Their study is the first to show that it’s possible to use artificial intelligence to detect coronary artery disease (CAD) by analysing a person’s face.
Earlier research shows that certain facial features are associated with an increased risk of heart disease.
These include ear lobe crease, thinning or grey hair, wrinkles, xanthelasmata (small, yellow deposits of cholesterol underneath the skin, usually around the eyelids) and arcus corneae (fat and cholesterol deposits that appear as a hazy white, grey or blue opaque ring in the outer edges of the cornea).
For the study, trained research nurses took four facial photos and collected data on socioeconomic status, lifestyle and medical history from 5,796 patients.
Radiologists reviewed the patients’ angiograms and assessed the degree of heart disease depending on how many blood vessels were narrowed by 50 per cent or more.
This information was used to create, train and validate the deep learning algorithm. The researchers then tested the algorithm on a further 1,013 patients, finding that it outperformed existing methods of predicting heart disease risk, correctly detecting CAD in 80 per cent of cases.
