Cardiosense has announced the publication of results from its SEISMIC-HF I study in the Journal of the American College of Cardiology: Heart Failure (JACC: Heart Failure).
The study demonstrates the ability of Cardiosense’s novel machine learning (ML) algorithm to non-invasively estimate pulmonary capillary wedge pressure (PCWP) in patients with heart failure with reduced ejection fraction (HFrEF).
The prospective, multicentre SEISMIC-HF I study enabled the development and evaluation of Cardiosense’s ML algorithm in 310 patients with HFrEF undergoing right heart catheterization (RHC).
The study population enrolled a diverse set of participants with demographics and clinical characteristics reflective of the broader population with HFrEF to ensure ML models generalise across all patients. Patients enrolled in the study wore the CardioTag device–a multi-modal sensor that captures seismocardiogram (SCG), electrocardiogram (ECG), and photoplethysmogram (PPG) signals–on their sternum during the RHC procedure used to provide gold standard assessment of an individual’s PCWP.
Key findings from the 15-site study show that the device estimated absolute PCWP values with accuracy on par with implantable haemodynamic sensors, suggesting that Cardiosense’s non-invasive, AI-enabled technology has the potential to increase access to hemodynamic-guided care. The device also demonstrated a consistent performance across sex, race, ethnicity, and body mass index.
“Heart failure management remains one of our most significant clinical challenges and most promising opportunities to directly impact patient outcomes,” said Liviu Klein (University of California San Francisco [UCSF], San Francisco, USA), lead author on the study and lead clinical advisor for Cardiosense. “We know that haemodynamic-guided care reduces hospital readmissions and mortality, yet routine cardiac pressure measurements remain inaccessible due to cost and risk. SEISMIC-HF I is a critical advancement toward the quest for non-invasive technology delivering on the promise of fast, safe, and accessible pressure-guided heart failure management.”
Initial results from Cardiosense’s SEISMIC-HF I study were presented as Late-Breaking Science at the American Heart Association’s 2024 scientific sessions.
“This analysis marks a real milestone in our journey to revolutionise heart failure management with our proprietary AI technology that leverages non-invasive sensor data to deliver critical information on cardiac function,” said Omer Inan, PhD, Cardiosense Co-founder and Chief Scientific Officer. “The results are a culmination of more than a decade of research and a testament to the cross-disciplinary collaboration from our team of medical and clinical experts, data scientists, and engineers.”
