Prediction of CKD Using Expert System Fuzzy Logic & AI
Naveen Kumar Pareek, Deepika Soni, Sheshang Degadwala
Abstract
The prevalence of chronic kidney disease (CKD) has grown to become an urgent global health issue. Medical personnel are better able to intervene early and effectively against CKD if the illness is detected and predicted correctly. The combination of expert systems, fuzzy logic, and AI methods has shown encouraging outcomes in several medical fields in recent years. The purpose of this research is to create a model for predicting CKD using an expert system that blends fuzzy logic and AI techniques. To improve prediction accuracy, the proposed expert system combines the knowledge and experience of nephrologists with the capabilities of fuzzy logic and AI algorithms. Age, blood pressure, serum creatinine, and urine protein levels are only few of the clinical and laboratory indicators included into the system. Fuzzy logic is used to process and analyze these variables to better describe and accommodate the inherent ambiguity and imprecision of medical data. In addition, the AI module draws on a variety of machine learning methods, including as decision trees, support vector machines, naive Bayes, and random forest, to analyze past patient records and develop accurate prognostic models. The technology will be tested on a large dataset of CKD patients and compared to conventional methods of diagnosis.