Determining Fuzzy Logic Parameters by using Genetic Algorithm for the Diagnosis of Liver Disease

Liver disease is one of the most common chronic liver problems and cirrhosis. Liver problems include a wide range of diseases and disorders that damage the liver tissue or its function. Early diagnosis and treatment of this disease can reduce the severity of the disease and mortality rate. In this descriptive-analytic study, database was consisted of 583 independent records, including 11 features in the UCI machine learning database and through using fuzzy logic that its parameters are determined by Genetic Algorithm (GA), a method for the diagnosis of liver disease is proposed. For this purpose, first, the features of the dataset were ranked using the entropy feature and then, the dataset data were optimized using GA. Ultimately, liver disease was diagnosed using the genfis2 and genfis3 Fuzzy Inference System (FIS). The results show that the accuracy of detection of liver disease using the genfis2 FIS with 8 features is 91.66% and using the genfis3 FIS with 6 features, it is equal to 89.87%. Moreover, the rates of error for genfis2 and genfis3 were 0.034 and 0.047 respectively. Liver disease is one of the most common diseases in population. Early diagnosis of disease while reducing costs can increase the chance of treatment success. According to the obtained results, the proposed model can identify people with liver disease with a fairly high degree of accuracy.