A Hybridized Artificial Neural Network and Optimization Algorithms for the Diagnosis Of Cardiac Arrhythmias
Abstract
In the recent years, the use of Intelligent Systems in Engineering Sciences, especially in the diagnosis of various diseases, is growing increasingly. In this paper, two intelligent methods for detecting cardiac arrhythmias based-on combination structure of artificial neural networks and the Optimization Algorithms are used. The optimization algorithms used in this study are Particle Swarm Optimization Algorithm and Genetic Algorithm, that have been used for optimization of weight coefficients and bias to minimize error. The results of implementing algorithms mentioned in reference data UCI from this method show a remarkable relative advantage of neural network based on PSO algorithm, with the Mean Squared Error and the Correct Classification Rate of 0.01204 and 85.36%, respectively.
Keywords
ECG arrhythmia; MLP neural network; Genetic Algorithm; Particle Swarm Optimization