Prediction Of Kidney Stones Using CNN

Kidney-related diseases such as cysts, stones, and tumors are among the most common urological disorders worldwide, and their early detection is critical for effective treatment and improved patient outcomes. Traditional manual interpretation of Computed Tomography (CT) scans by radiologists is time- consuming, subjective, and prone to diagnostic variability. To address these challenges, this project presents an AI-powered Kidney Scan Analyzer that leverages Convolutional Neural Networks (CNNs) for automated multi-class classification of kidney CT scans. The system classifies scans into four categories—Normal, Cyst, Stone, and Tumour—while providing confidence scores to support medical decision-making. Implemented as an end-to-end web application with a Flask backend and a React + TypeScript frontend, the solution enables seamless image upload, real-time prediction, and user-friendly visualization. Extensive testing demonstrated an average accuracy of approximately 90%, with response times under five seconds, validating the system’s efficiency and reliability. This project serves as a practical decision-support tool for radiologists, improves diagnostic consistency, and provides a foundation for future expansion into multi-organ disease detection.