Study of Quality Evaluation Indexes in Filtered ECG Signal

Abstract:

Electrocardiogram (ECG) is a kind of weak signal. It was disturbed by surrounding factors, even by patient him/herself. It was happened mostly in portable device. Filtering is an usual step in ECG signal processing. Therefore, the quality evaluation of ECG signal became necessary. In the paper, some indexes were proposed to evaluate the quality of filtered ECG signal. The definition and recommended values or limits of the indexes were discussed. The indexes covered from the aspects of signal procession and clinical diagnosis meanings. They were Signal-to Noise Ratio (SNR), Autocorrelation coefficient (AC), Transformation Ratio (StTR) and Voltage Amplitude Change (StTV) of ECG ST Segment. Median, Wavelet, and Morphology filters were selected in the experiments. From the experiment results, Wavelet performs best in controlling attenuation, but it distorted ST segment the most, both in shape and in its voltage amplitude. The shape change ratio may reach 25%, compare to 17% of median and 14% of morphology, and those filters were acceptable clinical evaluation. It was proved that the indexes can become the potential standard in quality evaluation in ECG signal filtering process.

• Periodical: Applied Mechanics and Materials (Volumes 284-287)
• Pages: 1671-1675
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.284-287.1671
• Online since: January 2013
• Authors: Gang Zheng, Ming Li Sun, Yuan Gu
• Keywords: ECG Signal, Filtering Process, Index, Signal Quality Evaluation

References

[1] J. Mateo, C. Sanchez, C. Vaya, R. Cervigon, J.J. Rieta, Computers in Cardiology, (2007) Sept. 30 -Oct. 3; Durham, NC

[2] G. Zheng, Y. Gu, M. Dai, Journal of Computational Information Systems, Vol. 7:5 (2011), p.1770

[3] Joyce Van de Vegte, Fundamentals of Digital Signal Processing, Pearson Education, Publisher: Prentice Hall, Canada (2003)

[4] Y. Gu, G. Zheng, M. Dai. Computing in Cardiology, (2011) Sept.18-21;Hangzhou,China

[5] L. Mao, JX. Sun, GM. Zhang, H. Ji, Signal Processing, Vol. 24:4 (2008)