Millions of people all around the world are afflicted with the common urological ailment known as kidney stones. Effective kidney stone treatment depends on early detection and accurate estimate of stone formation. The progression of kidney stones can now be predicted using patient-specific data from the recent developments in machine learning and data analytics. The suggested approach entails creating a CNN-based predictive model and training it using a large dataset made up of CT images. The model detects the stone and categorizes it into several stages using CNN technique. Making a diagnosis of the patient's condition—whether it be normal, cystic, stone- or tumor-related involves training the model on annotated medical images to automatically learn discriminative features using a convolutional neural network, enabling accurate classification and supporting clinicians in timely and reliable decision-making.