A Hybrid Decision Support System for Chronic Kidney Disease Diagnosis

Abstract

The incidence of chronic renal problems is increasing markedly due to hypertension, diabetes, anemia, and other related conditions. Patients with these diseases may remain unaware of initial symptoms, complicating the diagnostic process. Advanced data mining diagnostic and prediction technologies could facilitate patient self-assessment and assist medical professionals in forming an accurate evaluation of the patient. This research presents a paradigm for a clinical decision support system for Chronic Kidney Disease (CKD) derived from the knowledge and insights of professionals and experts. Diverse classification techniques are utilized and evaluated on the dataset to identify the condition and ascertain the development stage of CKD. The proposed methodology improves accuracy to 91.75% and reduces the cost of forecasting CKD stages by employing LMT algorithms on the dataset.