The classification of the electrocardiogram (ECG) into different patho-physiological disease categories is a complex pattern recognition task. In this work, it is proposed to integrate Neural Network clustering, principal component analysis (PCA) and particle swarm optimization (PSO) for ECG beat classification. The PCA is used to reduce drastically the dimensionality of the vectors to be classified. A back propagation neural network (BPNN) is employed as classifier. ECG samples attributing to five different cardiac disease are acquired from the MIT-BIH arrhythmias database for experiments. In this paper comparative study of performance of three different structures such as PCA-NN, RBF and PSO-PCA-NN are investigated. The test results suggest that PSO-PCA-NN structure can perform better and faster than other techniques.
