Kidney stones are a very common health issue around the world, causing a lot of pain and hospitalizations for many people. They are solid deposits of minerals and salts that form inside the kidneys and can obstruct the urinary tract. The causes and types of kidney stones vary, but they can be influenced by factors such as diet, hydration, genetics, and medications. There is various imaging techniques applied for diagnosing this condition; however, they have to be analysed and interpreted by an expert. The most common ones include ultrasound, X ray, intravenous urography, and computed tomography (CT). Of all, the most precise and reliable is the CT because it can even identify small, non-calcified stones. However, CT also involves a higher dose of radiation that a patient may have and requires manual inspection from radiologists regarding all the images. Computer-aided diagnosis systems are useful methods that can assist doctors with their decisions. Automating processes related to detection and measurement on the basis of medical images about kidney stones saves human error and workload. They can also offer supplementary information and insights that could help in the treatment and prevention of kidney stones. For example, they can categorize the stones based on their composition, location, and shape, and also estimate the probability of spontaneous passage or the need for surgical intervention. In this project, we applied a deep learning (DL) approach to come up with automatic diagnosis of kidney stones based on Computed Tomography (CT) images, technology that is new in the field. of AI.