AI outperforms Framingham risk score in predicting heart disease from retinal scan

This AI uses a scan of your retina to predict your risk of heart disease

The researchers then used QUARTZ software to analyze retinal pictures of 7,411 people aged 48-92 and combined that data with their health history, such as their smoking habits, statin usage, and prior heart attacks, to predict the risk of heart disease. The health of participants was monitored for seven to nine-years, and the results were compared with Framingham Risk Score (FRS) prediction.

The FRS is a common tool to estimate heart disease risk. It uses age, gender and total cholesterol as well as high density lipoprotein (HDL) cholesterol, smoking, and systolic pressure to determine the likelihood that someone will develop a heart condition within a certain time frame, typically 10 to 30.

The QUARTZ team compared its data with 10-year FRS forecasts and found that the algorithm was as accurate as the conventional tool.


This AI Uses a Scan of Your Retina to Predict Your Risk of Heart Disease