Sparse ECG Representation with a Multi-Scale Dictionary Derived from Real-World Signals

The electrocardiogram (ECG) was the first biomedical signal where digital signal processing techniques were extensively applied. By its own nature, the ECG is typically a sparse signal, composed of regular activations (the QRS complexes and other waveforms like the P and T waves) and periods of inactivity (corresponding to isoelectric intervals like the PQ or ST segments), plus noise and interferences. In this work, we show how to construct a realistic multi-scale dictionary using waveforms recorded from realworld patients and how to apply this dictionary to obtain a sparse representation of ECG signals. Simulations on realworld records from Physionet show the good performance of the proposed approach.