Compression of ECG Signal Based on Improving the Signal Sparsity Adopting QRS-Complex Estimation and Spatial Domain Transforms

Document Type : Original Article

Authors

1 Electrical and Electronics Eng. Dept., Faculty of Eng., Assiut University, Egypt.

2 Computer and Systems Dept., Faculty of Engineering, Minia University, Egypt.

Abstract

In this paper, an ECG signal compression technique based on improving the signal sparsity adopting QRS-complex estimation and spatial domain transforms is proposed. Compressive Sensing (CS) framework is utilized for this purpose; where the difference between the rate of change of a signal and the rate of change of its information contents is utilized. The proposed method starts with estimating the QRS-complex through the detection of the maximum amplitudes and the start and end points of the signal components. The error signal calculated as the difference between the estimated QRS-complex samples and original ECG samples is transformed using either DWT, DCT or FFT. QRS-complex estimation results in sparser timedomain error signal and the sparsity is increased by adopting the spatial-domain transforms. The proposed technique is assessed by calculating PRD and CR. Numerical results indicate that DWT gives a higher CR and lower PRD. The effect of increasing the signal sparsity, signal length, wavelet filters and wavelet decomposition levels have been studied. Comparison with recently published results [1] adopted only QRS-estimation and DWT indicate that the utilization of CS with both QRS-estimation and DWT yields to improved acceptable retrieved signal quality and higher CR. Moreover, comparison with other four CS-based algorithms [2]-[5] and other traditional ECG compression algorithm [6] indicates the superior performance of the proposed algorithm.
.

xt-stroke-width: 0px; "> [1] M. Abo-Zahhad, S. M. Ahamed and A. Zakaria, "ECG Signal Compression
Technique Based on Discrete Wavelet Transform and QRS-Complex
Estimation," Signal Processing An International Journal (SPIJ), vol. 4, no.
2, pp. 138-160, 2010.
[2] A.M.R. Dixon, E.G. Allstot, D. Gangopadhyay, and D.J. Allstot,
“Compressed sensing system considerations for ECG and EMG wireless
biosensors,” IEEE TBCAS, vol. 6, no. 2, pp. 156 –166, April 2012.
[3] L.F. Polania, R.E. Carrillo, M. Blanco-Velasco, and K.E. Barner,
“Compressive sensing exploiting wavelet domain dependencies for ECG
compression,” in SPIE Defense, Security, and Sensing, pp. 83650E–83650E,
April 2012.
[4] L.F. Polania, R.E. Carrillo, M. Blanco-Velasco, and K.E. Barner, “On
exploiting interbeat correlation in compressive sensing-based ECG
compression,” in SPIE Defense, Security, and Sensing, pp. 83650D–
83650D, April 2012.
[5] L.F. Polania, R.E. Carrillo, M. Blanco-Velasco, and K.E. Barner,
“Compressed sensing based method for ECG compression,” in Proceedings e-width: 0px; "> IEEE Int. Conf. on Acoustics, Speech, and Signal Processing, Prague, Czech
Republic, ICASSP, , pp. 761-764, May 2011.
[6] Z. Lu, D. Y. Kim, and W. A. Pearlman, “Wavelet compression of ECG
signals by the set partitioning in hierarchical trees algorithm,” IEEE T-BME,
vol. 47, no. 7, pp. 849 856, July 2000.
[7] [7] D. L. Donoho, “Compressed sensing,” IEEE Trans. Inform. Theory, vol.
52, pp. 1289-1306, 2006.
[8] [8] E. Candes, J. Romberg, and T. Tao, “Robust uncertainty principles:
exact signal reconstruction from highly incomplete frequency information,”
IEEE Trans. Inform. Theory, vol. 52, no. 2, pp. 489509, Feb. 2006.
[9] [9] M. M. Abo-Zahhad, Aziza I. Hussein, A. M. Mohamed, "Compression
of ECG Signal Based on Compressive Sensing and the Extraction of
Significant Features," Int. J. Communications, Network and System
Sciences, vol. 8, pp. 97-117, April 2015.
[10] [10] A. B. Roy, D. Dey and B. Mohanty and D. Banerjee, "Comparison of
FFT, DCT, DWT, WHT Compression Techniques on Electrocardiogram
and Photoplethysmography Signals" Special Issue of International Journal
of Computer Applications (Inte. Conf. on Computing, Communication and
Sensor Network, CCSN'2012), pp. 6-11, 2012.