Automatic segmentation of vertebrae in 3D CT images using adaptive fast 3D pulse coupled neural networks

  • Mina Zareie
  • Hossein ParsaeiEmail author
  • Saba Amiri
  • Malik Shahzad Awan
  • Mohsen Ghofrani
Scientific Paper


Two systems are presented for segmentation of vertebrae in a 3D computed tomography (CT) image. The first method extracts seven features from each voxel and uses a multi-layer perceptron neural network (MLPNN) to classify the voxel as vertebrae or background. In the second method, the segmentation is completed in two steps: first, a newly developed adaptive pulse coupled neural network (APCNN) directly applied to a given image segments vertebrae, then the result is refined using a median filter. In the developed APCNN, the values for the user-defined parameters of the pulse coupled neural networks (PCNN) are adaptively adjusted for each image individually, instead of using one value for all images as in conventional PCNN. The performance of both systems in terms of Dice index (DI) was evaluated and compared against the state-of-the-art segmentation methods using seventeen clinical and standard CT images. Overall, both systems demonstrated statistically similar and promising performance with average DI > 95%. Compared to existing PCNN-based segmentation algorithms, the accuracy of the proposed APCNN improved by 29.3% on average. The developed APCNN-based system is more accurate than MLPNN-based system and existing PCNN-based algorithms in segmentation of vertebrae with blurred and weak boundaries and in the images contaminated by salt- and- pepper noise. In terms of computation time, the APCNN-based system is 16 times faster than the MLPNN-based system. Consequently, the presented APCNN-based algorithm is both accurate and fast and could be used in clinical environment for segmentation of vertebrae in 3D CT images.


Feature extraction Multi-layer perceptron Pulse-coupled neural networks Vertebrae segmentation 



We would also like to thank the staffs of Medical Imaging center of Namazi hospital and Taba medical imaging center, Shiraz, Iran and Spine CT image site [36] for sharing the medical images used in this research and assisting with analysis of this images. The authors wish to thank Mr. H. Argasi at the Research Consultation Center (RCC) of Shiraz University of Medical Sciences for his invaluable assistance in editing this manuscript.


This study was funded by Shiraz University of Medical Sciences (Grant # 93-01-01-8772).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

For the part of this study that we used data (image) of patients, all the procedures performed in this work involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study whom their 3D vertebrae CT images acquired in Medical Imaging center of Namazi hospital and Taba medical imaging center, Shiraz, Iran.


