In China, medical experts and computer scientists are working together to develop AI-powered technology that can detect coronary artery disease through facial images.
In recent years, AI-powered applications have been used in daily clinical practice including analyzing electrocardiograms, tracking vital signs, and interpreting medical images.
In the latest study, Chinese researchers explored the possibility and feasibility of using AI to detect coronary artery disease via screening facial images.
Experts have always identified facial appearance as an indicator of cardiovascular risk. For instance, some of the most common predictors are features such as male pattern baldness, xanthelasmata, earlobe crease, and skin wrinkling.
For the study, researchers at China’s National Center for Cardiovascular Diseases and Tsinghua University have enrolled as many as 5,796 Chinese patients. The participants had to undergo several heart imaging tests and had their facial photos taken. They also had to answer questionnaires about their lifestyle, medical history, and social-economic status.
Based on this data, an AI algorithm was developed and trained. It was then tested on 1,013 facial images of other patients across nine Chinese hospitals.
According to the results, published in the European Heart Journal, the algorithm outperformed the traditional prediction model of coronary artery disease as it had a sensitivity of 80% and specificity of 54%.
Further, the researchers said that the studies are required to make a practical application of the algorithm. The algorithm’s low specificity raises concerns of false-positive results that may confuse both clinicians and patients.
Overall, the results holding promise for pre-test screening of the disease. It suggests a deep learning algorithm based on facial images that can assist in coronary artery disease detection.
In an editorial that was published in the same journal, researchers from the University of Oxford said “using selfies as a screening method can enable a simple yet efficient way to filter the general population towards more comprehensive clinical evaluation,” and “the full potential of such novel and out-of-the-box diagnostics lies ahead of us.“