Google Retinal Scans Predict Heart Attacks

913 2018-02-21

Scientists working for Google’s parent company Alphabet have used artificial intelligence to determine a person’s risk of having a heart attack from their retinal scan.

The method—detailed in a paper published on Monday in the Nature journal Biomedical Engineering—involves analyzing blood vessels in an area of the eye called the retinal fundus.

The researchers from Verily, formerly known as Google Life Sciences, developed the algorithm in the hope of making accurate assessments of patients’ cardiovascular health more quickly and easily than current methods.

Training deep-learning models on data from more than a quarter of a million patients, the scientists were able to predict the cardiovascular risk factors that were not previously thought to be present in retinal fundus images.

The risk factors include the person’s gender, smoking status, blood pressure and age—estimated to within four years of the patient’s actual age.

“Most cardiovascular risk calculators use some combination of these parameters to identify patients at risk of experiencing either a major cardiovascular event or cardiac-related mortality within a pre-specified time period, such as 10 years,” the paper states.

“However, some of these parameters may be unavailable…We therefore explored whether additional signals for cardiovascular risk can be extracted from retinal images, which can be obtained quickly, cheaply and non-invasively in an outpatient setting.”

Deep-learning networks have already been applied to produce algorithms capable of diagnosing diseases like melanoma and blindness caused by diabetes.

Further tests are required before this latest method can be used within a clinical setting.

The researchers concluded: “The opportunity to one day readily understand the health of a patient’s blood vessels, key to cardiovascular health, with a simple retinal image could lower the barrier to engage in critical conversations on preventive measures to protect against a cardiovascular event.”

Comments
Change the code
Facebook Comments