Classification of magnetic resonance images by using genetic algorithms

Abstract

A neural network trained by genetic algorithms (GANN) is presented. Each neuron of the network forms a closed region in the input space. The closed regions which are formed by the neurons overlap each other, like STAR. Genetic algorithms are used to improve the classification performances of the magnetic resonance (MR) images with minimized number of neurons. GANN is examined comparatively with a multilayer perceptron (MLP), and restricted coulomb energy (RCE). It is observed that GANN gives the best classification performance with less number of neurons after a short training time.