Deep Learning for Detection of Pulmonary Hypertension and Reduced Left Ventricular Ejection Fraction Using a Combined Digital Stethoscope and Three-lead Electrocardiogram

What this Trial Is About

Researchers are evaluating whether heart sounds and three-lead electrocardiograms (ECGs) recorded using the Eko CORE 500 digital stethoscope can help detect pulmonary hypertension (PH) and low left ventricular ejection fraction (EF ≤ 40%). PH is a condition with high blood pressure in the lungs' arteries that can lead to heart failure, and low EF means the heart pumps less effectively. These conditions often go undiagnosed because symptoms may be absent or unclear. This observational study aims to develop and validate artificial intelligence (AI) algorithms for earlier and simpler screening. Adults undergoing clinical echocardiograms at outpatient sites will participate in a single 20-minute visit. During this visit, heart sounds and three-lead ECGs will be recorded at four standard chest locations using the Eko CORE 500 device. If a 12-lead ECG was performed within 30 days of the echocardiogram, those data may also be included. The echocardiogram done within seven days before or after the study visit will confirm the presence of PH or low EF. Participants will provide demographic and clinical data, including medical history and lab results. Researchers will analyze the recordings with AI methods to measure how well the algorithms detect PH and low EF, using sensitivity and specificity as primary outcomes. Up to 3,850 participants will be enrolled to ensure about 3,500 complete the study. No results from the device or AI analyses will be shared with participants or added to medical records.

Deep Learning Detection of Pulmonary Hypertension and Low Ejection Fraction Via Digital Stethoscope and 3-Lead ECG