AI accurately detects cardiac dysfunction using smartwatch ECG: Study

For this purpose, the researchers digitally enrolled 2,454 unique patients from 46 states in the US and 11 countries who sent 125,610 ECGs to the data platform from August 2021 to February 2022.

The study led to the following findings:

• Four hundred twenty-one participants had at least one watch-classified sinus rhythm ECG within 30 d of an echocardiogram, of whom 3.8% had an EF ≤ 40%.
• The AI algorithm detected patients with low ejection fraction with an area under the curve (AUC) of 0.885 and 0.881, using the mean prediction within a 30-d window or the closest ECG relative to the echocardiogram that determined the EF, respectively.

"The results suggest that consumer watch ECGs, acquired in nonclinical environments, can be used for the identification of patients with cardiac dysfunction, which is a potentially life-threatening and often asymptomatic condition," the authors wrote in their study.

According to the experts, "the approach could find a place in clinical practice if the algorithm is validated in future research and implementation issues are worked out. However, the approach should not replace echocardiography or another way of proper measurement of LV ejection fraction but is potentially useful as a screening tool."

The researcher stated, "we have been advised that, without the approval of the US FDA, the AI algorithm cannot be used in routine practice outside the Mayo Clinic."

Reference:

Attia, Z.I., Harmon, D.M., Dugan, J. et al. Prospective evaluation of smartwatch-enabled detection of left ventricular dysfunction. Nat Med (2022). https://doi.org/10.1038/s41591-022-02053-1