Google Claims It Can Predict Heart Disease Through People’s Eyes

Google Claims It Can Predict Heart Disease Through People’s Eyes
The eyes of Serbian tennis player Novak Djokovic a match at the Australian Open in Melbourne on Jan. 22, 2018. (Saeed Khan/AFP/Getty Images)
Colin Fredericson
2/20/2018
Updated:
2/20/2018

Google claims its technology can predict heart disease risk by examining eye scans.

Google has a machine learning algorithm that can deduce health information nearly as accurately as leading methods, simply by analyzing eye scans, according to The Verge.

“They’re taking data that’s been captured for one clinical reason and getting more out of it than we currently do,” Luke Oakden-Rayner, a medical researcher from the University of Adelaide, told The Verge. “Rather than replacing doctors, it’s trying to extend what we can actually do.”

The research behind the power of the eye to reveal the health of a person is already well-established, according to The Verge. Numerous blood vessels at the back of the eye can give a picture of characteristics like age, blood pressure, and whether or not a person smokes. Taken together, they can indicate heart health.

Google ran tests on 300,000 people to build its technology. They used machine learning to build an algorithm based on eye scans and other health tests from the individuals that were part of the data set, The Verge reported.

“The caveat to this is that it’s early, (and) we trained this on a small data set,” Google’s Lily Peng, lead researcher on the project, told USA Today. “We think that the accuracy of this prediction will go up a little bit more as we kind of get more comprehensive data. Discovering that we could do this is a good first step. But we need to validate.”

Google’s algorithm isn’t perfect. But it’s imperfections are on par with established methods. The algorithm could correctly pick out a patient that experienced a heart attack, stroke, or a related event 70 percent of the time. That’s on par with the common method that requires blood tests and has 72 percent accuracy, according to The Verge.

Alun Hughes, a professor at University College London thinks Google may be on to something due to a “long history of looking at the retina to predict cardiovascular risk,” but he still thinks the algorithm merits further tests to be certain, The Verge reported.

The Verge thinks Google’s research offers a new potential for the use of artificial intelligence, and a new direction for diagnosing medical conditions.

From NTD.tv
Recommended Video:
Colin is a New York-based reporter. He covers Entertainment, U.S., and international news. Besides writing for online news outlets he has worked in online marketing and advertising, done voiceover work, and has a background in sound engineering and filmmaking. His foreign language skills include Spanish and Chinese.