Investigation of Brain Default Network's activation in autism spectrum disorders using Group Independent Component Analysis

Abstract

Autism Spectrum Disorders (ADS), with unknown etiology, is one of the most understudy fields of research worldwide that requires complicated and delicate analytical study methods. The purpose of this study was to compare active regions of Brain Default Mode Network (DMN) using Group Independent Component Analysis (6ICA) among resting state patients with Autism Disorder and healthy subjects. Default Mode Network consists of posterior cingulate cortex (PCC), lateral parietal cortex/angular gyrus retrosplenial cortex, medial prefrontal cortex, superior frontal gyrus, parahippocampal gyrus and temporal lobe shows more prominent activity in passive resting conditions. The diagnosis of autism disorder was According to the clinical ADI-R of ADOS test. After Data preprocessing of rs-fMRI dataset by SPM toolbox, Group independent component analysis was performed in three steps of data reduction, ICA, and back reconstruction using the group ICA of fMRI toolbox (GIFT). Group ICA revealed sixteen Default mode network components which five of them were chosen as DMN components to compare between both groups. The number of voxels in each cluster of autistic individuals was significantly lower than in healthy individuals. Spatial group ICA of fMRI can be a useful approach to determine and study of differences in DMN of brain in patients with autism disorder.