FDA issues EUA for heart-dysfunction screening tool using machine learning

May 14, 2020

The U.S. Food and Drug Administration (FDA) has issued Emergency Use Authorizations (EUA) for the Eko electrocardiogram (ECG) Low Ejection Fraction Tool (“ELEFT”)—a machine learning algorithm, the agency said in a press update.

Separately, the FDA also granted EUA for the Ascom teleCARE IP Nurse Call System.

For the Eko electrocardiogram (ECG) Low Ejection Fraction Tool is a machine-learning algorithm, the FDA concluded that the tool can be used to provide an assessment of Left Ventricular Ejection Fraction in situations where an electrocardiogram is not immediately available or is not medically indicated.

In explaining its decision, the agency noted the “published evidence that patients with cardiac comorbidities are at substantially increased risk for complications from COVID-19, and COVID-19 may exacerbate or introduce cardiac comorbidities, which is associated with higher risk of in-hospital mortality.”

According to the FDA, the ELEFT algorithm was trained on and evaluated with datasets containing linked ECG and echocardiogram records from electronic health records. The validation was performed using retrospective and prospective methods to define the performance of the algorithm.

Meanwhile, the teleCARE IP Nurse Call System is designed for use in healthcare environments, including temporary hospital facilities, for remote communication between patients and healthcare providers. The FDA said that the remote communication and monitoring capabilities of the nurse call system may decrease the amount of contact healthcare providers have with patients who are in isolation rooms, which would reduce healthcare providers’ risk of exposure to SARS-CoV-2, the virus that causes COVID-19.

Visit the FDA for more news