Enhancement and compression of the electrocardiogram signal using the discrete wavelet transform

The aim of this paper is to present an efficient process of the denoising and the compression of the electrocardiogram signal (ECG). This process is based on an enhanced algorithm based on the discrete wavelet transform. Two different algorithms are proposed in this process. The first technique is a hybrid algorithm of the discrete wavelet transform and the adaptive dual threshold filter for the ECG enhancement. The second technique is based on four levels for the ECG signal compression. The First is the ECG signal decomposition. The second is the hard thresholding step. The third is the run length encoding step, and the last step is the minimisation of the number of bits in each sample. The proposed methods are tested on some of the signals of the MIT-BIH arrhythmia database. These methods show high performances, both in the qualitative and the quantitative results, comparing to other methods recently published.