AI uses eyes to identify heart disease with up to 80% accuracy

Artificial Intelligence has been used to help doctors examine eye scans and see if patients are at risk for heart disease.

In eye scans, doctors can find that blood vessels in the retina, a thin light-sensitive tissue in the back of the eye, indicate an increased possibility of suffering from heart disease. A key to sensing light, the retina also has a vast network of blood vessels that can be damaged by heart conditions or high blood pressure, according to Harvard Medical School.

In a study published by the peer-reviewed journal Nature Machine Intelligence, researchers at the University of Leeds used deep learning techniques. This is a machine learning technique that allows computers to analyze data and predict outcomes, to train AI to instantly view eye scans and recognize patients who were likely to suffer a heart attack in the following year.

Researchers discovered that the AI had 70-80% accuracy in identifying patients who were at risk of heart disease.

“This technique opens up the possibility of revolutionizing the screening of cardiac disease,” said Professor Alex Frangi at the University of Leeds in a press release.

“Retinal scans are comparatively cheap and routinely used in many optician practices. As a result of automated screening, patients who are at high risk of becoming ill could be referred for specialist cardiac services.”

According to the Centers for Disease Control and Prevention, every 36 seconds, a person dies from cardiovascular disease in the United States, which makes heart disease the leading cause of death in America.

This practice allows for AI to find early signs of heart disease, which can allow doctors to prescribe preventative treatments to avoid the condition from developing further. Using eye scans serve as a noninvasive alternative to determining one’s risk of heart disease, according to the American Heart Association.

For more content like this, sign up for the Pulse newsletter here.