Efficient electrocardiogram data compression algorithm based on wavelet transform

Abstract

Physicians as the most important tool in the diagnosis of acute and chronic heart conditions leverage the Electrocardiogram (ECG) data. The huge volume of data that needs to be gathered; processed as well as transmitted burdens the communication channel by obstructing the implementation of the ECG applications in telemedicine and remote monitoring scenarios. This has triggered the need for computationally efficient and advanced compression methods permitting the reconstruction of the destination's original signal. This paper focuses on proposing a highly efficient compression method relying on both run length coding (RLC) and fast lifting wavelet transform (FLWT) for both 1 and 2 dimensional ECG data. The basic wavelets that had been used are Daubechies (Db4); Cohen-Daubechies-Feauveau (CDF 2.2) and Haar. The algorithm that has been proposed is deployed as well as tested on different ECH data signals that stem from an open MIT-BIH arrhythmia database. It has been seen that Db4 offers greater performance via lower error and greater compression ratio when opposed to the other wavelets stated here.