Automatic Diagnosis Algorithm Research Of Premature Ventricular Contraction In Electrocardiogram Signal

At present, there is an upward trend in the incidence of cardiovascular disease at present. Electrocardiogram(ECG) signal is an important basis for the diagnosis of cardiovascular disease. Therefore, the automatic processing and analysis of ECG signal become more and more important. Premature Ventricular Contraction(PVC) is a common type of abnormal heartbeat. Frequent PVC is often associated with complications like syncope, angina, and heart failure. Without early diagnosis and proper treatment, PVC may result in more serious harms. Diagnosis of PVC is of great importance in goal-directed treatment and preoperative prognosis.This paper proposes a method of automatic diagnosis of PVC. We use wavelet transform to remove the noise of ECG signals. First,R peak of ECG signal can be detected based on the analysis of energy and frequency range. Then, a single ECG heart beat can be cut out according to the position of R wave. Premature beat can be identified by the non-uniformity of RR intervals, which helps to significantly reduce the range and difficulty of PVC diagnosis. Next, ECG heart beat is reconstructed in phase-space. Parameters of phase space reconstruction are determined according to the experiment. All Lyapunov exponents are calculated in ECG beats. Different types of ECG beats have different Lyapunov exponent curve. Chaotic characteristics can be extracted based on the analysis of Lyapunov exponent. Finally, the characteristic parameters of each ECG beat are used as input to the neural network. Then PVCs can be distinguished from other arrhythmia.Data from MIT-BIH arrhythmia database is used in our method to verify the effectiveness and stability of the algorithm. The experiment result shows that the automatic analysis algorithm based on chaotic characteristics of ECG has a well performance on separating the PVCs to other beats. Automatic diagnosis of PVC is realized.