Non-invasive Pulmonary Artery Prediction

Study to Determine if Novel Wearable Monitoring System and Machine-Learning Algorithm Can Model Continuous Pulmonary Artery Pressure Recordings in Human Subjects

Cardiac remote monitoring devices have expanded our ability to track physiological changes used in the diagnosis and management of patients with cardiac disease. Implantable remote monitoring technologies have been shown to predict heart failure events, and guide therapy to reduce heart failure hospitalizations. The CardioMEMs System, the most studied and established remote monitoring system, relies on a pulmonary artery implant for continuous PAP measurement. However, there are no commercially available wearable systems that can reproduce continuous PAP tracings.

This study aims to determine if a machine-learning algorithm with data from a wearable cardiac remote-monitoring system incorporating EKG, heart sounds, and thoracic impedance can reproduce a continuous PAP tracing obtained during right heart catheterization.

Inclusion Criteria:

1. Subjects age 18+ years
2. Undergoing a right heart cardiac catheterization or in the cardiac care unit with active monitoring using an arterial line or Swan-Ganz catheter.

Exclusion Criteria:

1. Vulnerable population
2. Unable to consent for any reason
3. Unstable patient
4. Known skin reaction to latex or adhesives