Cardiac Arrhythmia Diagnosis with an Intelligent Algorithm using Chaos Features of Electrocardiogram Signal and Compound Classifier

Cardiac Arrhythmias are known as one of the most dangerous cardiac diseases. Applying intelligent algorithms in this area, leads to the reduction of the ECG signal processing time by the physician as well as reducing the probable mistakes caused by fatigue of the specialist. The purpose of this work is to introduce an intelligent algorithm for the separation of three cardiac arrhythmias using the chaos features of ECG signal and combining three types of the most common classifiers in these signal’s processing area. First, the ECG signals related to the three cardiac arrhythmias of atrial fibrillation, ventricular tachycardia, and post-supra ventricular tachycardia along with the normal cardiac signal from the arrhythmia database of MIT-BIH are gathered. Then the chaos features describing non-linear dynamic of the ECG signal are extracted by calculating the Lyapunov exponent values and signal’s fractal dimension. At the end, the compound classifier is used by combining multi-layer perceptron neural network, support vector machine network, and K-Nearest Neighbor. The results obtained are compared with the classifying method based on the features of time-domain and time-frequency domain, as a proof for the efficacy of the chaos features of the ECG signal. Likewise, in order to evaluate the efficacy of the compound classifier, each network is used both as separately and also as combined, and the results are compared. The obtained results show that using the chaos features of ECG signal and the compound classifier can classify cardiac arrhythmias with a 99.1% ± 0.2 accuracy and a 99.6% ± 0.1 sensitivity, and a specificity rate of 99.3% ± 0.1.