Share Email Print

Proceedings Paper

Artificial convolution neural network with wavelet kernels for disease pattern recognition
Author(s): Shih-Chung Benedict Lo; Huai Li; Jyh-Shyan Lin; Akira Hasegawa; Chris Yuzheng Wu; Matthew T. Freedman M.D.; Seong Ki Mun
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

A two-dimensional convolution neural network (CNN) with wavelet kernels (WK) has been developed for image pattern recognition. The structure of the CNN is a simplified version of the neocognitron. We used only a two-level structure and eliminated all complex-cell layers. Nets between two adjacent layers in the feature selection level of the CNN are selectively interconnected across groups. In this part of the CNN signals processing, each group in the receiving layer receives signals from a group of weights (i.e., kernels). For the forward signal propagation, the product obtained from the kernel convoluting the front layer is collected onto the corresponding matrix element of the receiving layer. In this paper, the convolution kernels of the CNN (CNN/WK) are wavelet based and are trained by a supervised manner. In the development of the CNN/WK, we forced each updated convolution kernel to be orthonormal. Therefore, features (transformed coefficients) selected on the transform domain are linearly independent. Hence, the fully connected layers in the classification level of the CNN can perform more effectively. The applications of the CNN for disease pattern recognition have been very successful. When isolated patterns were further processed by internal filtering and classification layers were built into the neural network structure, the disease patterns were more easily recognized. Although, we did not receive substantial improvement of the ROC performance using the CNN/WK, this method may assist us in the analysis of the trained kernels and eventually lead to the optimization of feature extraction in a course of disease pattern recognition.

Paper Details

Date Published: 12 May 1995
PDF: 10 pages
Proc. SPIE 2434, Medical Imaging 1995: Image Processing, (12 May 1995); doi: 10.1117/12.208730
Show Author Affiliations
Shih-Chung Benedict Lo, Georgetown Univ. Medical Ctr. (United States)
Huai Li, Georgetown Univ. Medical Ctr. (United States)
Jyh-Shyan Lin, Georgetown Univ. Medical Ctr. (United States)
Akira Hasegawa, Georgetown Univ. Medical Ctr. (United States)
Chris Yuzheng Wu, Georgetown Univ. Medical Ctr. (United States)
Matthew T. Freedman M.D., Georgetown Univ. Medical Ctr. (United States)
Seong Ki Mun, Georgetown Univ. Medical Ctr. (United States)

Published in SPIE Proceedings Vol. 2434:
Medical Imaging 1995: Image Processing
Murray H. Loew, Editor(s)

© SPIE. Terms of Use
Back to Top