Research On Motion Artifact Removal And Compression Method For ECG Signals

As a key bioelectrical signal that can reflect human physiological health,electrocardiogram(ECG)signal is widely used in clinical research of cardiovascular disease.However,the ECG signal will be severely disturbed by motion artifact during the acquisition process,which affects the judgment,diagnosis and treatment of the patient’s cardiac health status.At the same time,long-term real-time monitoring will also bring challenges to the wireless transmission and storage of data.To solve these problems,this paper studies the motion artifact removal method and compression method for ECG signals.The main research contents are as follows:(1)An ECG motion artifact removal method based on empirical wavelet transform(EWT)and artificial neural network(ANN)is proposed.Combined with spectral trend-based boundary prediction and EWT,the ECG signal with motion artifact is decomposed into noise dominant component and ECG dominant component.According to the spectral characteristics of the noise in the two components,the ANN is used to process the noise dominant component to remove the noise and restore the effective information of the ECG signal in the corresponding low frequency band,and then use the adaptive dual threshold method to remove the high frequency noise in the ECG dominant component.The experimental results show that the proposed method can effectively remove motion artifact in ECG signals,reduce waveform distortion,and retain the characteristic information in the original ECG signals.(2)An ECG signal compression method based on discrete wavelet transform(DWT)and variable length run length encoding(VL-RLE)is proposed.The time-domain ECG signal is converted to the energyconcentrated wavelet domain by DWT to obtain wavelet coefficients,and perform dead-zone quantization on the wavelet coefficients for encoding.According to the data distribution characteristics of data sequences and count sequences obtained by run length encoding,Huffman coding and length-value coding can effectively reduce the average bit length of the encoded data and improve the compression efficiency.Experimental results show that the proposed method provides a compression ratio of20.48 while ensuring lower reconstruction error,showing better compression performance compared with other ECG signal compression methods.