ECG Feature Analysis Based on Poincaré Plot and Symbolic Dynamics

Heart rate variability (HRV) can be used to predict, prevent heart related diseases and do prognosis evaluation. A method was proposed based on Poincaré plot and symbolic dynamics to analyze ECG feature. Firstly, HRV sequence was extracted from the ECG signal, and presented in Poincaré plot. Then, the splashes in different areas of the plot would be numbered and coded. The entropy of the ECG signal, calculated by the probability of each code, was applied to recognize and classify ECG signal as feature. The experiment results show that, the accuracy rates of classification in normal sinus rhythm and atrial fibrillation, normal sinus rhythm and premature beat are 86.67%, 90% respectively, the method can distinguish normal sinus rhythm from arrhythmia effectively.