Chronic Kidney Disease Analysis using Various Biological Parameters under Machine Learning Techniques

Kidney chronic disease (KCD) has become one of the most critical illnesses which should be diagnosed properly within a certain limit of time for better treatment. Nowadays different algorithms being used for treatment of patients. By using the best predictory model, doctor can easily predict the disease on time. For this verdict, Kidney Chronic Disease Prediction has been discussed in this article. In this article various datasets have been used by the different article authors to predict the disease which are pre-processed with different methods like Univariate Feature Selection (UFS) and Recursive Feature Elimination (RFE), Data mining techniques, etc. which are further used to train and test various ML models get the accuracy by predictory models. The scores gained by the predictory models is then validated to check the precision of the output accuracy of by the model using methods like performance matrices, various cross validation techniques, Mathew Correlation Coefficient (MCC). This whole article is all around the analyses of different articles based on the machine learning and validation techniques with respect to the accuracy. Most of the models attained the accuracy of 99 percent and above. In this article, along with machine learning models it also discussed about two neural networks, which attained the accuracy between 93 to 98% with a good precision value.