FDA Clears Heart Murmur Detection AI Software for Eko Smart Stethoscope

The US Food and Drug Administration on July 12 granted clearance to digital health company Eko for its Eko Murmur Analysis Software (EMAS) for detecting and characterizing heart murmurs in adult and pediatric patients, according to a company statement. EMAS allows the Eko smart stethoscope to differentiate between innocent and structural heart murmurs indicative of valvular heart disease.

Pathological heart murmurs that may signal valvular heart disease can go undetected in examinations using a traditional stethoscope in general practice. In fact, by some estimates as much as 57% of clinically significant valvular heart disease may be missed, according to the Eko statement.

EMAS AI-powered algorithms have consistently identified valvular heart disease with an overall sensitivity of 85.6% and specificity of 84.4%, Eko says. That accuracy increases in adults aged ≥18 years, where the algorithm detected structural murmurs with a sensitivity of 90.2% and specificity of 90.6%. The company cites data showing that when clinicians use standard stethoscopes, their sensitivity for murmur detection drops to 44% and specificity slips to 60%.

Eko smart stethoscopes were first FDA cleared in 2020 to automatically detect signs of atrial fibrillation and heart murmurs in adults. EMAS, a cloud-based machine learning algorithm, now will analyze heart sounds and phonocardiogram and ECG signals captured by the smart stethoscope. FDA cleared EMAS to automatically detect murmurs and also the murmur timing and character, including S1, S2, innocent heart murmurs, structural heart murmurs, and the absence of murmur.

"Combining pathologic murmur detection with the stethoscope, a tool already deeply embedded in the practice of medicine, will allow for more accurate and efficient screening of heart valve disease in the community," said Patrick McCarthy, MD, executive director of the Bluhm Cardiovascular Institute at Northwestern Medicine. "Eko's platform will help uncover early valvular heart disease in millions of people worldwide who may otherwise be missed and make a significant impact on our ability to treat patients with life-saving interventions."

The company cautions that the software should not be used as a sole method of diagnosis. The interpretations offer clinicians information that may support their own evaluation and clinical judgement.