Intelligent Arrhythmia Detection Using Genetic Algorithm and Emphatic SVM (ESVM)

In this paper, a new method of arrhythmia classification is proposed. At first we extract twenty two features from electrocardiogram signal. We propose a novel classification system based on genetic algorithm to improve the generalization performance of the SVM classifier. For this purpose, we have optimized the SVM classifier design by searching for the best value of the parameters that tune its discriminate function, and looking for the best subset of features that feed the classifier. We select appropriate features with our proposed Genetic-SVM approach. We also propose Emphatic SVM (ESVM), a new SVM classifier, with fuzzy constraints. It emphasizes on constraints of SVM formulation to give more ability to our classifier. We finally, classify the ECG signal with the ESVM. Experimental results show that our proposed approach is very truthfully for diagnosing cardiac arrhythmias. Our goal is classification of four types of arrhythmias which with this method we obtain 95% correct classification.