Multiscale analysis of fMRI data with mixture of Gaussian densities

We describe in this work an exploratory analysis of fMRI data. We regard the fMRI dataset as a set of spatiotemporal signals S_x indexed by their position x. The analysis is performed on the wavelet packet coefficients of the fMRI signals S_x, and we show that we can characterize the coefficients in terms of a mixture model of multivariate Gaussian distributions.