New study, published in the British Journal of Ophthalmology, opens the way to the development of rapid and inexpensive cardiovascular screenings, if its results are confirmed in future clinical trials. These screenings will allow people to learn about their risk of stroke and heart attack without the need for blood tests or even blood pressure measurements.

“This AI tool could let someone know in 60 seconds or less their level of risk,“ said the leading study author Alicja Rudnicka in the interview with The Guardian newspaper. The study found that the predictions were as accurate as those made by modern tests.

The software works by analyzing the network of blood vessels contained in the retina. It measures the total area covered by these arteries and veins, as well as their width and degree of curviness. All of these factors affect a person’s heart health, allowing the software to make predictions about heart disease risk just by looking at a non-invasive image of the eye.

“The study adds to a growing body of knowledge that the eye can be used as a window to the rest of the body,” says Pearse Keane, researcher in ophthalmology and Artificial Intelligence analysis, not connected to the study in the interview with The Verge. “Doctors have known for more than a hundred years that you could look in the eye and see signs of diabetes and high blood pressure. But the problem was manual assessment: the manual delineation of the vessels by human experts.“ The use of machine learning, according to Keanea, can overcome this challenge.

Using AI to diagnose diseases based on eye scans has turned out to be one of the fastest-growing areas of machine learning medicine. The first-ever FDA-approved AI diagnostic device was used to screen for eye diseases, and studies show that AI can detect a range of ailments this way, from diabetic retinopathy to Alzheimer’s disease (an area of Keane’s own research). Tools applying these findings are at various stages of development, but questions remain about the reliability and universality of their diagnoses.

This recent study, by a team from St George’s University of London, was tested only on scans of the eyes of white patients. The team obtained data for testing from the UK Biobank, a database that is 94.6% white (which includes UK demographics for the age range of patients included in the Biobank). Such biases must be balanced in the future so that any diagnostic tool is equally accurate across ethnic groups.

The researchers compared the results of their software, called QUARTZ (inventive acronym derived from the phrase “QUantitative Analysis of Retinal vessels Topology and siZe”, with 10-year risk predictions obtained using the standardized Framingham Risk Score test (FRS). They found that the two methods have “comparable performance“.