Fracture detection in pediatric wrist trauma X-ray images using YOLOv8 algorithm

Hospital emergency departments frequently receive lots of bone fracture cases, with pediatric wrist trauma fracture accounting for the majority of them. Before pediatric surgeons perform surgery, they need to ask patients how the fracture occurred and analyze the fracture situation by interpreting X-ray images. The interpretation of X-ray images often requires a combination of techniques from radiologists and surgeons, which requires time-consuming specialized training. With the rise of deep learning in the field of computer vision, network models applying for fracture detection has become an important research topic. In this paper, we use data augmentation to improve the model performance of YOLOv8 algorithm (the latest version of You Only Look Once) on a pediatric wrist trauma X-ray dataset (GRAZPEDWRI-DX), which is a public dataset. The experimental results show that our model has reached the state-of-the-art (SOTA) mean average precision (mAP 50). Specifically, mAP 50 of our model is 0.638, which is significantly higher than the 0.634 and 0.636 of the improved YOLOv7 and original YOLOv8 models. To enable surgeons to use our model for fracture detection on pediatric wrist trauma X-ray images, we have designed the application "Fracture Detection Using YOLOv8 App" to assist surgeons in diagnosing fractures, reducing the probability of error analysis, and providing more useful information for surgery.

GRAZPEDWRI-DX [Nagy et al. 2022] is a public dataset of 20,327 pediatric wrist trauma X-ray images released by the University of Medicine of Graz. These X-ray images were collected by multiple pediatric radiologists at the Department for Pediatric Surgery of the University Hospital Graz between 2008 and 2018, involving 6,091 patients and a total of 10,643 studies. This dataset is annotated with 74,459 image labels, featuring a total of 67,771 labeled objects. You can find the original GRAZPEDWRI-DX dataset here (unsplit).

In our study [Ju et al. 2023], we divided the GRAZPEDWRI-DX dataset randomly into three sets: training, validation, and test. The sizes are approximately 70%, 20%, and 10% of the original dataset, respectively. Specifically, our training set consists of 14,204 images (69.88%), validation set of 4,094 images (20.14%), and test set of 2,029 images (9.98%). In subsequent studies [Chien et al. 2024][Chien et al. 2025][Ju et al. 2026], we all followed this split.