Development of method to analyze factors of kidney disease by the use of fuzzy logic

Abstract

This research suggests a new method of analyzing factors associated with kidney disease through fuzzy logic. Fuzzy logic is utilized here to group parameters of interest associated with kidney health on a finer scale than mere measures of statistics. The data set of this study is wide-ranging clinical factors ranging from age, blood pressure, specific gravity, albumin, sugar, random blood glucose, blood urea, serum creatinine, sodium, potassium, hemoglobin, packed cell volume, white blood cell count, and red blood cell count. Fuzzy logic centroids are calculated to divide these factors into levels of low, medium, and high to better interpret clinical data.

This study makes use of fuzzy logic to handle natural uncertainty and variability of clinical measurements, unlike classical measures that only use mean, median, and standard deviation-based descriptive measures. The use of fuzzy memberships to sum up variables makes interpretation more adaptable, particularly where category thresholds are indeterminate. By unearthing hidden relationships that may have gone unobserved without it, fuzzy logic centroids give greater insight into trends within the range of clinical parameters. 

The systematic approach of analyzing causes of kidney disease using innovative combinations of classical statistical methods and fuzzy logic here enhance our understanding of processes of kidney disease by investigating in a deep manner interactive relationships of kidney factors. The findings have important implications on how predictive and diagnostic models of kidney disease can be developed, supplementing literature on health care use of fuzzy logic.