Investigation on the Compression of Electrocardiogram Signals Using Dual Tree Complex Wavelet Transform

Abstract

ABSTRACTElectrocardiogram (ECG) records the electrical potentials of the heart. ECG reveals a lot of useful information on the normal and abnormal conditions of heart. It is very difficult to analyse ECG signals as they are non-stationary in nature. There is a need to compress the ECG signal in an efficient way to reduce the amount of data that is transmitted, stored, and analysed without losing the significant clinical information. In this paper, compression using dual tree complex wavelet transform (DT-CWT) has been proposed, that results in many wavelet coefficients getting close to zero. To improve the compression ratio, Set Partitioning in Hierarchical Tree (SPIHT) coding is used along with DT-CWT to compress data. The proposed method gives better compression ratios and reduced reconstruction errors compared to stationary wavelet transform (SWT). Experimental results of DT-CWT based SPIHT are shown on many MIT-BIH records which show improved performance by 35.19% over existing methods namely, SWT and 19.02% over empirical wavelet transform.