Research on the analysis of cardiogram signal with fractals based upon wavelet transform and its application
小波变换的信号分形分析及其在心电信号处理中的应用研究

The nonlinear analysis of heart rate variability (HRV) has been the hotspot in the field of biomedical engineering There are several methods in the research of its dynamic behaviour, such as complexity, Lyapunov exponents, correlation dimension, approximate entropy, detrended fluctuation analysis. Because of the long\|range power\|law correlation characteristic of HRV, this paper proposeds another alternative technique with fractals based upon wavelet transform and gaves its application in the detection of atrial fibrillation. This technique extracts a characteristic value r . The results show that the value rises when atrial fibrillation begins, while the value falls when atrial fibrillation ends. Thus, the begin and the end of atricial fibrillation can be successfully detected according to the change of the characteristic value. In addition, since the value also indicates the complexity of HRV, the results show that the complexity of HRV decreases at the beginning of atricial fibrillation.