Optimization of neural network training using genetic algorithm for texture segmentation of MRI - images

Authors

  • В. А. Панчук National Technical University of Ukraine "KPI"
  • Д. Ю. Лебедев National Technical University of Ukraine "KPI"

DOI:

https://doi.org/10.18372/2310-5461.19.5585

Keywords:

Neural network, perceptron, back propagation algorithm, genetic algorithm

Abstract

An analysis of existing methods for image processing. These solutions allow you to create an automated system for diagnosing diseases based on data obtained from images of magnetic resonance imaging. Satisfy the requirements of the system diagnostic neural network based on three - layer perceptron. In the process simulation software package MatLab v R2007b, were investigated speed of network training on the number of neurons in the hidden layer for the conventional method of back propagation and combined with genetic algorithm. The study revealed the advantage of the combined method for image classification task. This method halves the number of training effects and requires fewer neurons in the hidden layer, which facilitates network architecture and reduces the computational cost

References

Хайкин С. Нейронные сети: полный курс /

С. Хайкин. — 2-e изд. : пер. с анrл. — М. : Изда-тельский дом «Вильямс», 2006. — 1104 с.

Рутковская Д. Нейронные сети, генетиче-ские алгоритмы и нечеткие системы: пер. с польск. И. Д. Рудинского / Д. Рутковская, М. Пилиньский, Л. Рутковский. — М. : Горячая линия-Телеком, 2006. — 452 c.

Zhen-Guo Che1. Feed-forward neural networks training: a comparison between genetic algorithm and back-propagation learning algorithm / Che1 Zhen-

Guo, Chiang Tzu-An and Che Zhen-Hua // International Journal of Innovative Computing, Information and ControlVolume 7, Number 10, October 2011.

Randall S. Sexton. Toward global optimization of neural networks: a comparison of the genetic algo-rithm and back propagation /. Sexton Randall S, Dorsey Robert E., and John D. Johnson // Songnian et al. BMC Neuroscience 2010

Ahmed Kharat. A Hybrid Approach for Auto-matic Classification of Brain MRI Using Genetic Al-gorithm and Support Vector Machine/ Kharat Ahmed, GasmiKarim // Leonardo Journal of Sciences July-December 2010. — С. 71–82.

Randall S. Sexton and Robert E. Dorsey1. Reli-able Classification Using Neural Networks: A Genetic Algorithm and Backpropagation Comparison.

Dorsey RE. The genetic adaptive neural net-work training (GANNT) for generic feed-forward artificial neural systems / RE Dorsey, JD Johnson, WJ Mayer // School of BusinessAdministration, University of Mississippi, University.

Sexton RS. Optimization ofneural networks: a comparative analysis of the genetic algorithm and simulated annealing / RS Sexton, RE Dorsey, JD Johnson // European Journal of Operational Research 1999; 114:589±601.

Ahmed Kharat. Automated Classification of Magnetic Resonance Brain Images Using Wavelet Genetic Algorithm and Support Vector Machine / Kharat Ahmed, Gasmi Karim, Abid Mohamed // Leo-nardo Journal of Sciences July-December 2011. — Р. 369–374.

Usama A. Badawi1. A Hybrid Memetic Algorithm (Genetic Algorithm andGreat Deluge Local Search) With Back-PropagationClassifier for Fish Recognition / Badawi1 Usama A., Khalil Mutasem, Alsmadi Sari // IJCSI International Journal of Computer Science Issues, Vol. 10, Issue 2, No 1, March, 2013.

Hiroaki Kitano. Empirical Studies on the Speed of Convergence of Neural Network Training using Genetic Algorithms / Kitano Hiroaki // From: AAAI-90 Proceedings. Copyright ©1999.

JIRI STASTNY*, VLADISLAV SKORPIL*. Genetic Algorithm and Neural Network // Proceedings of the 7th WSEAS International Conference on Applied Informatics and Communications, Athens, Greece, August 24–26, 2007.

Дьяконов В .П. MATLAB 6.5 SP1/7/7 SP1/7 SP2 + Simulink5/6. Инструменты искусственного интеллекта и биоинформатики/ В. П. Дьяконов,

В. В. Круглов. — М. : СОЛОН-ПРЕСС, 2006. —

с.

Downloads

Published

2013-11-28

How to Cite

Панчук, В. А., & Лебедев, Д. Ю. (2013). Optimization of neural network training using genetic algorithm for texture segmentation of MRI - images. Science-Based Technologies, 19(3), 285–289. https://doi.org/10.18372/2310-5461.19.5585

Issue

Vol. 19 No. 3 (2013)

Section

Information and Communication Systems and Networks

License

The scientific journal adheres to the principles of Open Access and provides free, immediate, and permanent access to all published materials without financial, technical, or legal barriers for readers.

All articles are published in Open Access under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

Copyright

Authors who publish their works in the journal:

  • retain the copyright to their publications;

  • grant the journal the right of first publication of the article;

  • agree to the distribution of their materials under the CC BY 4.0 license;

  • have the right to reuse, archive, and distribute their works (including in institutional and subject repositories), provided that proper reference is made to the original publication in the journal.