Revised GMDH-type neural networks with a feedback loop and their application to the medical image recognition

Abstract

In this paper, radial basis function networks are identified by using the revised GMDH (Group Method of Data Handling)-type neural network algorithm With a feedback loop and the identified neural networks are applied to medical image recognition of the brain. The GMDH-type neural networks can automatically organize themselves by using a heuristic self-organization method. The useful input variables, the number of the hidden layers, the number of the neurons in the hidden layers and optimum architectures of the neurons in the hidden layers are automatically selected so as to minimize an error criterion defined as Akaike's information criterion (AIC). The GMDH-type neural networks with a feedback loop can identify the complex nonlinear system very accurately by using the feedback loop calculations. In this paper, the revised GMDH-type neural networks with a feedback loop are applied to the medical image recognition of the brain and it is shown that this neural network algorithm is very useful in the medical image recognition of the brain.