×

Design and evaluation of matched wavelets with maximum coding gain and minimum approximation error criteria for \(R\) peak detection in ECG. (English) Zbl 1157.94313

Summary: Recently, several wavelet-based algorithms have been proposed for feature extraction in non-stationary signals such as ECG. These methods, however, have mainly used general purpose (unmatched) wavelet bases such as Daubechies and Quadratic Spline. In this paper, five new matched wavelet bases, with minimum approximation error and maximum coding gain criteria, are designed and applied to ECG signal analysis. To study the effect of using different wavelet bases for this application, two different wavelet-based \(R\) peak detection algorithms are implemented: (1) a conventional wavelet-based method; and (2) a modified wavelet-based \(R\) peak detection algorithm. Both algorithms are evaluated using the MIT-BIH Arrhythmia database. Experimental results show lower computational complexity (up to 76%) of the proposed R peak detection method compared to the conventional method. They also show considerable decrease in the number of failed detections (up to 55%) for both the conventional and the proposed algorithms when using matched wavelets instead of Quadratic Spline wavelet which, according to the literature, has generated the best detection results among all conventional wavelet bases studied previously for ECG signal analysis.

MSC:

94A11 Application of orthogonal and other special functions
94A08 Image processing (compression, reconstruction, etc.) in information and communication theory
92C55 Biomedical imaging and signal processing
65T60 Numerical methods for wavelets
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] DOI: 10.1109/TBME.2003.821031 · doi:10.1109/TBME.2003.821031
[2] Li C., IEEE Trans. Biomed. Eng. 42 pp 21– · doi:10.1109/10.362922
[3] Senhadji L., IEEE Eng. Med. Biol. pp 167–
[4] DOI: 10.1109/34.625106 · doi:10.1109/34.625106
[5] DOI: 10.1142/S0219691307002129 · Zbl 1135.92313 · doi:10.1142/S0219691307002129
[6] DOI: 10.1142/S0219691307001823 · doi:10.1142/S0219691307001823
[7] DOI: 10.1142/S0219691307002154 · Zbl 1146.65071 · doi:10.1142/S0219691307002154
[8] Masson R., Int. J. Wavelets Multiresolut. Image Process. 6 pp 749–
[9] Chapa J. O., IEEE Trans. Signal Process. 48 pp 3395–
[10] DOI: 10.1109/82.285705 · Zbl 0810.94010 · doi:10.1109/82.285705
[11] DOI: 10.1109/TSP.2004.828923 · Zbl 1369.94330 · doi:10.1109/TSP.2004.828923
[12] DOI: 10.1016/j.patrec.2005.09.002 · doi:10.1016/j.patrec.2005.09.002
[13] DOI: 10.1109/78.622941 · doi:10.1109/78.622941
[14] DOI: 10.1109/78.678504 · Zbl 1011.94517 · doi:10.1109/78.678504
[15] Burrus C. S., Introduction to Wavelets and Wavelet Transforms, A Primer (1998)
[16] DOI: 10.1109/18.382009 · Zbl 0820.62002 · doi:10.1109/18.382009
[17] DOI: 10.1109/97.475823 · doi:10.1109/97.475823
[18] DOI: 10.1016/j.sigpro.2005.08.004 · Zbl 1172.94535 · doi:10.1016/j.sigpro.2005.08.004
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.