Density Invariant Detection of Osteoporosis Using Growing Neural Gas

  • Igor T. Podolak
  • Stanisław K. Jastrzębski
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 226)


We present a method for osteoporosis detection using graph representations obtained running a Growing Neural Gas machine learning algorithm on X–ray bone images. The GNG induced graph, being dependent on density, represents well the features which may be in part responsible for the illness. The graph connects well dense bone regions, making it possible to subdivide the whole image into regions. It is interesting to note, that these regions in bones, whose extraction might make it easier to detect the illness, correspond to some graph theoretic notions. In the paper, some invariants based on these graph theoretic notions, are proposed and if used with a machine classification method, e.g. a neural network, will make it possible to help recognize images of bones of ill persons. This graph theoretic approach is novel in this area. It helps to separate solution from the actual physical properties. The paper gives the proposed indices definitions and shows a classification based on them as input attributes.


Healthy Bone Ball Density Edge Betweenness Grow Cell Structure Igraph Package 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Dlotko, P.: Private Communications (2012)Google Scholar
  2. 2.
    Efron, B.: Bootstrap methods: another look at the jackknife. Annals of Statistics 7, 1–26 (1979)MathSciNetzbMATHCrossRefGoogle Scholar
  3. 3.
    Fiser, D., Feigl, J., Kulich, M.: Growing Neural Gas Efficiently. Neurocomputing (2012) (in press)Google Scholar
  4. 4.
    Fritzke, B.: Kohonen Feature Maps and Growing Cell Structures a Performance Comparison. Advances in Neural Information Processing Systems 5, 115–122 (1993)Google Scholar
  5. 5.
    Fritzke, B.: A Growing Neural Gas network learns topologies. Advances in Neural Information Processing Systems 7, 625–632 (1995)Google Scholar
  6. 6.
    Fritzke, B.: A self-organizing network that can follow non-stationary distributions. In: International Conference on Artificial Neural Networks, Lausanne, pp. 613–618 (1997)Google Scholar
  7. 7.
    Garrahan, N.J., Mellish, R.W.E., Compston, J.E.: A new method for the two-dimensional analysis of bone structure in human iliac crest biopsies. J. Microsc. 142, 341–349 (1986)CrossRefGoogle Scholar
  8. 8.
    Hahn, M., Vogel, M., Pompesius–Kempa, M., Delling, G.: Trabecular bone pattern factor – a new parameter for simple quantification of bone micro architecture. Bone 13, 327–330 (1992)CrossRefGoogle Scholar
  9. 9.
    R: A language and environment for statistical computing. R Foundation for Statistical Computing (2005)Google Scholar
  10. 10.
    Saha, P.K., Wehrli, F.W.: A robust method for measuring trabecular bone orientation anisotropy at in vivo resolution using tensor scale. Pattern Recognition 37, 1935–1944 (2004)CrossRefGoogle Scholar
  11. 11.
    Tabor, Z., Rokita, E.: Quantifying deterioration of bone tissue from grey-level images. Md. Eng. Phys. 29, 497–504 (2007)CrossRefGoogle Scholar
  12. 12.
    Tabor, Z.: Detecting surfaces of minimal cut – a graph theoretical approach. Inżynieria Materiałowa 4, 439–442 (2008)Google Scholar
  13. 13.
    World Health Organization, Assessment of fracture risk and its application to screening for post-menopausal osteoporosis, Technical Report Series 843 (1994)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2013

Authors and Affiliations

  1. 1.Institute of Computer ScienceJagiellonian UniversityKrakówPoland

Personalised recommendations