Multimodal Functional and Structural Brain Connectivity Analysis in Autism: A Preliminary Integrated Approach With EEG, fMRI, and DTI

This paper proposes a novel approach of integrating different neuroimaging techniques to characterize an autistic brain. Different techniques like electroencephalography (EEG), functional magnetic resonance imaging (fMRI), and diffusion tensor imaging (DTI) have traditionally been used to find biomarkers for autism, but there have been very few attempts for a combined or multimodal approach of EEG, fMRI, and DTI to understand the neurobiological basis of autism spectrum disorder (ASD). Here, we explore how the structural brain network correlate with the functional brain network, such that the information encompassed by these two could be uncovered only by using the latter. In this paper, source localization from EEG using independent component analysis and dipole fitting has been applied first, followed by selecting those dipoles that are closest to the active regions identified with fMRI. This allows translating the high temporal resolution of EEG to estimate time varying connectivity at the spatial source level. Our analysis shows that the estimated functional connectivity between two active regions can be correlated with the physical properties of the structure obtained from DTI analysis. This constitutes a first step toward opening the possibility of using pervasive EEG to monitor the long-term impact of ASD treatment without the need for frequent expensive fMRI or DTI investigations.