Presenting a New Strategy to Extract Data Clustering Heartbeat Samples by Using Discrete Wavelet Transform

Abstract

This paper presents the improvement of detection system that normal and arrhythmia electrocardiogram classification. This classification is done to aid the ANFIS (Adaptive Neuro Fuzzy Inference System). The data used in this paper obtained from MIT-BIH normal sinus ECG database signal and MIT-BIH arrhythmia database signal. The main goal of our approach is to create an interpretable classifier that provides an acceptable accuracy. In this model, the feature extraction using DWT (Discrete Wavelet Transform) is obtained. The last stage of this extraction is introduced as the input of ANFIS model. In this paper, the ANFIS model has been trained with Quantum Behaved Particle Swarm Optimization (QPSO). In this study, for training of proposed model, four sample data have been used which result in acceleration of training data. On the test set, we achieved an outstanding sensitivity and accuracy 100%. Experimental results show that the proposed approach is very fast and accurate in improving classification. Using the proposed methodology and telemedicine technology can manage patient of heart disease.