Automatic Detection of Wrist Fractures From Posteroanterior and Lateral Radiographs: A Deep Learning-Based Approach

  • Raja EbsimEmail author
  • Jawad Naqvi
  • Timothy F. Cootes
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11404)


We present a system that uses convolutional neural networks (CNNs) to detect wrist fractures (distal radius fractures) in posterioanterior and lateral radiographs. The proposed system uses random forest regression voting constrained local model to automatically segment the radius. The resulting automatic annotation is used to register the object across the dataset and crop patches. A CNN is trained on the registered patches for each view separately. Our automatic system outperformed existing systems with a performance of 96% (area under receiver operating characteristic curve) on cross-validation experiments on a dataset of 1010 patients, half of them with fractures.


Medical image analysis with deep learning X-ray fracture detection Wrist fracture detection Computer-aided diagnosis 



The research leading to these results has received funding from Libyan Ministry of Higher Education and Research. The authors would like to thank Dr Jonathan Harris, Dr Matthew Davenport, and Dr Martin Smith for their help setting up the project.


  1. 1.
    Court-Brown, C., Caesar, B.: Epidemiology of adult fractures: a review. Injury 37(8), 691–697 (2006). Scholar
  2. 2.
    Goldfarb, C., Yin, Y., Gilula, L., Fisher, A., Boyer, M.: Wrist fractures: what the clinician wants to know. Radiology 219(1), 11–28 (2001). Scholar
  3. 3.
    Guly, H.: Injuries initially misdiagnosed as sprained wrist (beware the sprained wrist). Emerg. Med. J. 19(1), 41–42 (2002). Scholar
  4. 4.
    Petinaux, B., Bhat, R., Boniface, K., Aristizabal, J.: Accuracy of radiographic readings in the emergency department. Am. J. Emerg. Med. 29(1), 18–25 (2011). Scholar
  5. 5.
    Lindner, C., Thiagarajah, S., Wilkinson, J., Wallis, G., Cootes, T., The arcOGEN Consortium: Fully automatic segmentation of the proximal femur using random forest regression voting. IEEE Trans. Med. Imaging 32(8), 1462–1472 (2013). Scholar
  6. 6.
    Cootes, T.F., Ionita, M.C., Lindner, C., Sauer, P.: Robust and accurate shape model fitting using random forest regression voting. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012. LNCS, vol. 7578, pp. 278–291. Springer, Heidelberg (2012). Scholar
  7. 7.
    Kaufman, J., et al.: A neural network approach for bone fracture healing assessment. IEEE Eng. Med. Biol. Mag. 9(3), 23–30 (1990). Scholar
  8. 8.
    Ryder, D., King, S., Oliff, C., Davies, E.: A possible method of monitoring bone fracture and bone characteristics using a noninvasive acoustic technique. In: Proceedings of International Conference on Acoustic Sensing and Imaging, pp. 159–163. IEEE (1993)Google Scholar
  9. 9.
    Singh, V., Chauhan, S.: Early detection of fracture healing of a long bone for better mass health care. In: Proceedings of 20th International Conference of the Engineering in Medicine and Biology Society – EMBC 1998, vol. 6, pp. 2911–2912. IEEE (1998).
  10. 10.
    Tian, T.P., Chen, Y., Leow, W.K., Hsu, W., Howe, T.S., Png, M.A.: Computing neck-shaft angle of femur for X-ray fracture detection. In: Petkov, N., Westenberg, M.A. (eds.) CAIP 2003. LNCS, vol. 2756, pp. 82–89. Springer, Heidelberg (2003). Scholar
  11. 11.
    Jia, Y., Jiang, Y.: Active contour model with shape constraints for bone fracture detection. In: Proceedings of 3rd International Conference on Computer Graphics, Imaging and Visualisation – CGIV 2006, pp. 90–95. IEEE (2006).
  12. 12.
    Yap, D.H., Chen, Y., Leow, W., Howe, T., Png, M.: Detecting femur fractures by texture analysis of trabeculae. In: Proceedings of 17th International Conference on Pattern Recognition – ICPR 2004, vol. 3, pp. 730–733. IEEE (2004).
  13. 13.
