Electrocardiogram Compression using Optimized TQWT and Dead-Zone Quantizer

In the biomedical field, electrocardiogram (ECG) recording produces a large amount of data, which are stored in a digitized format for monitoring and diagnosis purposes. In this regard, it is essential to reduce data size due to memory constraints in ambulatory and tel-e-medicine systems. This paper proposes an algorithm using optimized tunable-Q wavelet transform (TQWT) to reduce the memory requirement. It has the flexibility to tune its parameters to obtain the desired compression. For optimizing the parameters of TQWT, nature-inspired algorithm ant colony optimization (ACO) is used. The compression is achieved by using a dead-zone quantizer (DZQ) and run-length encoding (RLE). Results illustrate that significant compression has been achieved at the cost of acceptable distortion in the signal quality. The performance of the proposed technique is evaluated using percentage-root-mean square difference (PRD), compression ratio (CR), and quality score (QS). The average value obtained of CR, PRD, and QS are given as 22.42, 4.52%, and 6.05, respectively.