Abstract:Diagnosing and treating abnormalities in the wrist, specifically distal radius, and ulna fractures, is a crucial concern among children, adolescents, and young adults, with a higher incidence rate during puberty. However, the scarcity of radiologists and the lack of specialized training among medical professionals pose a significant risk to patient care. This problem is further exacerbated by the rising number of imaging studies and limited access to specialist reporting in certain regions. This highlights the need for innovative solutions to improve the diagnosis and treatment of wrist abnormalities. Automated wrist fracture detection using object detection has shown potential, but current studies mainly use two-stage detection methods with limited evidence for single-stage effectiveness. This study employs state-of-the-art single-stage deep neural network-based detection models YOLOv5, YOLOv6, YOLOv7, and YOLOv8 to detect wrist abnormalities. Through extensive experimentation, we found that these YOLO models outperform the commonly used two-stage detection algorithm, Faster R-CNN, in fracture detection. Additionally, compound-scaled variants of each YOLO model were compared, with YOLOv8m demonstrating a highest fracture detection sensitivity of 0.92 and mean average precision (mAP) of 0.95. On the other hand, YOLOv6m achieved the highest sensitivity across all classes at 0.83. Meanwhile, YOLOv8x recorded the highest mAP of 0.77 for all classes on the GRAZPEDWRI-DX pediatric wrist dataset, highlighting the potential of single-stage models for enhancing pediatric wrist imaging.

What problem does this paper attempt to address?

The paper primarily addresses the diagnostic issues of common wrist abnormalities (particularly distal radius and ulna fractures) in children, adolescents, and young adults. It points out that wrist fractures are especially prevalent during adolescence, and timely and accurate assessment and treatment are crucial to prevent long-term adverse outcomes. However, the shortage of professional radiologists and the lack of specialized training among medical professionals in evaluating these injuries have led to an increased risk of diagnostic errors. Specifically, the error rate in emergency X-ray readings can be as high as 26%. To address these issues, the paper employs different versions of the advanced single-stage object detection model YOLO (YOLOv5, YOLOv6, YOLOv7, and YOLOv8) and conducts experiments on a large pediatric wrist X-ray image dataset named GRAZPEDWRI-DX. These models aim to improve the detection accuracy of wrist abnormalities, including fractures. Specifically, the research objectives include: 1. Testing the effectiveness of the latest YOLO detection models in identifying near-wrist fractures in pediatric X-ray images. 2. Comparing the performance of single-stage detection models with the widely used two-stage detection methods (such as Faster R-CNN) in wrist image analysis. 3. Investigating whether the compound scaled variants within each YOLO algorithm can enhance their performance in fracture detection. 4. Exploring whether YOLO models surpass traditional Convolutional Neural Network (CNN) architectures and DenseNets in terms of sensitivity in fracture recognition. The main contributions of the research include a thorough performance evaluation of the latest YOLO detection models on the newly released GRAZPEDWRI-DX dataset and an in-depth comparison between different variants within each YOLO model. Additionally, the paper reports achieving state-of-the-art mean Average Precision (mAP) scores on this dataset. Through these experiments, the paper aims to verify whether YOLO models and their compound scaled variants can effectively improve the detection efficiency of fractures in pediatric wrist X-ray images.

Enhancing Wrist Fracture Detection with YOLO

Pediatric Wrist Fracture Detection in X-rays via YOLOv10 Algorithm and Dual Label Assignment System

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

YOLOv8-ResCBAM: YOLOv8 Based on An Effective Attention Module for Pediatric Wrist Fracture Detection

YOLOv9 for fracture detection in pediatric wrist trauma X‐ray images

YOLOv8-AM: YOLOv8 Based on Effective Attention Mechanisms for Pediatric Wrist Fracture Detection

YOLOv9 for Fracture Detection in Pediatric Wrist Trauma X-ray Images

Learning from the few: Fine-grained approach to pediatric wrist pathology recognition on a limited dataset

Pediatric Wrist Fracture Detection Using Feature Context Excitation Modules in X-ray Images

Deep Learning Approach for Automatic Wrist Fracture Detection Using Ultrasound Bone Probability Maps

Critical evaluation of deep neural networks for wrist fracture detection

Detecting pediatric wrist fractures using deep-learning-based object detection

Detection of Spinal Fracture Lesions Based on Improved Yolov3

Detection of whole body bone fractures based on improved YOLOv7

An AI-Driven Framework for Detecting Bone Fractures in Orthopedic Therapy

Fracture Detection in Wrist X-ray Images Using Deep Learning-Based Object Detection Models

Lightweight G-YOLOv11: Advancing Efficient Fracture Detection in Pediatric Wrist X-rays

Global Context Modeling in YOLOv8 for Pediatric Wrist Fracture Detection

Artificial intelligence to identify fractures on pediatric and young adult upper extremity radiographs

An Integrated Approach using YOLOv8 and ResNet, SeResNet & Vision Transformer (ViT) Algorithms based on ROI Fracture Prediction in X-ray Images of the Elbow