Region-Aware Deep Localization Framework for Cervical Vertebrae in X-Ray Images

  • S. M. Masudur Rahman Al Arif
  • Karen Knapp
  • Greg Slabaugh
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10553)


The cervical spine is a flexible anatomy and vulnerable to injury, which may go unnoticed during a radiological exam. Towards building an automatic injury detection system, we propose a localization framework for the cervical spine in X-ray images. The proposed framework employs a segmentation approach to solve the localization problem. As the cervical spine is a single connected component, we introduce a novel region-aware loss function for training a deep segmentation network that penalises disjoint predictions. Using data augmentation, the framework has been trained on a dataset of 124 images and tested on another 124 images, all collected from real life medical emergency rooms. The results show a significant improvement in performance over the previous state-of-the-art cervical vertebrae localization framework.


Cervical spine X-ray Localization FCN Region-aware 


  1. 1.
    Platzer, P., Hauswirth, N., Jaindl, M., Chatwani, S., Vecsei, V., Gaebler, C.: Delayed or missed diagnosis of cervical spine injuries. J. Trauma Acute Care Surg. 61(1), 150–155 (2006)CrossRefGoogle Scholar
  2. 2.
    Morris, C., McCoy, E.: Clearing the cervical spine in unconscious polytrauma victims, balancing risks and effective screening. Anaesthesia 59(5), 464–482 (2004)CrossRefGoogle Scholar
  3. 3.
    Tezmol, A., Sari-Sarraf, H., Mitra, S., Long, R., Gururajan, A.: Customized Hough transform for Robust segmentation of cervical vertebrae from X-ray images. In: Proceedings of Fifth IEEE Southwest Symposium on Image Analysis and Interpretation, pp. 224–228. IEEE (2002)Google Scholar
  4. 4.
    Larhmam, M.A., Mahmoudi, S., Benjelloun, M.: Semi-automatic detection of cervical vertebrae in X-ray images using generalized Hough transform. In: 2012 3rd International Conference on Image Processing Theory, Tools and Applications (IPTA), pp. 396–401. IEEE (2012)Google Scholar
  5. 5.
    Glocker, B., Feulner, J., Criminisi, A., Haynor, D.R., Konukoglu, E.: Automatic localization and identification of vertebrae in arbitrary field-of-view CT scans. In: Ayache, N., Delingette, H., Golland, P., Mori, K. (eds.) MICCAI 2012. LNCS, vol. 7512, pp. 590–598. Springer, Heidelberg (2012). doi: 10.1007/978-3-642-33454-2_73 CrossRefGoogle Scholar
  6. 6.
    Glocker, B., Zikic, D., Konukoglu, E., Haynor, D.R., Criminisi, A.: Vertebrae localization in pathological spine CT via dense classification from sparse annotations. In: Mori, K., Sakuma, I., Sato, Y., Barillot, C., Navab, N. (eds.) MICCAI 2013. LNCS, vol. 8150, pp. 262–270. Springer, Heidelberg (2013). doi: 10.1007/978-3-642-40763-5_33 CrossRefGoogle Scholar
  7. 7.
    Al Arif, S.M.M.R., Gundry, M., Knapp, K., Slabaugh, G.: Global localization and orientation of the cervical spine in X-ray images. In: Yao, J., Vrtovec, T., Zheng, G., Frangi, A., Glocker, B., Li, S. (eds.) CSI 2016. LNCS, vol. 10182, pp. 3–15. Springer, Cham (2016). doi: 10.1007/978-3-319-55050-3_1 CrossRefGoogle Scholar
  8. 8.
    Shelhamer, E., Long, J., Darrell, T.: Fully convolutional networks for semantic segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 39(4), 640–651 (2016)Google Scholar
  9. 9.
    Noh, H., Hong, S., Han, B.: Learning deconvolution network for semantic segmentation. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1520–1528 (2015)Google Scholar
  10. 10.
    Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). doi: 10.1007/978-3-319-24574-4_28 CrossRefGoogle Scholar
  11. 11.
    Ruder, S.: An overview of gradient descent optimization algorithms, arXiv preprint arXiv:1609.04747 (2016)
  12. 12.
    NHANES-II Dataset. Accessed 19 Feb 2017

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • S. M. Masudur Rahman Al Arif
    • 1
  • Karen Knapp
    • 2
  • Greg Slabaugh
    • 1
  1. 1.City, University of LondonLondonEngland
  2. 2.University of ExeterExeterEngland

Personalised recommendations