Please use this identifier to cite or link to this item:
|Title:||Hybrid Model for Coronary Artery Disease Classification Based on Neural Networks and Evolutionary Algorithms|
|Authors:||AWAD, MOHAMMED $AAUP$Palestinian|
Multi-Layer Perceptron Neural Network
Genetic Algorithms, Particle Swarm Optimization, Biogeography-Based Optimization
|Publisher:||Journal of Information Science and Engineering / Institute of Information Science,|
|Citation:||*, Hybrid Model for Coronary Artery Disease Classification Based on Neural Networks and Evolutionary Algorithms, Journal of Information Science and Engineering|
|Series/Report no.:||Vol. 38;No. 6|
|Abstract:||The human body is a vital source of data such as images and signals. The signals are collected from different human organs and utilized in diagnosing different diseases. The designing and implementation of intelligent computer programs that try to emulate human intelligence are a sign of the integration of various sciences and areas of knowledge. The development of technologies associated with Artificial Intelligence (AI) techniques that are applied to medicine, represents a novel perspective, which can reduce costs, time, and medical errors. Coronary artery disease (CAD) killed many people in the world, it is considered the most common type of heart disease. This paper uses different evolutionary algorithms to optimize the neural network parameters to enhance the classification process of Coronary Artery Disease (CAD). A hybrid system that combines a Genetic Algorithm (GAs), Biogeography-Based Optimization (BBO) with neural networks (NNs) [GAsBBO-MLPNNs] is proposed to enhance the accuracy of CAD. This paper concentrates on the medical classification system for heart disease on CAD using the Z-Alizadeh Sani dataset as a benchmark. The proposed GAsBBO-MLPNNs outperformed other hybrid models such as; Biogeography-Based Optimization (BBO) and particle swarm optimization (PSO) methods combined with NNs, and previous works, where the method performance parameters result represented as 94.5%, 96.4%, and 94.8% accuracy, Sensitivity, and Specificity respectively|
|Appears in Collections:||Faculty & Staff Scientific Research publications|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.