Survey on Diagnosis and Prediction Methods of Chronic Kidney Diseases Using Data Mining Techniques

The significant growth of kidney disease, its effects and complications, and the costs that are imposed on the society, have caused the medical community to look for programs to predict and diagnose this disease early. In recent years, the use of data mining techniques in the field of medicine has gained great importance. The purpose of this article is to review methods based on data mining techniques in order to predict chronic kidney diseases. In this article, it is intended to collect and review a comprehensive review of chronic kidney disease, diagnosis methods in the field of medicine.  In this paper, chronic kidney diseases were examined first, and following this examination, we will discuss and analyze the field of early diagnosis of these diseases. In the following, we will introduce various data-mining algorithms in the diagnosis of chronic kidney diseases. After that, we will discuss the methods of diagnosing and predicting chronic kidney diseases, and we will review these methods in terms of their goals, limitations, and capabilities. In this article, in order to predict the failure disease, all the different points of view, which are: feature selection, classification method, used tools, are examined, which we show by analyzing the classification algorithms. The use of random forest algorithm, k-nearest neighbor algorithm and support vector algorithm along with feature selection methods can play an effective role in predicting kidney failure disease. Also, the results show that the UCI database and MATLAB tool are the most useful in diagnosing and predicting chronic kidney diseases.