Kristoffer Hougaard Madsen1,2, Lars Kai Hansen2, Morten Mrup2
1Danish Research Centre for Magnetic Resonance, Copenhagen University Hospital Hvidovre, Hvidovre, Greater Copenhagen, Denmark; 2DTU Informatics, Technical University of Denmark, Kgs. Lyngby, Greater Copenhagen, Denmark
To overcome poor signal-to-noise ratios in fMRI, data sets are often acquired over repeated trials that form a three-way array of space time trials. As fMRI data contain multiple inter-mixed signal components blind signal separation and decomposition methods are frequently invoked for exploratory analysis and as a preprocessing step for signal detection. Here we extend multi-linear decomposition to account for general temporal modelling within a convolutional representation. We demonstrate how this alleviates degeneracy and helps to extract physiologically plausible components. The resulting convolutive multi-linear decomposition can model realistic trial variability as demonstrated in fMRI data.