Enhancing Fracture Detection in Different Bones Using Deep Learning and YOLO Frameworks

Innovative techniques to improve the diagnosis and treatment of a variety of medical problems have been made possible by the quick developments in medical imaging technologies. Radiologists perform manual examinations as part of traditional procedures for diagnosing bone fractures, which can be subjective and time-consuming. This work introduces a unique method for automating the diagnosis of bone fractures in medical photographs by utilizing Deep Learning (DL) and Machine Learning (ML) approaches. For the purpose of training the model, a large collection of preprocessed X-ray pictures with both fractures and normal bone structures were gathered. The suggested hybrid model combines the YOLO technique for accurate fracture detection, and Convolutional Neural Networks (CNNs) for feature extraction. Fractures of the shoulder, wrist, knee, neck, femur, and hand are among the anatomical areas in which this model accurately detects fractures. The suggested model obtains an accuracy of 79.5%, according to the accuracy-based performance evaluation, indicating its potential to assist radiologists in the precise and effective identification of bone fractures.