Chronic kidney diseases commonly known as chronic renal illness, is a kind of kidney disease is characterized by abnormal kidney function or breakdown of renal function that takes a long time or months to complete. Chronic renal diseases are frequently discovered during those individuals are recognised to be at danger of renal obstacle, those who have high blood pressure or diabetes, for example, are tested, as well as those who have a blood family who has chronic kidney disease as a result, swiftly prognosis is a critical in battling the disease and providing effective therapy. Deep learning implementation techniques such as CNN; RNN for CKD is suggested in this work. Using the fewest features, the final outcome forecasts either the person has CKD or not. In our research we will be using algorithms the Convolution Neural network (CNN) as existing system; recurrent neural Network (RNN) as proposed system and compared in terms of accuracy. From the results its proved that Recurrent Neural network (RNN) performs better than Convolution Neural network (CNN).