Region of Interest Based Independent Component Analysis

  • Ingo R. Keck
  • Jan Churan
  • Fabian J. Theis
  • Peter Gruber
  • Elmar W. Lang
  • Carlos G. Puntonet
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4232)


The over-complete case remains a difficult problem in the field of independent component analysis (ICA). In this article we combine a technique called “region of interest” (ROI) with a standard complete ICA. We show how to create a mask using ICA, then using the masked data for a second ICA. At the same time this method eliminates a commonly necessary model-based step in fMRI data analysis. We also demonstrate our approach on a real world fMRI data set example.


fMRI Data Independent Component Analysis Blind Source Separation fMRI Data Analysis Blind Source Separation Technique 
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.
    Calhoun, V.D., Adali, T., Hansen, L.K., Larsen, J., Pekar, J.J.: Ica of funtional mri data: An overview. In: Proceedings of the ICA 2003, pp. 281–288 (2003)Google Scholar
  2. 2.
    Amari, S.: Natural gradient learning for over- and under-complete bases in ica. Neural Computation 11, 1875–1883 (1999)CrossRefGoogle Scholar
  3. 3.
    Theis, F.J., Jung, A., Puntonet, C.G., Lang, E.W.: Linear geometric ica: Fundamentals and algorithms. Neural Computation 15, 419–439 (2003)MATHCrossRefGoogle Scholar
  4. 4.
    Bach, F.R., Jordan, M.I.: Beyond independent components: Trees and clusters. Journal of Machine Learning Research 4, 1205–1233 (2003)CrossRefMathSciNetGoogle Scholar
  5. 5.
    Hyvärinen, A., Hoyer, P.: Topographic independent component analysis. Neural Computation 13, 1527–1558 (2001)MATHCrossRefGoogle Scholar
  6. 6.
    Meyer-Bäse, A., Theis, F.J., Lange, O., Puntonet, C.G.: Tree-dependent and topographic independent component analysis for fmri analysis. In: Puntonet, C.G., Prieto, A.G. (eds.) ICA 2004. LNCS, vol. 3195, pp. 782–789. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  7. 7.
    Hyvärinen, A., Oja, E.: Independent component analysis: Algorithms and applications of the hough transform. Neural Networks 13(4–5), 411–430 (2000)CrossRefGoogle Scholar
  8. 8.
    Belouchrani, A., Abed-Meraim, K., Cardoso, J.-F., Moulines, E.: A blind source separation technique using second-order statistics. IEEE Transactions on Signal Processing 45(2), 434–444 (1997)CrossRefGoogle Scholar
  9. 9.
    Hyvärinen, A., Oja, E.: A fast fixed-point algorithm for independent component analysis. Neural Computation 9, 1483–1492 (1997)CrossRefGoogle Scholar
  10. 10.
    Theis, F.J., Gruber, P., Keck, I.R., Tomé, A.M., Lang, E.W.: A spatiotemporal second-order algorithm for fMRI data analysis. In: Proc. CIMED 2005, Lisbon, Portugal, pp. 194–201 (2005)Google Scholar
  11. 11.
    Stone, J.V., Porrill, J., Porter, N.R., Wilkinsonq, I.W.: Spatiotemporal independent component analysis of event-related fmri data using skewed probability density functions. Neuroimage 15(2), 407–421 (2002)CrossRefGoogle Scholar
  12. 12.
    Statistical parametric mapping,
  13. 13.
    Schubotz, R., Frederici, A.D., von Cramon, D.Y.: Time perception and motor timing: A comon cortical and subcortical basis revealed by fmri. NeuroImage 11, 1–12 (2000)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Ingo R. Keck
    • 1
  • Jan Churan
    • 2
  • Fabian J. Theis
    • 1
  • Peter Gruber
    • 1
  • Elmar W. Lang
    • 1
  • Carlos G. Puntonet
    • 3
  1. 1.Institute of BiophysicsUniversity of RegensburgRegensburgGermany
  2. 2.Generation Research ProgramLMU MunichBad TölzGermany
  3. 3.Departamento ATCUniversidad de Granada/ESIIGranadaSpain

Personalised recommendations