Chronic Kidney Disease Prognosis Using Various Machine Learning Based Classifiers

Abstract

Chronic kidney disease (CKD) is a lifelong, dangerous disorder caused by either kidney cancer or decreased kidney function. The progression of this chronic illness can be stopped or slowed down to an end stage where the only ways to save a patient’s life are dialysis or surgery. Therefore, early-stage identification of kidney disease is necessary. Patients who are aware of potential risk factors will be able to recognize the dangers of CKD and receive treatment early. Chronic kidney impairment is more common in older adults, those with diabetes, and those with high blood pressure. In order to conquer these obstacles, methods based on machine learning have been employed to quickly diagnose diseases. Stochastic Gradient Boosting (SGB), XgBoost, Cat Boost Classifier, Decision Tree Classifier, Random Forest Classifier, AdaBoost Classifier, Gradient Boosting Classifier, Extra Trees Classifier, and KNN Classifier are some of the machine learning techniques that have been utilized in this work. These algorithms were trained using data from the UCI repository. Among all the classifiers that have used in work, it has been found that Extra Trees Classifier (ETC) performs really well and got the highest accuracy of 99 percent.

Ähnliche Arbeiten