Designing a Fuzzy Expert System of Diagnosing the Hepatitis B Intensity Rate and Comparing it with Adaptive Neural Network Fuzzy System

摘要

in this paper an adaptive neural fuzzy system has been designed for diagnosing the hepatitis B intensity rate. The main problem in determining the disease intensity is not having information about the data variation rate and its resulting effect on the system. Designing a fuzzy expert system and using a neural network for training then testing the system adaptively has resulted in a very good optimization. A Hepatitis B data bank has been collected in accordance with the recent medical findings about this disease and the endorsement of a liver specialist. This bank has 300 records and each record has 7 fields. This bank has been assembled from patients presenting at the liver biopsy department of Imam Reza hospital Mashad, Iran. Using specialist research and experience strong inference rules have been attained. Thus, the accuracy oft the system in diagnosing the hepatitis B intensity is 96.4±0.2%.