AI-Based MRI Analysis for Bone Fracture Detection

Abstract

Bone fractures are a common medical issue brought on by sports injuries, accidents, or underlying medical conditions like osteoporosis. Effective treatment and recovery depend on an early and accurate diagnosis, but standard MRI scan analysis takes a lot of time and specialized knowledge. In order to automate and improve fracture identification accuracy, this research proposes an AI-powered MRI scan analysis system that combines CNN and YOLO. The suggested model effectively detects fractures by utilizing CNN's deep feature extraction and YOLO's object detection capabilities, cutting down on diagnostic time and enhancing consistency. A. Using sophisticated image preprocessing methods including median filtering for noise reduction and transfer learning for improved generalization, the system is trained on a sizable dataset of MRI images. High diagnostic reliability is indicated by performance criteria such as precision (92%), recall (89%), and F1-score (90.5%). The technology connects scan center, physicians, and patients in a seamless manner to facilitate quicker decision-making in actual hospitals. Patients can upload MRI scans through the system's web-based interface, and the trained YOLO-CNN model analyzes them for possible fractures.