Improvement of an Algorithm to Detect Structural Heart Murmurs in Adult Patients Using Electronic Stethoscopes

Improvement of an Algorithm to Detect Structural Heart Murmurs in Adult Patients Using Electronic Stethoscopes

The main objective of this study is to evaluate a machine learning model's ability to detect murmurs indicative of structural heart disease ("structural murmur") by analyzing phonocardiogram waveforms-and simultaneous electrocardiogram waveforms when available-in multiple auscultatory positions per subject. Diagnosis of structural murmur will be confirmed by gold-standard echocardiography and reviewed by an expert panel of cardiologists.

Inclusion Criteria:

• 18+ years old
• Patient or patient's legal healthcare proxy consents to participation
• Documented history of SHD
• Undergoing (or has undergone, within 30 days) a complete echocardiogram
• Willing to have heart recordings done with two different electronic stethoscopes

Exclusion Criteria:

• Patient or proxy is unwilling/unable to give written informed consent
• Unable to complete a complete echocardiogram, or none recent completed within the last 30 days
• No documented history of SHD
• Experiencing a known or suspected acute cardiac event
• Mechanical ventricular support (such as ECMO, LVAD, RVAD, BiVAD, Impella, intra-aortic balloon pumps, TAH, VentrAssist, DuraHeart, HVAD, EVAHEART LVAS, HeartMate, Jarvik 2000)
• Unwilling or unable to follow or complete study procedures