Abstract: Chronic Kidney Disease (CKD) has emerged as a global health concern. Many people get diseases unexpectedly as a result of risk factors such as food, environment, and living standard. A normal kidney has one million filtering units. Glomerulas are the name given to each filtering unit. High blood pressure and diabetes both contribute significantly to kidney damage. The following kidney functions can be harmed by this disease. And they are: damaging their filtering units, collecting tubules, and causing scarring. A normal or healthy kidney can remove waste products from the blood and maintain an equal chemical level in the human body.
CKD is also defined as either kidney structural damage or a decrease in GFR of less than 60ml/min/1.73m2 for three months or more. It means that slow progressive loss of kidney functions over time caused by progressive destruction of renal mass. Because of this gradual loss of kidney function, CKD frequently goes undetected and undiagnosed until it worsens over time. Gradual kidney function loss can result in End Stage Renal Disease (ESRD), accelerated Cardio Vascular Disease (CVD), and death. This paper proposes a framework to predict the Chronic Kidney disease so that it reduces the time and effort required to detect and predict chronic kidney disease also to create a reliable analysis system. Using such models, most patients at risk of having proteinuria less than 1g/24 hrs can be classified as low risk and potentially treated slowly by their primary patient follow-up.
Keywords: Chronic Kidney Disease, Machine Learning, SVM, KNN, Random Forest, CKD
Copyright Form