Development of a Multilayer Fuzzy Expert System for the Diagnosis of Chronic Kidney Disease

In the initial stages of chronic kidney disease, it has some hidden characteristics that lead to a delay in its detection and diagnosis. The growth of kidney impairment or damage can be stopped or slowed down by making an early diagnosis. Hence, in this paper, to overcome this issue, a multilayer fuzzy expert system has been developed by using fuzzy logic, which assists in the diagnosis of chronic kidney disease at its early stages. The developed system has two layers in which the first layer is used to detect if a patient has clinical symptoms of having CKD. Similarly, the second layer of the system is used to evaluate the current stage of chronic kidney disease from which an individual is suffering. The input variables for layer 1 are age, diabetic Mellitus, smoking, hypertension and family history. Moreover, the laboratory tests are considered as the input variables for later 2 to confirm the stage of the disease. Hence, glomerular filtration rate, serum creatinine, albumin, blood urea nitrogen and pus cell in urine are the input variables for layer 2. This research work also evaluated the performance of the developed system on the basis of classification accuracy. As a result, this system provided 99.33% classification accuracy for the diagnosis of chronic kidney disease.