Artificial intelligence lets your smartwatch spot heart failure

Artificial intelligence lets your smartwatch spot heart failure

  • News
  • Published on
    Updated


    Reading 2 mins.

    An American study reveals how two simultaneous technological advances can transform a smartwatch into a real heart failure diagnostic tool.

    Wearing a smartwatch might just become the simplest cardiovascular health check. Indeed, according to a study published in the journal NatureNovember 14, a team of researchers from Minnesota demonstrated how the ECG (electrocardiogram) data of the “carrier”, translated by an intelligence algorithm, makes it possible to identify heart failure.

    An app and a modified algorithm to translate heart health

    2,454 people, in the United States and 11 different countries, have downloaded an app created by the Mayo Clinic Center for Digital Health to securely share their Apple Watch ECGs in their electronic health record. A total of 125,000 Apple Watch ECGs were recorded between August 2021 and February 2022. About 420 participants underwent an echocardiogram, a standard test using sound waves to produce images of the heart within 30 days of recording. an Apple Watch ECG in the app. Of these, 16 patients had a low ejection fraction confirmed by echocardiogram, which provided a comparison to refine the results.

    The Mayo researchers also interpreted single-lead ECGs from the Apple Watch by modifying an earlier algorithm developed for 12-lead ECGs, which has been shown to detect heart failure. Although the data is early, the modified AI algorithm achieves a measure of accuracy as good or slightly better than a medical treadmill diagnostic test.

    Wearing a watch is easier than having a CT scan

    If the technology can sometimes seem complicated, this new method on the contrary would come to simplify the path of the patient, thanks to the only wearing of this watch.

    “Currently, we diagnose ventricular dysfunction through an echocardiogram, CT or MRI, but these are expensive, time-consuming and sometimes inaccessible” recalls Paul Friedman, lead author of the study. Rather, the results here suggest the potential for a relatively inexpensive and widely available tool to detect heart failure in people outside of a clinical setting.

    “These data are encouraging because they show that digital tools are enabling convenient, inexpensive and scalable screening for important conditions. Using technology, we can remotely gather useful information about a patient’s heart in a way accessible that can meet the needs of people where they are,” completes Zachi Attia, co-author of the study.

    According to the scientists, the innovation would also make it possible to go faster in the health journey: people with a weak heart pump may not show symptoms, whereas this heart disease affects around 2% of the population and 9% of people over 60 years old. Early diagnosis is important because once identified, there are many treatments to improve quality of life and decrease the risk of heart failure and death.

    dts1