Deep Learning-Based Chronic Kidney Disease Prediction Using CNN and YOLOv8 with Ultrasonic X-ray Imaging

Chronic Kidney Disease (CKD) is a progressive condition that affects millions of people worldwide, often remaining undiagnosed until it reaches an advanced stage. Early detection is essential for improving patient outcomes and reducing long-term health complications. This project presents a deep learning-based approach for the prediction and detection of CKD using Convolutional Neural Networks (CNN) and the YOLOv8 algorithm, integrated with real-time camera input from ultrasonic X-ray imaging. A camera is used to capture ultrasonic X-ray visuals of the kidneys, which are then processed using CNN for feature extraction and classification. YOLOv8 is applied for real-time object detection, enabling accurate identification and localization of kidney stones and other abnormalities associated with CKD. The system is implemented using Python and trained on a dataset of kidney ultrasound images. By combining the power of deep learning with modern detection algorithms, this project aims to provide a reliable, non-invasive, and automated tool to assist healthcare professionals in early-stage CKD diagnosis.