Title: Intelligent Breast Cancer Diagnosis Based on Enhanced Pareto Optimal and Multilayer Perceptron Neural Network
Authors: Ashraf Osman Ibrahim; Siti Mariyam Shamsuddin
Addresses: Faculty of Computer and Technology, Alzaiem Alazhari University, Khartoum North 13311, Sudan; Nile College, Khartoum North 11111, Sudan ' UTM Big Data Centre, Universiti Teknologi Malaysia, Skudai, 81310 Johor, Malaysia
Abstract: Among the common cancer diseases is a breast cancer. Diagnosis of this disease depends on the human experience. It is time consuming and having an element of human error in the results. The Pareto optimal evolutionary multi-objective optimisation is used to obtain multiple final results in a single run for simultaneous parameter optimisation of artificial neural networks (ANNs). In this paper, a computer-based method of an automatic classifier for the breast cancer disease diagnosis task is proposed. The proposed method applied a multilayer perceptron (MLP) neural network based on enhanced non-dominated sorting genetic algorithm (NSGA-II) to achieve an accurate classification result for the breast cancer diseases diagnosed. Moreover, it is also used to optimise the network structure and reduce the error rate of the MLP neural network simultaneously. Compared to other methods found in the literature, the proposed method is viable in breast cancer disease diagnosis.
Keywords: breast cancer diagnosis; multilayer perceptron; MLP; Pareto optimal; NSGA-II.
International Journal of Computer Aided Engineering and Technology, 2018 Vol.10 No.5, pp.543 - 556
Received: 03 Jul 2015
Accepted: 18 Mar 2016
Published online: 11 Jun 2018 *