“School of Cognitive Sciences”

Back to Papers Home
Back to Papers of School of Cognitive Sciences

Paper   IPM / Cognitive Sciences / 7521
School of Cognitive Sciences
  Title:   How to Control the false positive rate in fuzzy Clustering : Application to fMRI Activation detection using a homodynamic-based feature space
  Author(s): 
1.  H. Jahanian
2.  G.A. Hossein Zadeh
3.  H. Soltanian Zadeh
  Status:   In Proceedings
  Proceeding: BioMed
  Year:  2003
  Supported by:  IPM
  Abstract:
FCM suffers from some drawbacks such as a priori definition of number of clusters, unidentified statistical significance of results, and instability of results when it is applied on raw fMRI time series. Using the randomization we developed a method to control the rate of false positive detection in FCM which gives a meaningful statistical significance to the results. Making use of it, we derived the optimum number of clusters. In this study we applied the FCM on a feature space that takes the variability of hemodynamic response function into account and compared it with the cross correlation feature space.


Download TeX format
back to top
scroll left or right