Innovative Fracture Diagnosis: MobileNet CNN Approach for Precise Bone Fracture Detection and Classification
Khushi Mittal, Kanwarpartap Singh Gill, Rahul Chauhan, Akanksha Kapruwan
Abstract
This work specifically addresses the important issue of recognizing and classifying bone fractures using advanced image analysis techniques. Early identification of bone fractures is essential for proper treatment due to their high occurrence rate and significant influence on patient outcomes. Swift and accurate identification is crucial in order to minimize complications and maximize overall recovery. The study utilizes a dataset consisting of 4906 X-ray images that are classified into two categories: fractured and non-fractured. The system utilizes a MobileNet Convolutional Neural Network (CNN) model. The CNN model achieves a remarkable accuracy of 98% in fracture classification. It was trained on a batch of 4099 pictures, tested on 401 photos, and validated using an additional 406 images. The technology progress showcases the precise capability of deep learning in medical imaging to offer accurate and effective fracture diagnosis and public participation, illustrating the potential of this innovative approach to enhance healthcare outcomes.