Abstract: In the article the application of fuzzy neural network NefClass? with Gaussian and triangular membership functions to pattern recognition of objects on medical images obtained by colposcopy. The characteristics of each group of diseases of cervix uterus are analyzed. Using the values of features by the color model RGB the class of the input sample is determined. Received samples are processed using fuzzy neural network RBF and a class of sample is determined. A comparative analysis of the recognition results using the fuzzy network with obtained results of crisp RBF neural network is performed.

Keywords: medical images, pattern recognition, fuzzy neural network

Link:

RECOGNITION OF OBJECTS ON OPTICAL IMAGES IN MEDICAL DIAGNOSTICS USING FUZZY NEURAL NETWORK NEFCLASS

Yuriy Zaychenko, Vira Huskova

http://www.foibg.com/ijima//vol05/ijima05-01-p02.pdf