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

TL;DR

This paper enhances YOLOv8 for pediatric wrist fracture detection in X-ray images by integrating feature context modules, resulting in improved accuracy and efficiency over existing models.

Contribution

Introduces four variants of FCE-YOLOv8 with different modules, achieving state-of-the-art performance in pediatric wrist fracture detection.

Findings

YOLOv8+GC-M3 improves mAP@50 to 66.32%

YOLOv8+SE-M3 achieves mAP@50 of 67.07%

Model reduces inference time compared to SOTA

Abstract

Children often suffer wrist trauma in daily life, while they usually need radiologists to analyze and interpret X-ray images before surgical treatment by surgeons. The development of deep learning has enabled neural networks to serve as computer-assisted diagnosis (CAD) tools to help doctors and experts in medical image diagnostics. Since YOLOv8 model has obtained the satisfactory success in object detection tasks, it has been applied to various fracture detection. This work introduces four variants of Feature Contexts Excitation-YOLOv8 (FCE-YOLOv8) model, each incorporating a different FCE module (i.e., modules of Squeeze-and-Excitation (SE), Global Context (GC), Gather-Excite (GE), and Gaussian Context Transformer (GCT)) to enhance the model performance. Experimental results on GRAZPEDWRI-DX dataset demonstrate that our proposed YOLOv8+GC-M3 model improves the mAP@50 value from 65.78% to 66.32%, outperforming the state-of-the-art (SOTA) model while reducing inference time. Furthermore, our proposed YOLOv8+SE-M3 model achieves the highest mAP@50 value of 67.07%, exceeding the SOTA performance. The implementation of this work is available at https://github.com/RuiyangJu/FCE-YOLOv8.