Computational Electrocardiology Group

Example VCG for a single ECG with multiple heartbeats. Self-Supervised Contrastive Learning Enables Robust ECG-Based Cardiac Classification We developed a contrastive self-supervised learning framework approach from large collections of unlabeled ECG recordings. We utilized three datasets in this study: a left ventricular ejection fraction-labeled ECG dataset, a large-scale unlabeled…

Body Surface Recordings Can be Used to Identify the Position of the Heart In this study we demonstrated a novel optimization strategy to identify the position of the heart within the chest using recordings from the body surface alone. This method leverages the relative differences in body surface ECG…