Noise Reduction And QRS Wave Recognition Method For ECG Signal

ECG is a comprehensive reflection of cardiac electrical activity on the body surface.ECG signals can objectively reflect the health status of the heart.With the development of science and technology,portable devices for automatic analysis of ECG signals have become a reality from requirements,and physical examinations can be performed at home through portable devices.How to improve the accuracy of portable device diagnostics is currently facing the difficulties of major manufacturers.ECG signals have relatively low amplitude-frequency characteristics and are easily interfered by noise.The most typical types of interference include three types:one is power frequency interference,the other is baseline drift,and the third is myoelectric interference.The presence of noise will cause the characteristic wave of the ECG signal to be difficult to extract and will reduce the accuracy of waveform recognition.Noise filtering is a prerequisite for the identification of various waveforms QRS wave is an important characteristic wave in ECG signal.Accurate and reliable QRS wave recognition is of great significance in the diagnosis of heart disease.This article mainly studies the ECG signal denoising and QRS wave recognition.In the ECG signal denoising algorithm,the band-pass filter designed by the MATLAB's own zero-phase filter is removed for the baseline drift noise,and the wavelet threshold denoising algorithm is used to filter out the power frequency interference and the myoelectric interference.In the QRS wave recognition,a new adaptive threshold method is proposed to identify the QRS wave.In this paper,the main work done:1.The zero-phase filter is studied,and it is designed as a band-pass filter to filter the baseline drift.The effect of the actual ECG signal with baseline drift is verified.2.The influence factors in the wavelet threshold denoising algorithm are studied,and an improved threshold function is proposed for the defects of the soft and hard threshold functions.The modified threshold function has good coherence,smoother waveform and higher signal-to-noise ratio.3.A new adaptive threshold method is proposed to identify the QRS wave.The performance of the proposed algorithm is verified by the ECG signal in the MIT-BIH database and the ECG signal collected in the hospital.This article uses a zero-phase filter to remove baseline drift.Experiments have shown that this method can effectively baseline drift.Using wavelet threshold denoising method to remove power frequency interference and myoelectric interference,the experiment shows that this method not only removes power frequency interference and myoelectric interference,but also maintains the signal without distortion.A new adaptive threshold method is used to identify the QRS wave.The experimental results show that the recognition rate of QRS wave is high and can be used as a method for waveform recognition of ECG signals in portable devices.