Towards Whole-body CT Bone Segmentation

  • André Klein
  • Jan Warszawski
  • Jens Hillengaß
  • Klaus Hermann Maier-Hein
Conference paper
Part of the Informatik aktuell book series (INFORMAT)


Bone segmentation from CT images is a task that has been worked on for decades. It is an important ingredient to several diagnostics or treatment planning approaches and relevant to various diseases. As high-quality manual and semi-automatic bone segmentation is very time-consuming, a reliable and fully automatic approach would be of great interest in many scenarios. In this publication, we propose a UNet inspired architecture to address the task using Deep Learning. We evaluated the approach on whole-body CT scans of patients suffering from multiple myeloma. As the disease decomposes the bone, an accurate segmentation is of utmost importance for the evaluation of bone density, disease staging and localization of focal lesions. The method was evaluated on an in-house data-set of 6000 2D image slices taken from 15 whole-body CT scans, achieving a dice score of 0.96 and an IOU of 0.94.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Krčah M, Székely G, Blanc R; IEEE. Fully automatic and fast segmentation of the femur bone from 3D-CT images with no shape prior. Proc ISBI. 2011; p. 2087–2090.Google Scholar
  2. 2.
    Clarke B. Normal bone anatomy and physiology. Clin J Am Soc Nephrol. 2008;3(Supplement 3):S131–S139.Google Scholar
  3. 3.
    Hillengass J, Delorme S. Multiples Myelom: Aktuelle Empfehlungen für die Bildgebung. Der Radiologe. 2012 Apr;52(4):360–365.Google Scholar
  4. 4.
    Buie HR, Campbell GM, Klinck RJ, et al. Automatic segmentation of cortical and trabecular compartments based on a dual threshold technique for in vivo micro-CT bone analysis. Bone. 2007;41(4):505–515.Google Scholar
  5. 5.
    Pinheiro M, Alves J. A new level-set-based protocol for accurate bone segmentationfFrom CT imaging. IEEE Access. 2015;3:1894–1906.Google Scholar
  6. 6.
    Younes LB, Nakajima Y, Saito T. Fully automatic segmentation of the femur from 3D-CT images using primitive shape recognition and statistical shape models. Int J Comput Assist Rad Surg. 2014;9(2):189–196.Google Scholar
  7. 7.
    Pérez-Carrasco JA, Acha-Piñero B, Serrano C. Segmentation of bone structures in 3D CT images based on continuous max-flow optimization. Proc SPIE. 2015; p. 94133Y.Google Scholar
  8. 8.
    Malladi R, Sethian JA, Vemuri BC. Shape modeling with front propagation: A level set approach. IEEE Trans Pattern Anal Mach Intell. 1995;17(2):158–175.Google Scholar
  9. 9.
    Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. arXiv preprint arXiv:170205747. 2017.
  10. 10.
    Ronneberger O, Fischer P, Brox T; Springer. U-net: Convolutional networks for biomedical image segmentation. Proc MICCAI. 2015; p. 234–241.Google Scholar
  11. 11.
    Nolden M, Zelzer S, Seitel A, et al. The medical imaging interaction toolkit: challenges and advances. Int J Comput Assist Rad Surg. 2013;8(4):607–620.Google Scholar

Copyright information

© Springer-Verlag GmbH Deutschland 2018

Authors and Affiliations

  • André Klein
    • 1
    • 2
  • Jan Warszawski
    • 2
  • Jens Hillengaß
    • 3
  • Klaus Hermann Maier-Hein
    • 1
  1. 1.Division of Medical Image ComputingDeutsches Krebsforschungszentrum (DKFZ)HeidelbergDeutschland
  2. 2.Medical FacultyUniversity of HeidelbergHeidelbergDeutschland
  3. 3.Section Multiple Myeloma, Department of Hematology, Oncology and RheumatologyUniversity of HeidelbergHeidelbergDeutschland

Personalised recommendations