This project proposes an automated kidney stone detection system leveraging deep learning techniques applied to CT (Computed Tomography) images. The system aims to accurately predict the presence or absence of kidney stones, aiding medical professionals in prompt diagnosis and treatment planning. By utilizing a convolutional neural network (CNN) architecture and transfer learning from pre-trained models, the system achieves efficient feature extraction and classification. Through rigorous training and optimization, including validation and performance evaluation, the system demonstrates its potential to enhance diagnostic accuracy and streamline clinical workflows. Kidney stones are a common medical illness which cause severe pain and difficulties if left untreated. Timely detection is crucial for effective management, yet manual interpretation of Computed Tomography (CT) images is labor-intensive and subject to human error. In response, this project proposes an automated kidney stone system harnessing the power of deep learning techniques applied to CT images. The system's primary objective is to accurately predict the presence or absence of kidney stones, thereby aiding medical professionals in prompt diagnosis and treatment planning. Central to the system's effectiveness is its utilization of a convolutional neural network (CNN) architecture, coupled with transfer learning from pre-trained models. This approach enables efficient feature extraction and classification, allowing the system to effectively discern subtle patterns indicative of kidney stones. Through rigorous training and optimization procedures, including validation and performance evaluation, the system showcases its potential to significantly enhance diagnostic accuracy compared to conventional methods. © 2025 Elsevier B.V., All rights reserved.