Off-the-Shelf Smartwatches, Enhanced with an "AI ECG" Algorithm, Prove Key for Early Diagnoses
Using an algorithm originally developed for 12-lead ECGs used in hospitals, patients' Apple Watches have been shown to offer early warnings.
Researchers at the Mayo Clinic have developed an artificial intelligence system capable of detecting patients' weak heart pumps through an off-the-shelf smartwatch — and have found that embedding AI into electrocardiogram (ECG) sensors can help identify patients at risk of strokes and cognitive decline up to 10 years earlier than traditional methods.
"Left ventricular dysfunction — a weak heart pump — afflicts two percent to three percent of people globally and up to nine percent of people over age 60," explains Paul Friedman, MD, chair of the Mayo Clinic's department of cardiovascular medicine. "It may have no symptoms, or be associated with shortness of breath, leg swelling, or racing heart beats.
"What is important is that once we know a weak heart pump is present, there are many lifesaving and symptom-preventing treatments available. It is absolutely remarkable that AI transforms a consumer watch ECG signal into a detector of this condition, which would normally require an expensive, sophisticated imaging test, such as an echocardiogram, CT scan, or MRI."
The researchers' work centered around single-lead ECG traces obtained from an Apple Watch, using a machine learning algorithm that had previously shown promise when used with the 12-lead ECG machines in common use in hospitals.
"Approximately 420 patients had a watch ECG recorded within 30 days of a clinically ordered echocardiogram, or ultrasound of the heart, a standard test to measure pump strength," explains Itzhak Zachi Attia, PhD, the department's lead AI scientist. "We took advantage of those data to see whether we could identify a weak heart pump with AI analysis of the watch ECG.
"While our data are early, the test had an area under the curve of 0.88, meaning it is as good as or slightly better than a medical treadmill test. AI analysis of the watch ECG is a powerful test to identify a weak heart pump."
"This test is the first step, as it demonstrates we can get medically useful information from a single-lead watch," adds Friedman. "Our next steps include global prospective studies to test this prospectively in more diverse populations and demonstrate medical benefit. This is what the transformation of medicine looks like: inexpensively diagnosing serious disease from your sofa."
At the same time, the Mayo Clinic has released a population-based study showing that an atrial fibrillation-detecting algorithm linked to ECG readings can provide a warning of stroke risk up to 10 years before a clinical diagnosis. "Application of this AI-ECG algorithm may be another way to screen individuals," explains Jonathan Graff-Radford, MD, corresponding author of the study, "not only to determine risk of atrial fibrillation, but also to identify future risk of cognitive decline and stroke."
The weak pump research was presented at the Heart Rhythm Society conference this weekend, but has not yet been published publicly; a copy of the atrial fibrillation study is available from the journal Mayo Clinic Proceedings (PDF).
