A novel fast fuzzy neural network backpropagation algorithm for colon cancer cell image discrimination

Authors:
Ephram Nwoye;Li C. Khor;Satnam S. Dlay;Wai L. Woo
Affiliations:
School of Electrical, Electronic and Computer Engineering, University of Newcastle upon Tyne, United Kingdom;School of Electrical, Electronic and Computer Engineering, University of Newcastle upon Tyne, United Kingdom;School of Electrical, Electronic and Computer Engineering, University of Newcastle upon Tyne, United Kingdom;School of Electrical, Electronic and Computer Engineering, University of Newcastle upon Tyne, United Kingdom
Venue:
ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part III
Year:
2006

Citing 6
Cited 0

Image Analysis Using Mathematical Morphology

IEEE Transactions on Pattern Analysis and Machine Intelligence
Neural Networks and Artificial Intelligence for Biomedical Engineering

Neural Networks and Artificial Intelligence for Biomedical Engineering
Digital Image Processing

Digital Image Processing
Self-adaptive neuro-fuzzy inference systems for classification applications

IEEE Transactions on Fuzzy Systems
Neuro-fuzzy rule generation: survey in soft computing framework

IEEE Transactions on Neural Networks
New dynamical optimal learning for linear multilayer FNN

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper a novel fast fuzzy backpropagation algorithm for classification of colon cell images is proposed. The experimental results show that the accuracy of the method is very high. The algorithm is evaluated using 116 cancer suspects and 88 normal colon cells images and results in a classification rate of 96.4%. The method automatically detects differences in biopsy images of the colorectal polyps, extracts the required image texture features and then classifies the cells into normal and cancer respectively. The net function computation is significantly faster. Convergence is quicker. It has an added advantage of being independent of the feature extraction procedure adopted, with knowledge and learning to overcome the sharpness of class characteristics associated with other classifiers algorithms. It can also be used to resolve a situation of in-between classes.