The power of AI: Advanced algorithm IDs heart issues using Apple Watch data
Artificial intelligence (AI) can be applied to electrocardiogram (ECG) data captured by smartwatches to identify patients with weak heart pumps, according to a new analysis published in Nature Medicine.[1]
A team of cardiologists from Mayo Clinic led the analysis, exploring ECG data from nearly 2,500 Mayo Clinic patients from all over the world. The mean patient age was 53 years old, 56% of patients were women and all ECGs were captured remotely from August 2021 to February 2022.
The group applied a 12-lead algorithm for low ventricular ejection fraction to the ECG data, finding that it had an area under the ROC curve of 0.88 for detecting signs of ventricular dysfunction. A total of 421 patients had an echocardiogram within 30 days of logging their ECG data into their Apple Watch, providing the team with a way to confirm the accuracy of their strategy.
“Currently, we diagnose ventricular dysfunction―a weak heart pump―through an echocardiogram, CT scan or an MRI, but these are expensive, time consuming and at times inaccessible,” senior author Paul Friedman, MD, chair of Mayo Clinic’s department of cardiovascular medicine, said in a prepared statement. “The ability to diagnose a weak heart pump remotely, from an ECG that a person records using a consumer device, such as a smartwatch, allows a timely identification of this potentially life-threatening disease at massive scale.”
“Building the capability to ingest data from wearable consumer electronics and provide analytic capabilities to prevent disease or improve health remotely in the manner demonstrated by this study can revolutionize health care,” added co-author Bradley Leibovich, MD, medical director for the Mayo Clinic Center for Digital Health. “Solutions like this not only enable prediction and prevention of problems, but also will eventually help diminish health disparities and the burden on health systems and clinicians.”
Mayo Clinic helped develop the advanced algorithm used for this analysis with nference, licensing it to Anumana. Some Mayo Clinic specialists have a financial interest in the algorithm. Also, Mayo Clinic fully funded this analysis—Apple provided no technical or financial support.
Michael has more than 18 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.