Chronic Kidney Disease Prediction Using Machine Learning Approach

Abstract

Machine learning is currently playing a very important role in various sectors. It also has a significant position in the healthcare sector for its working efficiency. Machine learning classification algorithms are popular in medical science for predicting various complex diseases. Kidney disease is a familiar name to all of us nowadays. Moreover, this disease has become a major public health problem for people worldwide. People are affected with kidney disease when they do not follow a proper diet in their daily life, lack of proper health awareness including drinking a very little amount of water. As a result, the disease later turned into a terrible disease, which is called CKD. Countless people all over the world are suffering from this disease and they are dying due to a lack of proper awareness and treatment. The treatment of this disease is extremely expensive and difficult. When a patient is affected with CKD, their kidneys stop working completely. There is no limit to the suffering of patients when the kidneys stop their activity. In this research, we have tried to predict CKD by applying various classification algorithms of machine learning. Accurate data is very important to do this work so that we have done this work with the proper dataset. Our goal is to identify and predict chronic kidney disease. To predict CKD, we used four popular machine learning classification algorithms. Which are respectively SVM, Decision Tree, Random Forest, KNN. We have basically completed our work in two steps. In the first step we have completed the training of the data and in the next step completed the testing. After trained by the machine learning algorithm, we got our desired result. In that case, the Decision tree algorithm has shown the best performance among the four classification algorithms.