    Lim, S., Xing, Y., Chen, Y., Leow, W., Howe, T., Png, M.: Detection of femur and radius fractures in X-ray images. In: Proceedings of 2nd International Conference on Advances in Medical Signal and Information Processing, pp. 249–256 (2004)Google Scholar
  14. 14.
    Lum, V., Leow, W., Chen, Y., Howe, T., Png, M.: Combining classifiers for bone fracture detection in X-ray images. In: Proceedings of International Conference on Image Processing – ICIP 2005, p. I-1149. IEEE (2005).
  15. 15.
    Bayram, F., Çakiroğlu, M.: DIFFRACT: DIaphyseal Femur FRActure Classifier SysTem. Biocybern. Biomed. Eng. 36(1), 157–171 (2016). Scholar
  16. 16.
    Roberts, M., Oh, T., Pacheco, E., Mohankumar, R., Cootes, T., Adams, J.: Semi-automatic determination of detailed vertebral shape from lumbar radiographs using active appearance models. Osteoporos. Int. 23(2), 655–664 (2012). Scholar
  17. 17.
    Cao, Y., Wang, H., Moradi, M., Prasanna, P., Syeda-Mahmood, T.: Fracture detection in X-ray images through stacked random forests feature fusion. In: Proceedings of 12th International Symposium on Biomedical Imaging – ISBI 2015, pp. 801–805. IEEE (2015).
  18. 18.
    Cootes, T., Edwards, G., Taylor, C.: Active appearance models. IEEE Trans. Pattern Anal. Mach. Intell. 23(6), 681–685 (2001). Scholar
  19. 19.
    Ebsim, R., Naqvi, J., Cootes, T.: Detection of wrist fractures in X-ray images. In: Shekhar, R., et al. (eds.) CLIP 2016. LNCS, vol. 9958, pp. 1–8. Springer, Cham (2016). Scholar
  20. 20.
    Ebsim, R., Naqvi, J., Cootes, T.: Fully automatic detection of distal radius fractures from posteroanterior and lateral radiographs. In: Cardoso, M.J., et al. (eds.) CARE/CLIP - 2017. LNCS, vol. 10550, pp. 91–98. Springer, Cham (2017). Scholar
  21. 21.
    Lindner, C., Bromiley, P., Ionita, M., Cootes, T.: Robust and accurate shape model matching using random forest regression-voting. IEEE Trans. Pattern Anal. Mach. Intell. 37(9), 1862–1874 (2015). Scholar
  22. 22.
    Kim, D., MacKinnon, T.: Artificial intelligence in fracture detection: transfer learning from deep convolutional neural networks. Clin. Radiol. 73(5), 439–445 (2018). Scholar
  23. 23.
    Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., Wojna, Z.: Rethinking the inception architecture for computer vision. In: Proceedings of 29th IEEE Conference on Computer Vision and Pattern Recognition – CVPR 2016, pp. 2818–2826. IEEE (2016).
  24. 24.
    Russakovsky, O., et al.: ImageNet large scale visual recognition challenge. Int. J. Comput. Vis. 115(3), 211–252 (2015). Scholar
  25. 25.
    Olczak, J., et al.: Artificial intelligence for analyzing orthopedic trauma radiographs: deep learning algorithms—are they on par with humans for diagnosing fractures? Acta Orthop. 88(6), 581–586 (2017). Scholar
  26. 26.
    Jia, Y., et al.: Caffe: convolutional architecture for fast feature embedding. In: Proceedings of 22nd International Conference on Multimedia – MM 2014, pp. 675–678. ACM (2014).
  27. 27.
    Zhong, S., Li, K., Feng, R.: Deep convolutional hamming ranking network for large scale image retrieval. In: Proceedings of 11th World Congress on Intelligent Control and Automation – WCICA 2014, pp. 1018–1023. IEEE (2014).
  28. 28.
    Rajpurkar, P., et al.: MURA large dataset for abnormality detection in musculoskeletal radiographs. arXiv:1712.06957v1 (2017)
  29. 29.
    Glorot, X., Bengio, Y.: Understanding the difficulty of training deep feedforward neural networks. In: Teh, Y., Titterington, M. (eds.) Proceedings of 13th International Conference on Artificial Intelligence and Statistics – AISTATS 2010, PMLR, vol. 9, pp. 249–256. PMLR (2010)Google Scholar
  30. 30.
    Kingma, D., Ba, J.: Adam: a method for stochastic optimization. arXiv:1412.6980 (2014)

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Centre for Imaging SciencesThe University of ManchesterManchesterUK
  2. 2.Health Education North West School of RadiologyManchesterUK

Personalised recommendations