Key points are not available for this paper at this time.
MyKidney is an accessible, web-based software platform designed to assist in the automated detection of kidney stones from axial computed tomography (CT) images. By leveraging a custom convolutional neural network (CNN), the system classifies CT slices and visually highlights potential stone regions, aiding in preliminary diagnosis. The model achieved a classification accuracy of 98.80% , demonstrating a superior balance between sensitivity and specificity when compared to established architectures like ResNet-18 and VGG19. The platform provides a cost-effective, real-time solution for clinics, particularly in under-resourced or rural areas where access to radiology expertise may be limited. Users can upload images directly via the web interface, and the system processes the data using a trained deep-learning model hosted on a secure backend. This software provides a potential entry point for further Artificial Intelligence (AI) assisted diagnostic tools and can be integrated into telehealth systems. We welcome potential collaborators interested in further validating or extending this tool for broader medical imaging applications.
Abdalla et al. (Wed,) studied this question.