Digital stethoscope with artificial intelligence may detect aortic stenosis

Screening for significant aortic stenosis was fast and effective through the assessment of phonocardiograms by a digital stethoscope and machine learning, according to results presented at the American Society of Echocardiography Scientific Sessions.

“A machine-learning algorithm trained on heart sounds can rapidly and accurately detect a murmur in patients with clinically significant aortic stenosis,” Steve Pham, MD, vice president of clinical and research affairs at Eko Devices, told Cardiology Today. “Front-line clinicians may be able to use Eko stethoscopes (Eko CORE) with this algorithm to refer patients for echocardiography to confirm aortic stenosis.”

Brent E. White, MD, of the Bluhm Cardiovascular Institute at Northwestern Memorial Hospital in Chicago, and colleagues analyzed 639 recordings from 161 patients who were undergoing transthoracic echocardiography. The 15-second phonocardiogram recordings were obtained from the digital stethoscope, which is wirelessly paired with a mobile app (Eko Mobile).

“[The device] fits into clinician workflow,” Pham said in an interview. “All clinicians are trained to use a stethoscope from day 1 of medical school, but most are not trained to use ultrasound or other advanced valve disease diagnostics. It is far more affordable than ultrasound. The ubiquity of stethoscopes allow Eko to be an early detector of a notoriously ‘silent’ disease.”

A machine-learning algorithm assessed the recordings for the presence or absence of a murmur that indicated clinically significant aortic stenosis.

Of the patients included in the study, 8.7% had significant aortic stenosis on transthoracic echocardiography.

The receiver-operating characteristic curve of the device had an area of 0.964. This resulted in a specificity of 86.4% (95% CI, 84-88.7) and a sensitivity of 97.2% (95% CI, 84.7-99.5) for the detection of aortic stenosis.

“Further research that is needed includes developing algorithms that can differentiate between valve diseases and that can stage the diseases, which can determine intervention strategy,” Pham told Cardiology Today. “We also need to study the health economics and outcomes impact on the use of valve disease algorithms in front-line medicine.” – by Darlene Dobkowski


White BE, et al. Abstract P2-068. Presented at: American Society of Echocardiography Scientific Sessions; June 21-25, 2019; Portland, Ore.

Disclosures: White reports he has a financial relationship with Eko Devices. Pham is an employee of Eko Devices. Please see the abstract for all other authors’ relevant financial disclosures.



Please enter your comment!
Please enter your name here