Over 350 Total Lots Up For Auction at Two Locations - NY 03/21, FL 03/22

Apple Watch algorithm developed at Mayo Clinic can diagnose weak heart pump

by John R. Fischer, Senior Reporter | May 18, 2022
Artificial Intelligence Cardiology
A new algorithm uses ECG data from the Apple Watch to identify weak heart pumps in people.
Mayo Clinic has developed an algorithm that, when used in conjunction with an Apple Watch, can effectively identify patients with a weak heart pump.

Clinically known as left ventricular dysfunction, a weak heart pump affects 2% to 3% of people worldwide and up to 9% of people over 60. It can have no symptoms but when it does, it is associated with shortness of breath, leg swelling or racing heart beats. Identifying the condition as early as possible can allow patients access to many lifesaving and symptom-preventing treatments.

Standard ECG uses 12 electrode leads strategically placed on a person’s chest, arms and legs to create a tracing that can evaluate the electrical signals of the heart. The AI application is less costly and picks up single-lead ECG tracings from the watch and securely transfers them. An abstract was presented on May 1 as late-breaking research at the Heart Rhythm Society conference. “The Apple Watch ECGs can readily be acquired from home, or while traveling, thus facilitating frequent assessments. It also can be done without having to remove your shirt or any clothing, whereas conventional ECGs are done in a medical environment and require partial disrobing,” Dr. Paul Friedman, chair of the department of cardiovascular medicine at Mayo Clinic in Rochester, told HCB News.

The app was developed by the scientists and Mayo Clinic’s Center for Digital Health to enable participants to use single lead ECGs from their watches. Friedman and his colleagues modified an established 12-lead algorithm for low ventricular ejection fraction, the main sign of a weak heart pump. The technology is licensed to Anumana, an AI-driven health technology company set up by nference and Mayo Clinic. They chose the Apple Watch because the company released its ECG data.

Enrolling via email, patients in the decentralized, prospective study just had to download the app. A total of 2,454 Mayo Clinic patients with an iPhone, the Mayo Clinic App and a series 4 or later Apple Watch took part in the study. Due to the high volume of participation, the research team says there is a possibility that a scalable tool could be developed for screening and monitoring heart patients wherever they are.

Participants then securely transmitted 125,610 ECGs from 46 states and 11 countries over six months. The app sent all previous watch ECGs and additional ones as they were recorded by patients to a Mayo secure data platform where they were analyzed. According to the authors, average app use was about twice a month and 92% used the app more than once. Each patient recorded many ECGs, with the scientists choosing the cleanest readings to assess.

They say it is only a first step and plan to conduct global prospective studies among more diverse populations to determine medical benefits and see if the watch helps create more equitable access to high-quality diagnostic care in real time. Friedman says that the same approach and technology could be used to diagnose other conditions. “We have used artificial intelligence to identify many cardiac and non-cardiac conditions, including the presence of silent arrhythmias (not present during the ECG acquisition), valvular heart disease, various heart muscle diseases, blood chemistry abnormalities and liver disorders. However, we would need to test, vet and verify that those AI ECG tests that have been developed using a 12-lead ECG also work on the Apple Watch.”

The study was funded by Mayo Clinic. Apple did not provide technical or financial support.

You Must Be Logged In To Post A Comment