<ul><li>ProtoECGNet is a prototype-based deep learning model for interpretable, multi-label ECG classification.</li><li>It employs a structured, multi-branch architecture integrating different CNNs with global and time-localized prototypes for rhythm classification, morphology-based reasoning, and diffuse abnormalities.</li><li>The model is trained using a prototype loss designed for multi-label learning, combining clustering, separation, diversity, and contrastive loss.</li><li>ProtoECGNet demonstrates competitive performance compared to state-of-the-art models and provides faithful, case-based explanations, making it a practical solution for transparent and trustworthy deep learning models in clinical decision support.</li></ul>

ProtoECGNet: Case-Based Interpretable Deep Learning for Multi-Label ECG Classification with Contrastive Learning

Discover more