Google New Al Algorithm Can Predict Heart Disease by Scanning Your Eyes

Doris Richards
February 23, 2018

Google scientists trained their new algorithm to make predictions on data from more than 300,000 patients in the United States and United Kingdom, according to a report on the technology published in the Nature Biomedical Engineering Journal. The research was also shared before peer review last September.

Using retinal images, Google says it was able to quantify this link and predict a patient's risk of a heart attack or other major cardiovascular event.

The idea of analyzing cardiovascular risk based on mere eye scans may sound rather absurd, but it indeed has a strong medical basis.

Google's AI was able to differentiate patients who suffered a major cardiac event in the following five years and those who didn't with a 70 percent accuracy.

Data from over 280,000 patients was used to create the models; images from over 13,000 of these patients were used to validate them. "Because blood vessels can be non-invasively visualized from retinal fundus images, various features in the retina, such as vessel caliber, bifurcation or tortuosity, microvascular changes and vascular fractal dimensions, may reflect the systemic health of the cardiovascular system as well as future risk", the authors write.

Study coauthor Dr Michael McConnell, a medical researcher at Verily, said: 'Cardiovascular disease is the leading cause of death globally. Predicting the factors that put a person at risk of a heart attack or stroke was an offshoot of the original research.

The robot used a deep learning analysis to spot patterns in this information, learning to link telltale signs in the eye scans with the metrics needed to predict disease risk, such as age or blood pressure.

Lily Peng, a doctor and lead researcher on the project, says Google was surprised by the results. The exercise was then expanded to predicting whether the person was a smoker or what their blood pressure was.

This kind of technological solution could produce fast, cheap and noninvasive tests that could be administered in a range of settings.

Google scientists have reported that their method is even less accurate than other more traditional and more invasive techniques, so it needs more improvement before it is clinically exploited, which will take a few years.

Other reports by Iphone Fresh

Discuss This Article