Chronic kidney disease (CKD) is a global health issue with a high mortality rate and it is the root cause of many other diseases. Patients fail to recognise the disease as there aren't any obvious signs in the beginning. Early symptom identification is essential for providing effective treatment. To solve this problem, CKD was predicted using machine learning algorithms. In this study, CKD was predicted using convolutional neural networks (CNN) and long short-term memory (LSTM). We examine the accuracy, precision, F1-score, and recall of CNN and LSTM to see which performs better. The UCI Machine Learning repository is where the dataset was gathered