Scientists at Google have discovered a new eye test that can tell if a person is susceptible to heart disease, according to new findings published in the journal Nature Biomedical Engineering
The team made this discovery by using a combination of retina scans and artificial intelligence to detect the chances a patient will suffer a heart attack or stroke. That allowed them to predict with 70 percent accuracy if a patient would experience a major cardiovascular event within five years.
Such results show the tests were just as accurate as invasive methods — including sticking a needle into the arm — when it came to predicting cardiovascular disease.
To get their analysis, scientists took results based on data from 284,335 patients and validated it on two independent data sets of 12,026 and 999 patients. The technique allowed researchers to build a “heatmap.” That graphical representation of the data revealed which pixels in an image were the most important for predicting a specific risk factor.
For example, the team’s algorithm focused much more on blood vessels for making predictions about blood pressure.
“Pattern recognition and making use of images is one of the best areas for AI right now,” explained Harlan M. Krumholz, a professor of medicine at Yale University who was not involved in the study, according to USA Today.
However, as interesting as the new findings are, much more research needs to be done before any conclusive findings can be made. Even so, the study could help scientists better understand certain processes and diagnoses in new ways.
That time is a few years off, but researchers also believe the AI used in the study could have other applications as well.
“I am very excited about what this means for discovery,” said study co-author Lily Peng, a researcher at Google, according to UPI. “We hope researchers in other places will take what we have and build on it.”