ROCHESTER, Minn.— Scientists are developing a way for Apple Watches to detect people with weak heart pumps. The AI algorithm could enable early intervention measures to prevent symptoms from worsening.
Previously, detecting a weak heart pump required 12 electrodes placed around a person’s chest, arms and legs to create a full picture of the heart’s electrical signals. But working with Anumana Inc., an AI-focused health technology company, Mayo Clinic scientists tweaked the 12-lead algorithm for low ventricular ejection fraction into a single signal monitoring derivation.
“Left ventricular dysfunction – a weak heart pump — affects 2-3% of people worldwide and up to 9% of people over the age of 60. It may have no symptoms or be associated with shortness of breath, swelling in the legs, or racing heartbeat. What’s important is that once we know a weak heart pump is present, there are many treatments that can save lives and prevent symptoms. It is absolutely remarkable that AI transforms a consumer watch ECG signal into a detector for this condition, which would normally require an expensive and sophisticated imaging test, such as an echocardiogram, CT scan, or MRI,” says Dr. Paul Friedman, chairman of Mayo’s Department of Cardiovascular Medicine. Clinic, in a statement.
The study involved recruiting participants who downloaded an app that securely transferred ECG data from their Apple Watch in the background. The scientists used an adaptation technique to translate the single-lead readings into signals the algorithm can understand.
The algorithm recorded 125,610 ECG signals from Apple Watches in more than 46 states and 11 countries over a six-month period. People commonly use the app twice a month with an overall 92% app engagement.
“About 420 patients had a watch ECG recorded within 30 days of a clinically ordered echocardiogram, or ultrasound of the heart, a standard test for measuring pump strength. We took advantage of this data to see if we could identify a weak heart pump with an AI analysis of the watch’s ECG Although our data is early, the test had an area under the curve of 0.88, meaning it’s as good as or slightly better than a medical treadmill test. The AI analysis of the watch’s ECG is a powerful test to identify a weak heart pump,” says Dr. Attia.
“Ongoing AI research in cardiology is part of Mayo’s commitment to bringing digital transformation to healthcare. Advanced diagnostics that once required a trip to a clinic can be performed with precision, as this Apple Watch ECG study shows, from the wrist of a patient, whether they live in Brazil or Baton Rouge,” says the Dr. Bradley Leibovich, Medical Director of the Center for Digital at Mayo Clinic. Health ads. “App-based access to a medical center can help reduce health disparities by making high-level diagnostics accessible to more people in real time.”
The study was presented at the Heart Rhythm Society conference.