Tree-Dependent and Topographic Independent Component Analysis for fMRI Analysis

  • Anke Meyer-Bäse
  • Fabian J. Theis
  • Oliver Lange
  • Carlos G. Puntonet
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3195)

Abstract

Recently, a new paradigm in ICA emerged, that of finding “clusters” of dependent components. This striking philosophy found its implementation in two new ICA algorithms: tree–dependent and topographic ICA. Applied to fMRI, this leads to the unifying paradigm of combining two powerful exploratory data analysis methods, ICA and unsupervised clustering techniques. For the fMRI data, a comparative quantitative evaluation between the two methods, tree–dependent and topographic ICA was performed. The comparative results were evaluated based on (1) correlation and associated time–courses and (2) ROC study. It can be seen that topographic ICA outperforms all other ICA methods including tree–dependent ICA for 8 and 9 ICs. However, for 16 ICs topographic ICA is outperformed by both FastICA and tree–dependent ICA (KGV) using as an approximation of the mutual information the kernel generalized variance.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Arfanakis, K., Cordes, D., Haughton, V., Moritz, C., Quigley, M., Meyerand, M.: Combining independent component analysis and correlation analysis to probe interregional connectivity in fmri task activation datasets. Magnetic Resonance Imaging 18, 921–930 (2000)CrossRefGoogle Scholar
  2. 2.
    McKeown, M., Jung, T., Makeig, S., Brown, G., Jung, T., Kindermann, S., Bell, A., Sejnowski, T.: Analysis of fmri data by blind separation into independent spatial components. Human Brain Mapping 6, 160–188 (1998)CrossRefGoogle Scholar
  3. 3.
    Biswal, B., Ulmer, J.: Blind source separation of multiple signal sources of fmri data sets using independent component analysis. Journal of Computer Assisted Tomography 23, 265–271 (1999)CrossRefGoogle Scholar
  4. 4.
    Ziehe, A., Müller, K.: Tdsep - an efficient algorithm for blind separation using time structure. In: Proc. ICANN, vol. 2, pp. 675–680 (1998)Google Scholar
  5. 5.
    Cardoso, J.F., Souloumiac, A.: Blind beamforming for non gausssian signals. IEE Proceedings-F 140, 362–370 (1993)Google Scholar
  6. 6.
    Hyvarinen, A., Oja, E.: Independent component analysis: algorithms and applications. Neural Networks 13, 411–430 (2000)CrossRefGoogle Scholar
  7. 7.
    Bach, F.R., Jordan, M.I.: Beyond independent components: Trees and clusters. Journal of Machine Learning Research 4, 1205–1233 (2003)CrossRefMathSciNetGoogle Scholar
  8. 8.
    Hyvarinen, A., Hoyer, P.: Topographic independent component analysis. Neural Computation 13, 1527–1558 (2001)CrossRefGoogle Scholar
  9. 9.
    Chow, C.K., Liu, C.N.: Approximating discrete probability distributions with dependence trees. IEEE Transaction on Information Theory 14, 462–467 (1968)MATHCrossRefGoogle Scholar
  10. 10.
    Hyvarinen, A., Hoyer, P.: Emergence of phase- and shift-invariant features by decomposition of natural images into independent feature subspaces. Neural Computation 12, 1705–1720 (2000)CrossRefGoogle Scholar
  11. 11.
    Kohonen, T.: Emergence of invariant-feature detectors in the adaptive-subspace self-organizing map. Biological Cybernetics 75, 281–291 (1996)MATHCrossRefGoogle Scholar
  12. 12.
    Cardoso, J.F.: Multidimensional independent component analysis. In: Proc. IEEE ICASSP, Seattle, vol. 4, pp. 1941–1944 (1998)Google Scholar
  13. 13.
    Wismüller, A., Lange, O., Dersch, D., Leinsinger, G., Hahn, K., Pütz, B., Auer, D.: Cluster analysis of biomedical image time–series. International Journal on Computer Vision 46, 102–128 (2002)CrossRefGoogle Scholar
  14. 14.
    Woods, R., Cherry, S., Mazziotta, J.: Rapid automated algorithm for aligning and reslicing pet images. Journal of Computer Assisted Tomography 16, 620–633 (1992)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Anke Meyer-Bäse
    • 1
  • Fabian J. Theis
    • 1
    • 2
  • Oliver Lange
    • 1
    • 3
  • Carlos G. Puntonet
    • 4
  1. 1.Department of Electrical and Computer EngineeringFlorida State UniversityTallahasseeUSA
  2. 2.Institute of BiophysicsUniversity of RegensburgRegensburgGermany
  3. 3.Department of Clinical RadiologyLudwig–Maximilians UniversityMunichGermany
  4. 4.Department of Architecture and Computer TechnologyUniversity of GranadaSpain

Personalised recommendations