Spatial ICA of fMRI data in time windows

Abstract

In this paper the concept of window ICA is proposed for the analysis of fMRI data. In window ICA, spatial ICA is applied in sliding time windows. The approach avoids the problems related to the high number of signals and may give an insight to small changes during the experiment. Usefulness of the proposed approach is demonstrated in an experiment where a subject listened to auditory stimuli consisting of sinusoidal sounds (beeps) and words in varying proportions. The window ICA finds different auditory activations patterns related to the beeps respectively the words.