Assessment of deep neural network for prediction of chronic kidney disease

This study uses deep learning techniques to detect chronic kidney disease at earlier stage to aid in its prevention. Disorders that affect the kidney's natural function are referred to as kidney diseases. This research recommends using a perceptron classifier with multiple layers classifier that is based on deep neural networks to identify CKD in patients. In order to diagnose chronic kidney disease, a database of 400 individuals with 25 attributes was employed. For the tests, the means of the respective attributes were used to replace each missing value in the original data set. The deep neural network's optimal parameters were then produced by fine-tuning the parameters and running several trials. In this research, A deep neural network is designed and developed. The performance of proposed model is compared with various cutting-edge machine learning techniques. Experiments show that the proposed model performs with 99.85% testing accuracy in classification tasks.