Google AI Researchers Find Heart Attack Warnings in Eye Scans

Google AI Researchers Find Heart Attack Warnings in Eye Scans

- March 9, 2018

Cardiovascular disease is the leading cause of death in the world. But now Google has developed an artificial intelligence software that can come close to predicting a person’s risk of heart attack just by looking into their eyes.

Nature Biomedical Engineering journal recently published a study to the effect that an algorithm developed by Google AI and Verily Life Sciences, through a photograph of the retina, can forecast if a person is likely to have a major cardiovascular event, such as a heart attack or stroke within five years.

The retina seems to be a corridor into what ails the body.  Physicians can spot such medical problems as diabetes, high blood pressure, high cholesterol, and some cancers during eye exams. According to a report in Business Insider, the eye scan gave enough info to let the algorithm predict a person’s age to within 3.26 years, smoking status with 71% accuracy and blood pressure within 11 units of the upper number reported in their measurement.

According to the study, data from 150 patients yielded encouraging results. “When the researcher presented the algorithm with two retina images and asked it to predict which one would suffer a major cardiac event or stroke, it predicted the correct scan 70% of the time.”

Still, Dr. Michael McConnell, the Head of Cardiovascular Health Innovations at Verily wrote that more work must be done to verify these findings on larger patient cohorts.

The new study demonstrates that deep learning is transforming how scientists study the body. According to the article, machine learning can take scans that we already have and use them to generate a more complete picture of human health.

Leave a Comment