Interpretable Nonnegative Incoherent Deep Dictionary Learning for FMRI Data Analysis

Abstract

Extracting information from fMRI data constitutes a broad active area of research. Current techniques still present several limitations; some ignore relevant aspects regarding the brain functioning or lack of interpretability. In an effort to overcome such limitations, we introduce an extension of the sparse matrix factorization approach to a multilinear decomposition. The proposed model is built upon natural justifiable assumptions and better accommodates the brain behavior. Tests on realistic synthetic as well as real fMRI datasets demonstrate significant performance gains over existing methods of this kind.