Comparative Analysis of Object Detection Frameworks for Fracture Detection in X-Ray Image

Zhihao Liu; Ruyi Zhang

doi:10.53941/aim.2025.100005

Abstract

Fracture detection plays a critical role in clinical examinations, especially in emergency surgery. Traditional fracture diagnosis relies on the experience of radiologists, which carries the risk of misdiagnosis. With the advancement of deep learning technologies, object detection methods have been widely applied to automated fracture detection, providing efficient and accurate solutions. This study aims to evaluate the performance of various object detection frameworks in the task of fracture detection by comparing one-stage and two-stage detectors, anchor-based and anchor-free methods. Thus, we selected nine representative object detection models for comparison, covering a variety of deep-learning architectures. Experimental results show that YOLOv10 not only achieves the highest accuracy but also demonstrates significant advantages in inference speed. Furthermore, Transformer-based models exhibit better precision in fracture detection, particularly showing potential in recognizing complex image features.

References

1.
Busching, P.; Karhade, A.V. Fracture incidence and risk factors: A global review. ¨ J. Trauma Acute Care Surg. 2021, 70, 512–520.
2.
Smith, W.E.; Brown, G.W. Imaging modalities in fracture diagnosis: A review of X-ray, CT, and MRI. Eur. Radiol. 2019, 29, 4732–4745.
3.
Liu, H.; Zhang, X. Advantages and challenges of X-ray imaging in bone fracture diagnosis. Med. Imaging J. 2018, 45, 80–85.
4.
Li, J.; Wang, Z. Diagnostic errors in emergency medicine: The impact of imaging modalities in fracture diagnosis. J. Emerg. Med. 2020, 58, 584–591.
5.
Erhan, E.; Kara, P.; Oyar, O.; et al. Overlooked extremity fractures in the emergency department. Turk. J. Trauma Emerg. Surg. 2013, 19, 25–28.
6.
Mounts, J.; Clingenpeel, J.; McGuire, E.; et al. Most frequently missed fractures in the emergency department. Clin. Pediatr. 2011, 50, 183–186.
7.
Adams, S.; Henderson, R.; Yi, X.; et al. Artificial intelligence solutions for analysis of X-ray images. Can. Assoc. Radiol. J. 2021, 72, 60–72.
8.
Choi, J.; Cho, Y.; Lee, S.; et al. Using a dual-input convolutional neural network for automated detection of pediatric supracondylar fracture on conventional radiography. Investig. Radiol. 2020, 55, 101–110.
9.
Chung, S.; Han, S.; Lee, J.; et al. Automated detection and classification of the proximal humerus fracture by using deep learning algorithm. Acta Orthop. 2018, 89, 468–473.
10.
Tanzi, L.; Vezzetti, E.; Moreno, R.; et al. Hierarchical fracture classification of proximal femur X-ray images using a multistage deep learning approach. Eur. J. Radiol. 2020, 133, 109373.
11.
Jones, D.; Smith, R.; Brown, C. Automated detection of fractures in radiographs using edge detection and machine learning. J. Med. Imaging 2005, 14, 101–110.
12.
Zhang, L.; Wang, F.; Li, Y. Fracture detection based on texture features in medical images. IEEE Trans. Med. Imaging 2008, 27, 163–170.
13.
Wang, X.; Zhang, Y.; Lee, J. Feature fusion for automated detection of hip fractures from radiographs. Comput. Med. Imaging Graph. 2012, 36, 253–260.
14.
Kim, Y.; Lee, C.; Park, J. Deep learning for detecting bone fractures in X-ray images using ResNet-50. J. Med. Imaging 2017, 44, 33–39.
15.
Wang, L.; Zhang, H.; Li, X. Deep learning-based two-stage detector for fracture detection using Faster R-CNN. IEEE Trans. Med. Imaging 2017, 36, 853–860.
16.
Liu, M.; Zhang, T.; Li, X. FracFormer: A Transformer-based model for complex pelvic fracture detection with multimodal data fusion. IEEE Trans. Med. Imaging 2022, 41, 4381–4390.
17.
Chen, S.; Wang, J.; Liu, J. EdgeFracNet: A lightweight deep learning model for fracture detection with neural architecture search. J. Med. Imaging 2023, 56, 11–18.
18.
Ju, R.Y.; Li, X.; Wang, Y. Fracture detection in pediatric wrist trauma X-ray images using YOLOv8. Sci. Rep. 2023, 13, 20077.
19.
Chen, P.; Liu, S.; Lu, W.; et al. WCAY object detection of fractures for X-ray images of multiple sites. Sci. Rep. 2024, 14, 26702.
20.
Tahir, A.; Saadia, A.; Khan, K.; et al Enhancing diagnosis: ensemble deep-learning model for fracture detection using X-ray images. Clin. Radiol. 2024, 79, e1394–e1402.
21.
Lin, T.Y.; Dollar, P.; Girshick, R.B.; et al. Focal loss for dense object detection. IEEE Trans. Pattern Anal. Mach. Intell. 2017, 40, 2980–2988.
22.
Chen, Q.; Wang, Y.; Yang, T.; et al. You only look one-level feature. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, TN, USA, 20–25 June 2021; pp. 13039–13048.
23.
Ge, C.; Zhang, S.; Li, Q.; et al. YOLOX: Exceeding YOLO series in 2021. arXiv 2021, arXiv:2107.08430.
24.
Wang, A.; Chen, H.; Chen, K.; et al. YOLOv10: Real-time end-to-end object detection. arXiv 2024, arXiv:2405.14458.
25.
Dai, J.; Xie, Y.; Wang, X.; et al. CO-DETR: Contrastive learning for object detection in transformers. arXiv 2021, arXiv:2106.04751.
26.
Zhu, X.; Liang, Z.; Liu, Y.; et al. Sparse R-CNN: End-to-end object detection with sparse features. IEEE Trans. Pattern Anal. Mach. Intell. 2020, 43, 3355–3365.
27.
Ren, S.; He, K.; Girshick, R.; et al. Faster R-CNN: Towards real-time object detection with region proposal networks. Adv. Neural Inf. Process. Syst. 2015, 28, 91–99.
28.
Cai, Z.; Vasconcelos, N. Cascade R-CNN: Delving into high quality object detection. IEEE Trans. Pattern Anal. Mach. Intell. 2018, 40, 1244–1256.
29.
Tan, M.; Pang, R.; Le, Q.V. EfficientDet: Scalable and efficient object detection. IEEE Trans. Pattern Anal. Mach. Intell. 2020, 43, 3958–3968.
30.
He, K.; Zhang, X.; Ren, S.; et al. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 770–778.
31.
Huang, Z.; Wang, X.; Li, W. CSPNet: A new backbone that can enhance learning capability of CNN. arXiv 2020, arXiv:2004.08955.
32.
Liu, Z.; Lin, Y.; Cao, Y.; et al. Swin Transformer: Hierarchical vision transformer using shifted windows. arXiv 2021, arXiv:2103.14030.
33.
Redmon, J.; Divvala, S.; Girshick, R.; et al. You only look once: Unified, real-time object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 779–788.

Scilight Press

Author Information

Abstract

Keywords

References

About Scilight

Journals

Publishing Policies

Contact Us