Stimulus Related Data Analysis by Structured Neural Networks

  • Bernd Brückner
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4234)


In the analysis of biological data artificial neural networks are a useful alternative to conventional statistical methods. Because of its advantage in analyzing time courses the Multilevel Hypermap Architecture (MHA) is used for analysis of stimulus related data, exemplified by fMRI studies with auditory stimuli. Results from investigations with the MHA show an improvement of discrimination in comparison to statistical methods. With an interface to the well known BrainVoyager software and with a GUI for MATLAB an easy usability of the MHA and a good visualization of the results is possible.

The MHA is an extension of the Hypermap introduced by Kohonen. By means of the MHA it is possible to analyze structured or hierarchical data (data with priorities, data with context, time series, data with varying exactness), which is difficult or impossible to do with known self-organizing maps so far.


Input Vector Auditory Cortex fMRI Data Learn Vector Quantization Hierarchical Data 
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.
    Smith, S.M.: Overview of fmri analysis. Br. J. Radiol. 77(Spec No 2), S167–S175 (2004)CrossRefGoogle Scholar
  2. 2.
    Ngan, S.C., Yacoub, E.S., Auffermann, W.F., Hu, X.: Node merging in Kohonen’s self-organizing mapping of fMRI data. Artif. Intell. Med. 25(1), 19–33 (2002)CrossRefGoogle Scholar
  3. 3.
    Ylipaavalniemi, J., Vigario, R.: Analysis of auditory fmri recordings via ica: A study on consistency. In: Proceedings of the 2004 International Joint Conference on Neural Networks (IJCNN 2004), Budapest, Hungary, vol. 1, pp. 249–254 (2004)Google Scholar
  4. 4.
    Meyer-Baese, A., Wismueller, A., Lange, O.: Comparison of two exploratory data analysis methods for fMRI: unsupervised clustering versus independent component analysis. IEEE Trans. Inf. Technol. Biomed. 8(3), 387–398 (2004)CrossRefGoogle Scholar
  5. 5.
    Voultsidou, M., Dodel, S., Herrmann, J.M.: Neural networks approach to clustering of activity in fMRI data. IEEE Trans. Med. Imaging 24(8), 987–996 (2005)CrossRefGoogle Scholar
  6. 6.
    Brückner, B., Franz, M., Richter, A.: A modified hypermap architecture for classification of biological signals. In: Aleksander, I., Taylor, J. (eds.) Artificial Neural Networks, 2, pp. 1167–1170. Elsevier Science Publishers, Amsterdam (1992)Google Scholar
  7. 7.
    Brückner, B., Wesarg, T.: Modeling speech processing and recognition in the auditory system using the multilevel hypermap architecture. In: Seiffert, U., Jain, L. (eds.) Self-Organizing Neural Networks. Recent Advances and Applications. Springer Series on Studies in Fuzziness and Soft Computing, vol. 78, pp. 145–164. Springer, Heidelberg (2001)Google Scholar
  8. 8.
    Kohonen, T.: The hypermap architecture. In: Kohonen, T., Mäkisara, K., Simula, O., Kangas, J. (eds.) Artificial Neural Networks, pp. 1357–1360. Elsevier, Helsinki (1991)Google Scholar
  9. 9.
    Kohonen, T.: What generalizations of the self-organizing map make sense? In: Marinaro, M., Morasso, P. (eds.) ICANN 1994, pp. 292–297. Springer, London (1994)Google Scholar
  10. 10.
    Brückner, B., Wesarg, T., Blumenstein, C.: Improvements of the modified hypermap architecture for speech recognition. In: Proc. Inter. Conf. Neural Networks, Perth, Australia, vol. 5, pp. 2891–2895 (1995)Google Scholar
  11. 11.
    Brückner, B.: Improvements in the analysis of structured data with the multilevel hypermap architecture. In: K., et al. (eds.) Progress in Connectionist-Based Information Systems, Proceedings of the ICONIP 1997, vol. 1, pp. 342–345. Springer, Singapore (1997)Google Scholar
  12. 12.
    Brückner, B., Gaschler-Markefski, B., Hofmeister, H., Scheich, H.: Detection of non-stationarities in functional mri data sets using the multilevel hypermap architecture. In: Proceedings of the IJCNN 1999, Washington D (1999)Google Scholar
  13. 13.
  14. 14.
    Kohonen, T.: Self-Organization and Associative Memory. Springer, New York (1988)MATHGoogle Scholar
  15. 15.
    Baumgart, F., Gaschler-Markefski, B., Woldorff, M.G., Heinze, H.J., Scheich, H.: A movement-sensitive area in auditory cortex. Nature 400(6746), 724–726 (1999)CrossRefGoogle Scholar
  16. 16.
    Scheich, H., Ohl, F., Schulze, H., Hess, A., Brechmann, A.: What is reflected in auditory cortex: Properties of sound stimuli or what the brain does with them? In: König, R., Heil, P., Budinger, E., Scheich, H. (eds.) The Auditory Cortex A Synthesis of Human and Animal Research, pp. 383–408. Lawrence Erlbaum Associates, Hillsdale (2005)Google Scholar
  17. 17.
    Cichocki, A., Amari, S.I.: Adaptive Blind Signal and Image Processing. John Wiley, Chichester (2002)CrossRefGoogle Scholar
  18. 18.
    Wesarg, T., Brückner, B., Schauer, C.: Modelling speech processing and recognition in the auditory system with a three-stage architecture. In: Vorbrüggen, J.C., von Seelen, W., Sendhoff, B. (eds.) ICANN 1996. LNCS, vol. 1112, pp. 679–684. Springer, Heidelberg (1996)Google Scholar
  19. 19.
    Blumenstein, C., Brückner, B., Mecke, R., Wesarg, T., Schauer, C.: Using a modified hypermap for analyzing moving scenes. In: Ichi Amari, S., et al. (eds.) Progress in Neural Information Processing, Proceedings of the ICONIP 1996, vol. 1, pp. 428–431. Springer, Singapore (1996)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Bernd Brückner
    • 1
  1. 1.Leibniz Institute for NeurobiologyMagdeburgGermany

Personalised recommendations