FUZZY EXPERT SYSTEM FOR THE PROGNOSIS OF LIVER CIRRHOSIS

Abstract

Decision support through expert systems becomes part of everyday life. The aim of this study is to design a fuzzy expert system for the diagnosis of Cirrhosis which is one of the common diseases of the liver. The designed system is based on the sequential
combination of the Child Pugh score. Input fields are based on scoring system of Child pugh score and the output field refers to the risk of cirrhosis. The ChildPugh score (or its associated Child-Pugh grade) is often used as a means to give a very general description of the clinical state of patients with cirrhosis of the liver, and to indicate the severity of the condition. CTP is used to
assess the likelihood of mortality in cirrhotic patients who were undergoing portosystemic shunt surgery to prevent further variceal bleeds. Each patient was assigned a Child’s grading (of A, B or C) to stratify the individual with regard to risk of death due to the procedure. Child’s grade A patients were believed to have the best prognosis, and Child’s grade C patients the worst. Five variables are considered: presence of ascites, encephalopathy, serum levels of albumin, total billirubin, and prolongation of the clotting time. Each of these variables is assigned a score between 1 and 3 according to its severity or degree of abnormality. The system uses Mamdani Inference method. The results obtained from the designed system are compared with the actual data
of patients in the database and observed results of the designed system are well within the limits set by the domain expert. The system can be used as decision support for the prediction of the cirrhosis and can avoid the need of the liver biopsy.