Title :
Designing an expert system of liver disorders by using neural network and comparing it with parametric and nonparametric system
Author :
Neshat, Mehdi ; Yaghobi, Mehdi ; Naghibi, Mohammad
Author_Institution :
Dept. of Comput. Eng., Azad Univ. press Mashhad, Mashhad
Abstract :
The In this essay, we are going to design a medical expert system by using neural network. In order to test the system, we used the data in the intended bank e.g. BUPA liver disorders. We compare the operation of the system with parametric methods -like Bayesian decision making theory- and non parametric methods. By analysis of the data, we first discovered an undesirable field in the bank which causes to decrease learning rate of the system. When we omitted this field, we concluded very good results. By comparing the three above systems, we concluded that the neural network has the best operation and effect in liver disorder diagnosis. This result has also been improved so far.
Keywords :
Bayes methods; decision theory; liver; medical diagnostic computing; medical expert systems; neural nets; Bayesian decision making theory; liver disorder diagnosis; medical expert system design; neural network; nonparametric system; parametric system; Bayesian methods; Biomedical engineering; Computer architecture; Computer networks; Design engineering; Expert systems; Liver; Medical expert systems; Neural networks; Signal design; Bayesian theory; Medical expert system; liver disorder; neural network; parametric method;
Conference_Titel :
Systems, Signals and Devices, 2008. IEEE SSD 2008. 5th International Multi-Conference on
Conference_Location :
Amman
Print_ISBN :
978-1-4244-2205-0
Electronic_ISBN :
978-1-4244-2206-7
DOI :
10.1109/SSD.2008.4632783
