Abstract #1101

Murua A, Cordes D, Stanberry L

In this paper we introduce a clustering method based on the Potts spin model that provides a powerful tool to determine functionally connected regions of the brain. The Potts spin model is a thermodynamic model used in statistical physics and provides an intuitively simple approach to clustering since it requires no assumption of the number of clusters or their underlying distribution. The method is demonstrated on fMRI data using a conventional finger-tapping task to obtain functional connectivity maps of motor activity.