A Hybrid 2D ECG Compression Algorithm using DCT and Embedded Zero Tree Wavelet

Electrocardiogram (ECG) signal plays a key role in the identification of heart diseases. ECG data is recorded to identify different heart abnormalities and a large amount of data is produced at the time of recording. Therefore, compression of ECG signal is highly recommended for the effective utilization of storage and transmission resources. In this regard, a compression algorithm is proposed for 2D ECG signals to utilize both interbeat and intra-beat redundancies. It is proposed using the combined application of discrete cosine transform and embedded zero tree wavelet. Furthermore, different wavelets such as db10, haar, coif5, sym2, and bior3.5 have been used to select the most suitable wavelet based on the comparative analysis. The proposed method is able to improve the transform domain's sparsity, which increases the compression efficiency at the cost of a small degradation in reconstruction quality. The experimental work has been tested on the MIT-BIH arrhythmia database. The performance assessment of the proposed algorithm has been done in terms of percent root-mean-square difference, compression ratio, and signal-to-noise ratio.