Google builds AI that can predict heart disease

Algorithm assesses risk by scanning patient’s eyes

Google
The system uses eye scans to assess factors such as blood pressure
(Image credit: Chris Jackson/Getty Images)

Medical professionals may soon be able to predict heart disease through a non-invasive eye test, thanks to a new artificial intelligence (AI) programme developed by Google and its subsidiary Verily.

A newly published study in the journal Nature Biomedical Engineering reveals how scientists from the two firms use an AI algorithm to accurately identify a patient’s blood pressure, age and whether they smoke, by analysing a scan of the back of their eye.

The programme then combines the information in order to evaluate the patient’s risk of suffering a major cardiac event such as a heart attack, The Verge reports.

Subscribe to The Week

Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

SUBSCRIBE & SAVE
https://cdn.mos.cms.futurecdn.net/flexiimages/jacafc5zvs1692883516.jpg

Sign up for The Week's Free Newsletters

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.

Sign up

Luke Oakden-Rayner, a medical researcher at the University of Adelaide who specialises in machine learning analysis, told the website that the AI algorithm could improve existing analytic tools in the industry.

“They’re taking data that’s been captured for one clinical reason and getting more out of it than we currently do,” he said. “Rather than replacing doctors, it’s trying to extend what we can actually do.”

But the study “isn't without limitations”, says Engadget, as the scientists “only surveyed eye images with a 45-degree field of view”.

Further research will help determine whether the programme needs to be adjusted to analyse larger or smaller images, the website adds, while a “larger data set” would help improve the AI system’s accuracy.

“In other words, it’s not yet ready for clinical testing, but it’s a promising start for non-invasive evaluation of cardiovascular health,” Engadget concludes.