Efficient Use of Bi-orthogonal Wavelet Transform for Cardiac Signals

Abstract

The ECG finds its importance in the detection of cardiac abnormalities. ECG signal processing in an embedded platform is a challenge which has to deal with several issues. Noise reduction in ECG signal is an important task of biomedical science. ECG signals are very low frequency signals of about 0.5Hz-100Hz. There are various artifacts which get added in these signals and change the original signal , therefore there is a need of removal of these artifacts from the original signal. The noises that commonly disturb the basic electrocardiogram are power line interference, electrode contact noise, motion artifacts, electromyography (EMG) noise, and instrumentation noise. These noises can be classified according to their frequency content. In this paper, these we have used wavelet transform based approach for removing these noise. In this paper, the discrete wavelet transform (DWT) at level 8 was applied to the ECG signals and decomposition of the ECG signals was performed. After removal of noise component using thresholding technique, decomposed signal is again constructed using Inverse discrete wavelet transform (IDWT). Here for de-noising the ECG signal, bi-orthogonal wavelet transform is used and the most efficient idea for noise removal process is concluded with this wavelet transform. The simulation has been done in MATLAB environment. The experiments are carried out on MIT-BIH database. Performance analysis was performed by evaluating Mean Square Error (MSE), Signal-to-noise ratio (SNR), Peak Signalto-noise ratio (PSNR) and visual inspection over the de-noised signal from each algorithm.

Automated Emotion Recognition Employing Wearable ECG Sensor and Deep-Learning

“…Discrete wavelet transform (DWT) provides time localization and is, therefore, well-suited for processing nonstationary signals. Several wavelets are shown to provide optimal results in isolating the QRS complex in ECG signals, including coif5 [11], db4 [12], and several biorthogonal wavelets including bior3.9, bior4.4 and bior6.8 [13]. The bior6.8 wavelet is used in this study, with detailed comparison of results using different wavelets left for future work.…”

Prediction of daily rainfall by a hybrid wavelet-season-neuro technique

“…As a result, large amount of computation time and resources are required for the calculation of CWT. As opposed to the CWT, DWT deals with functions that are defined over a range of integers transform and it is described as an implementation of the wavelet transform using a discrete set of the wavelet scales and translations obeying some defined rules (Sharma et al, 2014). DWT is achieved by modifying the wavelet representation to the form given in Eq.…”