The Effective Detection Of T-wave Alternans

With the development of economy and the progress of the society, the number of sudden cardiac death is increasing year by year, and the malignant ventricular arrhythmia is a major cause of sudden cardiac death. A huge number of studies show that TWA exist in malignant ventricular arrhythmias of ECG. T-wave alternans (TWA) is a manifestation of ECG variation, such as the shape, polarity and amplitude of T-wave changes alternately beat to beat. Microvolt T-wave Alternans (MTWA) have differences in microvolt level and important clinical value in avoiding sudden death.This paper focus on the effectiveness of TWA algorithm, the main work are listed as following:Synthetic analog TWA ECG signals. Firstly a normal and single beat of the heart is copied, so as to form ECG signals with no noise (the EMG interference, baseline drift frequency and harmonic interference) and no T-wave alternans, then we add real physiological noise and T-wave alternans factors into the ECG signals, finally we get ECG signals that have high similar degree with actual clinical data. By the synthetic process of ECG, we can get the basic parameters of ECG waveforms ahead.The pretreatment of simulate ECG signals. Firstly, we apply simple integer coefficient 50Hz notch filter, digital filter and bior2.2 biorthogonal wavelet filter to analog ECG signals, we can remove the noise of ECG efficiently, so as to get the ECG signal feature obviously, label each peak position on its effectiveness and lay the groundwork for the next step of the detection of TWA algorithm. Using of accurate, simple and convenient calculation of two order differential Marr-wavelet for multilevel wavelet decomposition, and then determine the T-peak position.The detection of the TWA algorithm is the point. Firstly, we seperate the sampling points of the adjacent T-wave into two parts, including the group of even matrix vector and the group of odd matrix vector. The T-wave interval (Ts-Te) and the mean of the RR interval should be collected and record, then we should calculate the time ratio of Ts-Te/RR in T-wave band, analyze the difference and correlation between the adjacent T wave for obtaining the correlation with T wave alternans; Secondly, based on the spectrum analysis,we proposed a methed of using the value of WRr to calculate ECG signal in time domain, so as to get the correlation between the number of WRr and TWA phenomenon in ECG; Finally, using synthetic TWA signals to verify the effectiveness of the Poincare plot method, the relationship between the external forms and T-wave alternans could be fingered out. We analyzed and summarized the experimental results.This paper focuses on using synthetic TWA signals. The basic parameters of the ECG waveform can be known in advance. We can analyze the different shape of T-wave, T-wave alternans amplitude and signal-to-noise ratio of the ECG signals by improving the T-wave window method. There are two improved algorithms:the ratio of Ts-Te/RR and the ratio of T-wave spectrum alternating. These purposes are to improve the detection of TWA phenomenon. At present, domestic and foreign research on the detection of TWA are not common, The detection of artificial TWA simulation need to be further studied.