Medical image diagnosis of liver cancer by hybrid feedback GMDH-type neural network using principal component-regression analysis

Abstract

organize the neural network architectures using heuristic self-organization method [3,4] which is a type of the evolutionary computation. In this study, hybrid feedback GMDH-type neural network algorithm is applied to the medical image diagnosis of liver cancer. The learning calculations of the weights are the principal componentregression analysis and the accurate and stable predicted values are obtained. This hybrid feedback GMDH-type neural network has a feedback loop and the optimum neural network architecture is automatically selected from three types of neural network architectures such as the sigmoid function neural network, the radial basis function (RBF) neural network and the polynomial neural network so as to minimize the prediction error criterion defined as AIC [5] or PSS [6][7][8]. The hybrid feedback GMDHtype neural network algorithm is applied to the medical image diagnosis of liver cancer and the recognition results are compared with those obtained using the conventional neural networks trained using the back propagation algorithm. Hybrid feedback GMDH-type neural networkThe architecture of the hybrid feedback GMDH-type neural network developed in this paper has a feedback loop as shown in Fig. 1. The hybrid feedback GMDH-type neural network algorithm uses three types of neuron architectures which are the sigmoid function neuron, the RBF neuron and the polynomial neuron. In the hybrid feedback GMDH-type neural network, optimum neuron architectures fitting the characteristics of the nonlinear system are automatically selected using AIC [5] or PSS [6][7][8]. The hybrid feedback GMDH-type neural network is shown as follows.Abstract The hybrid feedback group method of data handling (GMDH)-type neural network is proposed and applied to the medical image diagnosis of liver cancer. In this algorithm, the principal component-regression analysis is used for the learning calculation of the neural network, and the accurate and stable predicted values are obtained. Furthermore, this neural network has the feedback loop and the complexity of neural network architecture is increased using the feedback loop calculations. The neural network architecture is automatically organized so as to fit the complexity of the nonlinear system using the prediction error criterion defined as Akaike's information criterion (AIC) or prediction sum of squares (PSS). The recognition results show that the hybrid feedback GMDH-type neural network algorithm is useful for the medical image diagnosis of liver cancer since the optimum neural network architecture is automatically organized.

Correlation of Experimental Liquid–Liquid Equilibrium Data for Ternary Systems Using NRTL and GMDH-Type Neural Network

“…GMDH algorithm was first proposed as a polynomial neural network for identification and modeling complex systems by Ivakhnenko . GMDH-type NN has been widely used in many engineering applications. − …”

Hybrid feedback GMDH-type neural network using principal component-regression analysis and its application to medical image recognition of heart regions

“…The heuristic self-organization method is a type of the evolutional computation. In this study, the hybrid feedback GMDH-type neural network using principal component-regression analysis [1], [2] is applied to the medical image recognition of heart regions. In this feedback GMDH-type neural network algorithm, the structural parameters such as the number of feedback loops, the number of neurons in the hidden layers and the relevant input variables are automatically selected so as to minimize the prediction error criterion defined as Akaike's Information Criterion (AIC) [7] or Prediction Sum of Squares (PSS) [8].…”

“…The GMDH-type neural network algorithms were proposed in our early works [1]- [4] and the GMDH-type neural networks were applied to the medical image diagnosis of lung cancer [1], liver cancer [2] and the medical image analysis of the heart [3]. The GMDH-type neural network can automatically organize the neural network architectures by using heuristic selforganization method [5], [6] which is the basic premise of the GMDH algorithm [5], [6], [8].…”

Medical Image Diagnosis of Lung Cancer by Deep Feedback GMDH-Type Neural Network

“…First, the lung regions were recognized by the logistic GMDHtype neural network and these lung regions were extracted and then the lung cancer regions were extracted using the image post processing of the extracted lung regions. In another previous our work 4 , we applied the hybrid feedback GMDH-type neural network to medical image diagnosis of liver cancer. The liver cancer regions were recognized and extracted using the same image processing using the hybrid feedback GMDH-type neural network.…”