Denoising and Compression of Biosignals (ecg, Eeg & Emg) Using Wavelets

Biomedical signals from many sources including heart, brain and endocrine system pose a challenge to researchers who may have to separate weak signals arriving from multiple sources contaminated with artifacts and noise. The analysis of these signals is important both for research and for medical diagnosis and treatment. The main difficulty in dealing with biomedical signals is the extreme variability of the signals and the necessity to operate on case to case basis. Another important aspect of biomedical signals is that the information of interest is often a combination of features that are well localized temporally or spatially and other that are more diffused. This requires the use of analysis methods sufficiently versatile to handle events that can be at opposite extremes in terms of their timefrequency localization. So, a robust method is to be designed which work in most circumstances rather than under very specific assumptions. Biomedical signals due to its enormous virtues are widely applied in several medical applications; Electrocardiogram, Electroencephalogram and Electromyogram are to name a few. However, these signals experience the addition of noise and result in an inefficient performance. A detailed analysis of Discrete Wavelet Transform (DWT) denoising using various wavelet families on biomedical signals (ECG, EMG and EEG) is presented in the thesis. The main intention of the work is to explore the wavelet function that is optimal in identifying and denoising the various biomedical signals. Nevertheless, Wavelet transforms offer better results for denoising the bio-medical signals, identification of the optimal Wavelet type is crucial.The proposed Wavelet Transform based frequency thresholding is used for noise removal of the noise corrupted decomposed biomedical signals. Then the signal is reconstructed using inverse wavelet reconstruction method. To reduce the storage size, hybrid wavelet Shannon-Fano coding is used for compression of denoised signal. Efficiency of the method used vis a vis existing techniques for the denoising of ECG, EEG and EMG signals has been evaluated and compared in terms of Signal to Noise Ratio (SNR), Percent Root Mean Square Difference (PRD), Mean Square Error (MSE) and Compression Ratio (CR)