Support Vector Machine (SVM) and Artificial Neural Networks (ANN) based Chronic Kidney Disease Prediction
C R Vishwanatha, V Asha, Arpana Prasad, Shyamal Das, Sunay Kumar, S P Sreeja
Abstract
Despite the high frequency and severe morbidity and death rates associated with cardiovascular disease, chronic kidney disease (CKD) remains woefully understudied. The majority of current practice recommendations are based on tiny underpowered prospective trials, retrospective evaluations, and the assumption that patients with chronic kidney disease would benefit equally from drugs and therapies as those with normal renal function. Furthermore, due to the prior absence of a clear definition of chronic kidney disease and how to quantify renal function, definitions of the degree of renal impairment have varied greatly, further complicating the data. Surprisingly, despite the fact that patients with chronic kidney disease are at higher risk of cardiovascular problems than those with diabetes mellitus, patients with chronic kidney disease are routinely excluded from therapeutic studies. Improvements in renal dosage guideline compliance are required in all areas where data is available, based on limited published data. All studies dealing with renal dosage guideline adherence in inpatient, long-term care, or ambulatory settings were considered. Six articles that met our inclusion criteria were examined. More research is needed to determine the appropriateness of renal dosage in ambulatory settings and to guide quality improvement initiatives across the board. The paper uses the concept of support vector machine and artificial neural network to predict Chronic Kidney Disease. The proposed algorithm has yielded better results in predicting Chronic Kidney Disease.