Deep Feedback GMDH-Type Neural Network Using Principal Component-Regression Analysis and Its Application to Medical Image Recognition of Abdominal Multi-Organs

Authors
Tadashi Kondo, Junji Ueno, Shoichiro Takao
Corresponding Author
Tadashi Kondo
Available Online 1 September 2015.
DOI
https://doi.org/10.2991/jrnal.2015.2.2.6
Keywords
Deep neural networks, GMDH, Medical image recognition, Evolutionary computation
Abstract
The deep feedback Group Method of Data Handling (GMDH)-type neural network is proposed and applied to the medical image recognition of abdominal organs such as the liver and spleen. 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. The neural network architecture is automatically organized so as to fit the complexity of the medical images using the prediction error criterion defined as Akaike’s Information Criterion (AIC) or Prediction Sum of Squares (PSS). The recognition results show that the deep feedback GMDH-type neural network algorithm is useful for the medical image recognition of abdominal organs.

Copyright
© 2013, the Authors. Published by ALife Robotics Corp. Ltd.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)