AI-Powered Stethoscope Could Diagnose Heart Disease in Minutes

1. A smart stethoscope developed by Eko Health leverages artificial intelligence to detect structural heart disease, such as valvular abnormalities, with high sensitivity.

2. The algorithm received FDA clearance after demonstrating diagnostic performance comparable to echocardiography in multiple validation studies.

Advances in cardiovascular diagnostics have increasingly integrated machine learning to enable earlier detection and triage of structural heart disease. A novel device by Eko Health combines a digital stethoscope with an FDA-cleared deep learning algorithm trained to identify murmurs consistent with valvular heart disease, including aortic stenosis. In a multicenter validation study involving over 3,400 adult patients, the AI algorithm demonstrated a sensitivity of 93.2% (95% CI, 89.7%–96.0%) and specificity of 86.0% (95% CI, 82.1%–89.4%) in identifying moderate-to-severe aortic stenosis compared to echocardiographic gold standard, with p < .001 for overall diagnostic concordance. The study population included diverse primary care and cardiology settings across the U.S., enhancing generalizability to real-world use.

The prospective trial, known as Eko CORE-HF, enrolled participants with suspected cardiac abnormalities based on physical exam or risk profile. The AI model was embedded within the digital stethoscope and processed cardiac auscultation in under one minute. Primary outcomes included concordance with echocardiography results, inter-observer reliability among clinicians using the tool, and usability scores based on a standardized implementation framework. Notably, in patients over 65 years old, the device maintained robust sensitivity across comorbid conditions including hypertension and diabetes, with no significant decline in performance (interaction P = 0.47). Clinician feedback also suggested reduced diagnostic uncertainty and expedited referral decisions in resource-limited environments.

The tool has since been deployed in select U.S. primary care clinics and rural health systems, where access to echocardiography may be delayed or unavailable. By identifying high-risk patients at the point of care, this AI-enabled device offers a rapid and cost-effective means to reduce missed or late diagnoses of valvular heart disease. Its integration into routine clinical workflows may be particularly impactful in underserved populations, offering scalable precision screening without requiring imaging infrastructure.

Relevant reading: Deep Learning Algorithm for Automated Cardiac Murmur Detection via a Digital Stethoscope Platform

©2025 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.