Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/122505
Title: A Novel Approach for Diagnosing Hepatitis Viruses Using Generalized Regression Neural Network
Researcher: C.Mahesh
Guide(s): E.Kannan
Keywords: Neural Networks, Hepatitis A, Hepatitis B, Hepatitis C, Generalized Regression Neural Network
University: Vel Tech Dr. R R and Dr. S R Technical University
Completed Date: 18-11-2016
Abstract: The most important issues in the field of medical science are the diagnosis of diseases. Medical diagnostics is pretty difficult which is almost done by medical experts. There are two most common problems in automatic diagnostic field. One is selecting the significant parameter set for the perfect diagnostics; the next is designing a steady and powerful algorithm which must not require longer time to run. Nowadays Neural Networks have become a widely used in the field of medical diagnosis. A very good solution could be attained for medical issued by employing neural network algorithms. newline This work analyzed the artificial intelligence application in conventional hepatitis virus diagnosis. By taking into account of the importance of the issues of medical diagnosis, our work investigates the application of an intelligent system based on artificial neural network for decision making for Hepatitis disease. In this work, an expert system with an innovative algorithm using Generalized Regression Neural Network is deployed and implemented for the diagnosis of hepatitis disease. newline Diagnosis process requires a medical expert in handling the approximation which is not available in the timely fashion. Computer-aided medical diagnosis is essential to assist doctors for making certainty without direct consultation with the medical expertise. Hepatitis disease is a serious liver infection which is influenced by the hepatitis virus. Hepatitis disease is a significant physical issue and the viral hepatitis is one of the most serious types of viral in the world. Mortality rate could be increased by computer aided programs or system designed by software, or software application by imitation medical expert s intelligence which can be used by the normal physician while diagnosing the disease. In such case the help of medical expert is not necessary for the normal physician to take decision in the diagnosis process. newline newline We have used Generalized Regression Neural Network as our expert system to diagnose hepatitis disease. A novel algorithm is been implemented using Generalized Regression Neural Network which improves accuracy of hepatitis diagnosis. Using this algorithm it is possible to use only the significant hepatitis attributes for the diagnosis. By attribute filter method significant attributes are extracted. For attribute filtering rough set method is been used. Proper pre-processing is done with the help of normalization method. The main objective of implementing this novel algorithm in Generalized Regression Neural Network is to diagnose hepatitis disease more effectively in terms of accuracy by using very less number of hepatitis attributes. newline We have proposed one more approach for the diagnosis of hepatitis disease using Generalized Regression Neural Network with less number of training. Probability density function algorithm is been employed in Generalized Regression Neural Network in such a way that only less number of training is necessary to train the Generalized Regression Neural Network. The objective of this method is to diagnose hepatitis disease more accurately using Generalized Regression Neural Network with minimum number of training cycles. newline A classifier with a new algorithm is proposed to classify the type of hepatitis virus. Logical inference method is implemented to classify the virus. Finally we have done the performance comparison of the above two approaches and the performance comparison with the previous works done. Our proposed system is tested in two hospitals and it has been proved that our proposed system works better. newline newline
Pagination: 
URI: http://hdl.handle.net/10603/122505
Appears in Departments:Department of Computer Science and Engineering

Files in This Item:
File Description SizeFormat 
abstract.pdfAttached File167.75 kBAdobe PDFView/Open
acknowledgement.pdf85.41 kBAdobe PDFView/Open
appendix.pdf223.76 kBAdobe PDFView/Open
biography.pdf85.54 kBAdobe PDFView/Open
certificates.pdf455.16 kBAdobe PDFView/Open
chapter 1 .pdf845.84 kBAdobe PDFView/Open
chapter 2.pdf259.76 kBAdobe PDFView/Open
chapter_3.pdf677.34 kBAdobe PDFView/Open
chapter_4.pdf498.57 kBAdobe PDFView/Open
chapter_5.pdf338.54 kBAdobe PDFView/Open
chapter_6.pdf82.42 kBAdobe PDFView/Open
chapter_7.pdf82.95 kBAdobe PDFView/Open
contents.pdf160 kBAdobe PDFView/Open
publications.pdf180.17 kBAdobe PDFView/Open
references.pdf278.83 kBAdobe PDFView/Open
title.pdf178.48 kBAdobe PDFView/Open


Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.

Altmetric Badge: