Bone Fracture Detection using YOLOv8 and OpenCV

This study introduces a deep learning method for automatically detecting bone fractures with the YOLOv8 object detection model, developed to aid radiologists in accurately and efficiently interpreting X-ray images. The model was developed using a carefully selected dataset of categorized X-ray pictures, utilizing convolutional neural networks (CNNs) to capture important spatial characteristics for fracture detection. The main parts of the experimental setup involved splitting the dataset, fine-tuning model settings, and assessing performance with precision, recall, F1-score, and mean Average Precision (mAP) metrics. The findings showed a high level of accuracy in classifying fractures, with the model successfully identifying and categorizing them while also striking a balance between precision and recall. Sample detections were displayed for qualitative evaluation, and error analysis identified areas for model improvement, specifically in detecting minor fractures. The results show that the YOLOv8 detection model is a promising method for quickly identifying bone fractures, which could improve accuracy in diagnosis and aid in medical imaging decision- making.