regular blood pressure and cholesterol checks can help identify those who are at risk, and taking medication can reduce the likelihood of an individual developing the condition – as can giving up smoking, eating a healthy diet and keeping to a moderate weight.

But what if we could identify those at risk simply by analysing a photograph showing the veins and arteries in the back of the eye? The findings of a recent study suggest that artificial intelligence (AI) software trained on thousands of fundus photographs can predict with a high degree of accuracy a person’s risk of developing heart disease (Rudnicka et al, 2022). The research opens up the possibility that a high-street optometrist, equipped with a colour fundus camera, could provide patients with a near-instant CVD risk score. Patients could then consult their GP about appropriate preventive steps.

The study builds on existing knowledge that changes to the vessels in the eye can be indicative of problems elsewhere. “There has been a lot of work for quite some time looking at how the retinal vessels relate to traditional cardiovascular risk factors,” says Alicja Rudnicka, Professor in Statistical Epidemiology at St George’s, University of London, and study lead. The microvascular structure of the eye, she points out, matches the microvasculature elsewhere in the body – but the eye vessels are the only ones we can study in a completely non-invasive way.