Raised Cosine Dictionary for Modeling and Compressed Sensing of Multi-lead ECG Signals

nowadays, modern technology has a very crucial role in healthcare and medical systems. An important issue in biomedical and signal processing area, is compressing biomedical signals. One of the most important signals is Electrocardiogram (ECG) which plays an important role in diagnosing heart’s anomalies. In this paper, the compressed sensing (CS) theory is exploited in order to compress multi-lead ECG signals. To achieve the sparse representation of the ECG signals, a basis matrix using raised Cosine functions is constructed. Raised Cosine kernels nicely fit ECG signals and their low excess Kurtosis makes these functions a suitable choice to model ECGs. Finally, CS theory is used to achieve the compressed version of the signals. Exploiting some optimization techniques the reconstructed signals are obtained from their compressed version. Finally the proposed method is evaluated by three known performance metrics.

کلیدواژه ها:

compressed sensing ، raised Cosine kernel ، ECG signal ، sparse representation