Improving the Predictive Performance of Chronic Kidney Disease Models Through Data Refinement

Abstract

Chronic kidney disease (CKD) represents a major global health challenge, largely due to its asymptomatic progression during early stages, which hinders timely prediction and intervention. This study explores the application of machine learning models specifically Support Vector Machine (SVM) and Naive Bayes for the accurate prediction and staging of CKD. The incorporation of the estimated glomerular filtration rate (eGFR), calculated using the Modification of Diet in Renal Disease (MDRD) formula, significantly enhances the predictive performance of these models. By integrating this essential clinical biomarker into our dataset, we improve data quality and increase its relevance for CKD diagnosis and monitoring. Experimental results demonstrate that both models can effectively predict CKD at early stages, offering a promising tool for early diagnosis. These findings underscore the importance of clinical feature integration and data refinement in boosting model accuracy and real-world applicability in healthcare contexts.

Ähnliche Arbeiten