An Empirical Evaluation of Savitzky-Golay (SG) Filter for Denoising ST Segment
Electrocardiogram (ECG) is the graphical illustration of the heart’s electrical activity and it is acquired through linking electrodes to body surface. It is useful in determining cardiac diseases non-invasively. The general concerns in ECG signal processing is the restrain of noise. This paper has been encouraged by the necessary to recognize the ability of Savitzky-Golay (SG) filter for denoising the ST segment of ECG signals. It essentially implicates the extraction of essential cardiac components by refusing the background noise with the aid of filtering technique. The simulation is carried out using MATLAB and the experiments are performed on physionet database. Peak Signal-to-Noise Ratio (PSNR), Signal-to-Noise Ratio (SNR), Mean Square Error (MSE) and Percentage Root mean square Difference (PRD) are used as performance evaluation metrics. This work uses SG filter to denoise the ECG signal and comparison is provided for SG filter with Median, Butterworth and SOS filters. The comparison results states the performance of the SG filter is better than that of the other filters.
KeywordsST segment Denoising ECG signal Filtering Cardiac diseases
- 6.Mjahad, A., Rosado-Muñoz, A., Bataller-Mompeán, M., Francés-Víllora, J.V., Guerrero-Martínez, J.F.: Ventricular fibrillation and tachycardia detection from surface ECG using time-frequency representation images as input dataset for machine learning. Comput. Methods Programs Biomed. 41, 119–127 (2017)CrossRefGoogle Scholar
- 15.Roopa, C.K., Harish, B.S.: A survey on various machine learning approaches for ECG analysis. Int. J. Comput. Appl. 163, 25–33 (2017)Google Scholar