Statistical inference in fuzzy cluster analysis of functional MRI

Fuzzy cluster analysis (FCA) of functional magnetic resonance images, suffers from some drawbacks such as a priori definition of number of clusters and unidentified statistical significance of results. Here, we introduce a method to control the rate of false positive detection in FCA which gives a meaningful statistical significance to the results. Using this method, we also derive the optimal number of clusters. In this study by measuring the rate of false alarm detection while analyzing 6 experimental datasets, we evaluate the introduced method for making statistical inference.