  1. 1.
    Roberts MG, Cootes TF, Adams JE Vertebral shape: automatic measurement with dynamically sequenced active appearance models. In: International conference on medical image computing and computer-assisted intervention. Springer, p. 733–740Google Scholar
  2. 2.
    Gupta G, Kaur T (2012) Denoising of computed tomography images using curvelet transformation with log Gabor filter. Int J Sci Eng Res 3:1000–1003Google Scholar
  3. 3.
    Mastmeyer A, Engelke K, Fuchs C, Kalender WA (2006) A hierarchical 3D segmentation method and the definition of vertebral body coordinate systems for QCT of the lumbar spine. Med Image Anal 10:560–577CrossRefGoogle Scholar
  4. 4.
    Suzani A, Rasoulian A, Fels S, Rohling RN, Abolmaesumi P (2014) Semi-automatic segmentation of vertebral bodies in volumetric MR images using a statistical shape + pose model. SPIE Med Imaging 2014:90360Google Scholar
  5. 5.
    Kim Y, Kim D (2009) A fully automatic vertebra segmentation method using 3D deformable fences. Comput Med Imaging Graph Elsevier 33:343–352CrossRefGoogle Scholar
  6. 6.
    Lim PH, Bagci U, Bai L (2013) Introducing Willmore flow into level set segmentation of spinal vertebrae. IEEE Trans Biomed Eng 60:115–122CrossRefGoogle Scholar
  7. 7.
    Huang J, Jian F, Wu H, Li H (2013) An improved level set method for vertebra CT image segmentation. Biomed Eng Online BioMed Central Ltd 12:48CrossRefGoogle Scholar
  8. 8.
    Mirzaalian H, Wels M, Heimann T, Kelm BM, Suehling M (2013) Fast and robust 3D vertebra segmentation using statistical shape models. In: Engineering in Medicine and Biology Society (EMBC), 2013 35th annual international conference of the IEEE. IEEE, pp 3379–3382Google Scholar
  9. 9.
    Ibragimov B, Likar B, Pernuš F, Vrtovec T (2014) Shape representation for efficient landmark-based segmentation in 3-d. IEEE Trans Med Imaging 33:861–874CrossRefGoogle Scholar
  10. 10.
    Yao J, Burns JE, Forsberg D, Seitel A, Rasoulian A, Abolmaesumi P et al (2016) A multi-center milestone study of clinical vertebral CT segmentation. Comput Med Imaging Graph 49:16–28CrossRefGoogle Scholar
  11. 11.
    Li Y, Liang W, Tan J, Zhang Y (2015) A novel automatically initialized level set approach based on region correlation for lumbar vertebrae CT image segmentation. In 2015 IEEE international symposium on medical measurements and applications (MeMeA), IEEE, pp 291–296Google Scholar
  12. 12.
    Klinder T, Ostermann J, Ehm M, Franz A, Kneser R, Lorenz C (2009) Automated model-based vertebra detection, identification, and segmentation in CT images. Med Image Anal 13:471–482CrossRefGoogle Scholar
  13. 13.
    Haykin SO (2008) Neural networks and learning machines, 3 edn. Pearson, New YorkGoogle Scholar
  14. 14.
    Lindblad T, Kinser JM, Lindblad T, Kinser JM (2005) Image processing using pulse-coupled neural networks. Springer, BerlinGoogle Scholar
  15. 15.
    Eckhorn R, Reitboeck HJ, Arndt M, Dicke P (1990) Feature linking via synchronization among distributed assemblies: simulations of results from cat visual cortex. Neural Comput 2:293–307CrossRefGoogle Scholar
  16. 16.
    Johnson JL, Padgett M, Lou (1998) PCNN models and applications. IEEE Trans Neural Netw 10:480–498CrossRefGoogle Scholar
  17. 17.
    Monica Subashini M, Sahoo SK (2014) Pulse coupled neural networks and its applications. Expert Syst Appl 41:3965–3974CrossRefGoogle Scholar
  18. 18.
    Wang Z, Ma Y, Cheng F, Yang L (2010) Review of pulse-coupled neural networks. Image Vis Comput 28:5–13CrossRefGoogle Scholar
  19. 19.
    Zhan K, Shi J, Wang H, Xie Y, Li Q (2017) Computational mechanisms of pulse-coupled neural networks: a comprehensive review. Arch Comput Methods Eng 24:573–588CrossRefGoogle Scholar
  20. 20.
    Chang Q, Shi J, Xiao Z (2009) A new 3D segmentation algorithm based on 3D PCNN for lung CT slices. In: BMEI. IEEE, pp 1–5Google Scholar
  21. 21.
    Murugavel M, Sullivan JM (2009) Automatic cropping of MRI rat brain volumes using pulse coupled neural networks. NeuroImage 45:845–854CrossRefGoogle Scholar
  22. 22.
    Wei S, Hong Q, Hou M (2011) Automatic image segmentation based on PCNN with adaptive threshold time constant. Neurocomputing 74:1485–1491CrossRefGoogle Scholar
  23. 23.
    Hassanien AE, Kim T (2012) Breast cancer MRI diagnosis approach using support vector machine and pulse coupled neural networks. J Appl Log 10:277–284CrossRefGoogle Scholar
  24. 24.
    Fu JC, Chen CC, Chai JW, Wong STC, Li IC (2010) Image segmentation by EM-based adaptive pulse coupled neural networks in brain magnetic resonance imaging. Comput Med Imaging Graph 34:308–320CrossRefGoogle Scholar
  25. 25.
    Lindblad T, Kinser JM (2013) The PCNN and ICM. In: Lindblad T, Kinser JM (eds) Image process using pulse-coupled neural networks. Applications in python. Springer, Berlin, pp 57–86CrossRefGoogle Scholar
  26. 26.
    Jin X, Zhou D, Yao S, Nie R, Yu C, Ding T (2016) Remote sensing image fusion method in CIELab color space using nonsubsampled shearlet transform and pulse coupled neural networks. J Appl Remote Sens 10:025023CrossRefGoogle Scholar
  27. 27.
    Yi-de M, Qing L, Zhi-bai Q (2004) Automated image segmentation using improved PCNN model based on cross-entropy. In: Proceedings of 2004 international symposium on intelligent multimedia, video and speech processing, 2004. IEEE, pp 743–746Google Scholar
  28. 28.
    Amiri S, Movahedi MM, Kazemi K, Parsaei H (2017) 3D cerebral MR image segmentation using multiple-classifier system. Med Biol Eng Comput 55:353–364CrossRefGoogle Scholar
  29. 29.
    Patel SP, Lee JJ, Hecht GG, Holcombe SA, Wang SC, Goulet JA (2016) Normative vertebral hounsfield unit values and correlation with bone mineral density. J Clin Exp Orthop 2:14CrossRefGoogle Scholar
  30. 30.
    Tuceryan M, Jain AK (1993) Texture analysis. Handb Pattern Recognit Comput Vis 2:207–248Google Scholar
  31. 31.
    Baraldi A, Parmiggiani F (1995) An investigation of the textural characteristics associated with gray level cooccurrence matrix statistical parameters. Geosci Remote Sens 33:283–304Google Scholar
  32. 32.
    Miller AS, Blott BH (1992) Review of neural network applications in medical imaging and signal processing. Med Biol Eng Comput 30:449–464CrossRefGoogle Scholar
  33. 33.
    Yu X, Efe MO, Kaynak O (2002) A general backpropagation algorithm for feedforward neural networks learning. Neural Netw IEEE Trans IEEE 13:251–254CrossRefGoogle Scholar
  34. 34.
    Chang Q, Shi J, Xiao Z (2009) A new 3D segmentation algorithm based on 3D PCNN for lung CT slices. In: Biomedical engineering and informatics 2009. BMEI09 2nd international conference. IEEE, pp 1–5Google Scholar
  35. 35.
    Lanzara RG (1994) Weber’s law modeled by the mathematical description of a beam balance. Math Biosci 122:89–94CrossRefGoogle Scholar
  36. 36.
  37. 37.
    Dice LR (1945) Measures of the amount of ecologic association between species. Ecology JSTOR 26:297–302CrossRefGoogle Scholar
  38. 38.
    Parsaei H, Stashuk D (2013) EMG Signal Decomposition Using Motor Unit Potential Train Validity. IEEE Trans Neural Syst Rehabil 21:265–274CrossRefGoogle Scholar
  39. 39.
    Parsaei H, Stashuk DW (2011) Adaptive motor unit potential train validation using MUP shape information. Med Eng Phys 33:581–589CrossRefGoogle Scholar

Copyright information

© Australasian College of Physical Scientists and Engineers in Medicine 2018

Authors and Affiliations

  1. 1.Department of Medical Physics and EngineeringShiraz University of Medical SciencesShirazIran
  2. 2.Shiraz Neuroscience Research CenterShiraz University of Medical SciencesShirazIran
  3. 3.School of Computing, Electronics and MathematicsPlymouth UniversityPlymouthUK

Personalised recommendations