Stoneuro: Stone Detection Unveiled Through Advanced AI Models

Renal lithiasis, an epidemic defined by the development of stones in the kidneys, represents an important global health issue due to its severe signs and prevalence. These stones, made from minerals including calcium, oxalate, and phosphorus, differ with regard to size and frequently lead to tremendous discomfort while they traverse the urinary tract. Being susceptible to renal lithiasis can be affected by variables like dehydration, eating habits, heredity, and age, with males demonstrating higher risk. Kidney stone detection that is both accurate and early is essential for prompt intervention. Our paper presents a novel deep learning method for nephrolith diagnosis that makes use of the VGG16, MobileNetV2, and a Convolutional Neural Network models. These models have been instructed using medical pictures (CT scans) to recognize the presence of calculus, cancerous growths, or cyst in the kidneys. Our results demonstrate the superior performance of CNN, achieving an accuracy of 98%, followed by VGG16 (96%) and MobileNetV2 (94%). The high accuracy of the CNN model showers its potential in terms of a powerful tool to aid medical professionals in diagnosing renal lithiasis and guiding appropriate treatment strategies.