AI algorithm detects asymptomatic heart disease from Apple Watch ECG data

A new algorithm developed by researchers at Mayo Clinic can effectively detect patients with severe heart failure using data collected by Apple Watch. A large trial is now underway looking to test the clinical benefit of the algorithm in 1 million people.

Although Apple introduced electrocardiogram (ECG) capabilities into its smartwatch in 2018, the technology has only recently begun to provide useful health reports. While the ECG feature on the Apple Watch can currently detect a type of arrhythmia called arterial fibrillation, it’s still often recommended as a tool to help detect heart abnormalities in conjunction with a doctor.

A traditional EKG test involves up to 12 electrodes attached to different parts of the body. These electrodes allow doctors to record the heart’s electrical activity and then detect a number of different heart abnormalities.

The Apple Watch ECG collects data from only one point on the wearer’s wrist, so of course it can’t be an accurate diagnostic tool like a 12-hour ECG test in the clinic. To turn the Apple Watch’s ECG data into something clinically useful, researchers must turn to artificial intelligence algorithms, designed to identify precise signals in the data that could correspond to heart problems.

Two years ago, a team at the Mayo Clinic developed a new algorithm that can automatically detect a heart condition known as left ventricular dysfunction using traditional 12-sided ECG data. The condition, informally referred to as heart pump dysfunction, is often asymptomatic and is estimated to affect nearly seven million Americans.

“Left ventricular dysfunction — an impairment of the heart’s pumping action — affects 2 percent to 3 percent of people globally and up to 9 percent of people over the age of 60,” explained Paul Friedman, a Mayo Clinic researcher working on the project. . Shortness of breath, leg swelling, or a fast heartbeat. The catch is that once we know the heart pump is weak, there are many life-saving treatments and symptom prevention. “

The new research, which was not published in the peer-reviewed journal, was recently presented at the Society of Cardiology’s conference. Led by Itzhak Zachi Attia, the research team adapted a previously developed algorithm to effectively interpret single-lead Apple Watch ECG data as if it were from a 12-sided device.

A small, initial six-month study looked at data from a number of people participating in the ongoing Mayo Clinic study. Participants shared a large volume of ECG data along with Apple Watch data to test how effective the new algorithm was in detecting a weak heart pump.

Attia said, “About 420 patients enrolled in an echocardiogram within 30 days had a clinically required echocardiogram, or cardiac ultrasound, which is a standard test to measure pump strength. We took advantage of this data to see if we could identify a weak heart pump. With AI analysis of the ECG clock. While our data is early, the test area was under the 0.88 curve, which means it’s as good or slightly better than the medical treadmill test.”

In early April, the Mayo Clinic launched a massive study looking to test a set of algorithms designed to predict heart disease with Apple Watch ECG data. The study hopes to enroll one million people who will be followed for one year, and provide Apple Watch ECG data to an anonymous database. This data will be retrospectively compared with each participant’s medical records to assess the quality of the AI ​​algorithm’s health predictions.

So, at best, it may be a year or two before the Apple Watch begins diagnosing people with this heart condition. However, Friedman is optimistic that this type of AI technology will transform medicine in the future, as increasingly inexpensive wearables and health tracking devices will be able to catch serious diseases in their early stages without requiring patients to be hospitalized and undergo expensive or cumbersome tests.

“It is quite remarkable that AI turns the ECG signal of a consumer’s watch into a detector for this condition, which would normally require an expensive and sophisticated imaging test, such as an echocardiogram, CT scan, or MRI,” Friedman added. The transformation of medicine looks like this: the inexpensive diagnosis of serious diseases from the couch. “

Source: Mayo Clinic

Leave a Comment

Your email address will not be published.