Search

A revised Group Method of Data Handling (GMDH)-type neural network algorithm for medical image diagnosis is proposed, and is applied to medical image diagnosis of lung cancer. In this algorithm, the knowledge base for medical image diagnosis are used for organizing the neural network architecture for medical image diagnosis, and the revised GMDH-type neural network algorithm can identify the characteristics of the medical images accurately. The optimum neural network architecture fitting the complexity of the medical images is automatically organized so as to minimize the prediction error criterion defined as Prediction Sum of Squares (PSS), and it is shown that the revised GMDH-type neural network can be easily applied to the medical image diagnosis. [ABSTRACT FROM PUBLISHER]

The feedback Group Method of Data Handling (GMDH) -type neural network algorithm is applied to the medical image diagnosis of lung cancer. 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) or Prediction Sum of Squares (PSS). The identification results show that the feedback GMDH-type neural network algorithm is useful for the medical image diagnosis of lung cancer since the optimum neural network architecture is automatically organized so as to fit the complexity of the medical images. [ABSTRACT FROM PUBLISHER]

A hybrid Group Method of Data Handling (GMDH)-type neural network algorithm using artificial intelligence technology for medical image diagnosis is proposed and is applied to medical image diagnosis of the liver cancer. In this algorithm, the knowledge base for medical image diagnosis is used for organizing the neural network architecture for medical image diagnosis. Furthermore, the revised GMDH-type neural network algorithm has a feedback loop and can identify the characteristics of the medical images accurately using feedback loop calculations. The optimum neural network architecture fitting the complexity of the medical images is automatically organized so as to minimize the prediction error criterion defined as Akaike's information criterion (AIC) or Prediction Sum of Squares (PSS). It is shown that the hybrid GMDH-type neural network is accurate and a useful method for the medical image diagnosis of the liver cancer. [ABSTRACT FROM PUBLISHER]

In this study, a revised Group Method of Data Handling (GMDH)-type neural network self-selecting optimum neuron architectures is applied to the computer aided image diagnosis (CAD) of lung cancer. The GMDH-type neural network algorithm has an ability of self-selecting optimum neural network architecture from three neural network architectures such as sigmoid function neural network, radial basis function (RBF) neural network and polynomial neural network. The GMDH-type neural network also has abilities of self-selecting the number of layers, the number of neurons in hidden layers and useful input variables. This algorithm is applied to CAD and it is shown that this algorithm is useful for CAD of lung cancer and is very easy to apply practical complex problem because optimum neural network architecture is automatically organized. [ABSTRACT FROM PUBLISHER]