An Analysis for Predicting Chronic Kidney Disease- Using Machine Learning Approaches

Abstract

Machine learning is the biggest help now-a-days to develop models of the actual biological systems which are both predictive and informative. Bioscience is the field where we can collect a huge amount of data. By using data mining processes, machine learning has become a growing use in biomedical technology because of the ever-increasing volume and complexity of biological data. Chronic kidney disease or chronic kidney failure is a term when kidneys are damaged and the wastes couldn’t be filtered as kidneys always do. High blood pressure and diabetes are the main causes of CKD. The bad situation occurs when the damage leads to kidney transplant. So, early detection is very much needed in this situation and for predicting CKD I have employed some ML techniques by using a dataset with four hundred clinical data. I have used most of the Machine learning algorithms but four of them(Random forest, XGBoost, Ada Boost, LGBM Classifier) gave me promising results and this model's performance is the best to predict CKD with the given dataset.