Functional Connectivity In Children With Low-function Autism Spectrum Disorders:A Group Independent Component Analysis Study

Objective:Neuroimaging studies have shown that autism spectrum disorders(ASDs)may be associated with abnormalities in brain structures and functions at rest as well as during cognitive tasks.However,it remains unclear if functional connectivity(FC)of all brain neural networks is also changed in these subjects.Materials and Methods:In this study,we acquired functional magnetic resonance imaging scans from 93 children with ASD and 79 matched healthy subjects.Group independent component analysis was executed for all of the participants to estimate FC.One-sample t-tests were then performed to obtain the networks for each group.Group differences in the different brain networks were tested using two-sample t-tests.Then,multiple comparisons were performed by using the false discovery rate criterion to acquire the most significant results of intergroup two-sample t-tests(p<0.05,FDR corrected).Finally,relationships between abnormal FC and clinical variables were investigated with Pearson's correlation analysis.Results:The results from one-sample t-tests revealed nine networks with similar spatial patterns in these two groups.When compared with the controls,children with ASD showed increased connectivity in the right dorsolateral superior frontal gyrus and left middle frontal gyrus(MFG)within the occipital pole network(OPN).Children with ASD also showed decreased connectivity in the left gyrus rectus,left middle occipital gyrus,right angular gyrus,right MFG and right inferior frontal gyrus,orbital part within the lateral visual network(LVN),the left inferior frontal gyrus,right precuneus,and right angular gyrus within the left frontoparietal(cognition)network(FPN).Furthermore,the mean FC values within the LVN showed significant positive correlations with total score of the Childhood Autism Rating Scale.Conclusion:Our findings indicate that abnormal FC extensively exists within the OPN,LVN and left FPN in children with ASD.This abnormal FC may constitute a biomarker of ASD.Our results are an important contribution to the study of neuropathophysiological mechanisms in children with ASD.