Medical Image Recognition of Abdominal Multi-organs by Hybrid Multi-layered GMDH-type Neural Network Using Principal Component-Regression Analysis

In this study, hybrid multi-layered Group Method of Data Handling (GMDH) type neural network algorithm using principal component-regression analysis is applied to recognition problems of the abdominal multi-organs such as the liver and spleen. In this GMDH-type neural network, principal component-regression analysis is used as the learning algorithm of the weights in the GMDH-type neural network which is a type of the deep neural network with many hidden layers. The architecture of the deep neural network with many hidden layers is automatically organized using the heuristic self-organization method, so as to minimize the prediction error criterion defined as Akaike´s information criterion (AIC) or Prediction Sum of Squares (PSS). The heuristic self-organization method is a type of the evolutional computation. In the GMDH-type neural network, the multi-co linearity occurs and the prediction values become unstable because the architecture of the neural network has many hidden layers whose characteristics are very complex. In the GMDH-type neural networks in this study, multi-co linearity does not occur and stable and accurate prediction values are obtained. This new algorithm is applied to the medical image recognitions of the liver and spleen. The optimum neural network architectures, which fit the complexity of the liver and spleen images, are automatically organized from the multi-detector row CT (MDCT) image of the abdominal regions and the liver and spleen regions are automatically recognized and extracted by the organized GMDH-type neural networks. The recognition results are compared with the conventional sigmoid function neural network trained using back propagation method and it is shown that this GMDH-type neural networks are useful for the medical image recognition problems of the abdominal multi-organs.