Foundation Models for Cardiovascular Disease Detection via BioSignals from Digital Stethoscopes

Auscultation of the heart and the electrocardiogram (ECG) are two central components of the cardiac exam. Recent innovations of the stethoscope have enabled the simultaneous acquisition of a high-quality digital acoustic signal and single or three-lead ECG during a routine cardiac exam. We present foundation models trained on phonocardiogram (PCG) and ECG data collected from digital stethoscopes during regular clinical practice. We show that these foundation models that are pre-trained on large unlabeled datasets in a self-supervised manner can be fine-tuned for a variety of cardiovascular disease detection tasks including detection of structural murmurs, atrial fibrillation and low ejection fraction. This is the first study that builds foundation models for PCG and ECG data and specifically for synchronously captured PCG and ECG data. Our approach is based on the recently developed masked autoencoder framework and we extend the framework to handle multiple signals with different modalities and that are synchronously captured from patients. This paradigm makes it possible to use large capacity models leading to superior performance even though the size of high-quality curated datasets with medical label annotations may be limited.