Eye scan can reveal risk of having a heart attack or stroke

A simple digital photograph of the back of the eye can predict a major cardiovascular event—such as a heart attack or stroke—set to happen in the next decade with 70% accuracy, according to research supported by the British Heart Foundation and the National Institute for Health and Care Research (NIHR).

Researchers believe the routine retinal scans could also be used to track people’s heart health over time, as they also found links between someone’s three-year change in risk score and their odds of a major cardiovascular event.

The eye scan is analyzed using AI, which takes just a fraction of a second to produce a personalized risk prediction.

People at the highest risk could be referred to a GP, who might end up prescribing blood pressure tablets or statins to lower their cholesterol. In the future, researchers hope anyone getting an eye test could get an alert on their heart health pinged to their smartphone.

Dr. Ify Mordi, British Heart Foundation Research Fellow at the University of Dundee and consultant cardiologist, led the study published in the journal Cardiovascular Diabetology.

He said, “It may be surprising, but the eyes are a window to the heart.

“If there is damage or narrowing of the blood vessels at the back of the eye, there is a good chance that will also be seen in the blood vessels further inside the body, supplying the heart, which could lead to a heart attack or stroke.

“This is a one-stop scan which is routinely performed and takes less than a minute. It could be an important part of the package, alongside blood pressure and cholesterol checks, in identifying people who could benefit from medication or lifestyle changes.”

Researchers at the University of Dundee developed AI technology to analyze digital retinal photographs, which are usually part of a routine eye test on the high street.

First, the AI was instructed to look for red flags such as blood vessel narrowing, blockages and damage, which could be signs of impending heart problems. Then a “black box” approach was taken, allowing the technology to use machine deep-learning to search for any detail it chose in the images, which could range from the size to the arrangement of blood vessels.

After being trained on around 4,200 images, the black box AI was judged on how well the technology could predict those people who went on to have a heart attack or stroke, or die from cardiovascular disease—also usually a heart attack or stroke—in the next decade. The team found that black box AI predicted 70% of these cases when tested on eye scans from more than 1,200 people.

Some of these people had also undergone a second scan, three years after their first. When the researchers analyzed how risk scores identified by AI changed between the scans, they found the fifth of this group with the largest increase in their risk score had a 54% higher risk of a major cardiovascular event than the rest of the group.

This 54% higher risk was seen in people whose AI score increased by as little as 3% in three years—for example whose 10-year risk of a cardiovascular event rose from 20% to 23%.

The team also compared the AI technology to the prediction people get currently from routine health checks with their GP—their “cardiovascular risk score”—which provides a percentage risk of someone suffering a major cardiac event in the next decade, based on factors like their age, sex, blood pressure reading, cholesterol level and whether they smoke. Researchers found that the risk score and retinal scan identified almost exactly the same proportion of people at risk of a heart attack, stroke or death from cardiovascular disease.

When both a clinical risk score, the retinal scan and a further genetic test were combined, the accuracy rose to 73%. That means the three together could potentially identify an extra three people in every 100 people at risk of a heart attack, stroke or death from cardiovascular disease.

The AI tool was trialed in the study on people with diabetes, as they get routine retinal scans on the NHS to check for complications affecting the eyes. But researchers say analysis of blood vessels in the eye to judge cardiovascular risk should work for most people, and not just those with diabetes.

The study was a collaboration between clinical researchers who trialed the technology, including Dr. Mordi and Dr. Alex Doney, at the University of Dundee, alongside computer scientists who developed the AI technology, led by the university’s Professor Emanuele Trucco and Dr. Mohammad Syed.

Professor Bryan Williams OBE, chief scientific and medical officer at the British Heart Foundation said, “The more accurately we can detect someone’s risk of a heart attack or stroke, the better the opportunities to prevent these happening.

“Cutting-edge innovations, like the use of retinal scans alongside health checks, could play a role in improving risk prediction, which is important if we are to reach the British Heart Foundation’s goal to prevent 125,000 heart attacks and strokes in the UK by 2035.

“However, more research is needed to show this prediction accuracy is robust, and to determine the feasibility of incorporating retinal scans into clinical practice.”