Decreased centrality of cortical volume covariance networks in autism spectrum disorders

Author(s)

Balardin, Joana Bisol ^[1] ; Comfort, William Edgar ^[1] ; Daly, Eileen ^[2] ; Murphy, Clodagh ^[2] ; Andrews, Derek ^[2] ; Murphy, Declan G. M. ^[2] ; Ecker, Christine ^[2] ; Sato, Joao Ricardo ^[1] ; Consortium, MRC AIMS

Affiliation

^[1] Univ Fed ABC, Ctr Math Computat & Cognit, BR-09210580 Santo Andre, SP - Brazil

^[2] Kings Coll London, Inst Psychiat Psychol & Neurosci, Dept Forens & Neurodev Sci, Sackler Inst Translat Neurodev, London WC2R 2LS - England

• Document type: Journal article
• Source: JOURNAL OF PSYCHIATRIC RESEARCH; v. 69, p. 142-149, OCT 2015.
• Web of Science Citations: 8

Abstract

Autism spectrum disorders (ASD) are a group of neurodevelopmental conditions characterized by atypical structural and functional brain connectivity. Complex network analysis has been mainly used to describe altered network-level organization for functional systems and white matter tracts in ASD. However, atypical functional and structural connectivity are likely to be also linked to abnormal development of the correlated structure of cortical gray matter. Such covariations of gray matter are particularly well suited to the investigation of the complex cortical pathology of ASD, which is not confined to isolated brain regions but instead acts at the systems level. In this study, we examined network centrality properties of gray matter networks in adults with ASD (n = 84) and neurotypical controls (n = 84) using graph theoretical analysis. We derived a structural covariance network for each group using interregional correlation matrices of cortical volumes extracted from a surface-based parcellation scheme containing 68 cortical regions. Differences between groups in closeness network centrality measures were evaluated using permutation testing. We identified several brain regions in the medial frontal, parietal and temporo-occipital cortices with reductions in closeness centrality in ASD compared to controls. We also found an association between an increased number of autistic traits and reduced centrality of visual nodes in neurotypicals. Our study shows that ASD are accompanied by atypical organization of structural covariance networks by means of a decreased centrality of regions relevant for social and sensorimotor processing. These findings provide further evidence for the altered network-level connectivity model of ASD. (C) 2015 Elsevier Ltd. All rights reserved. (AU)

• FAPESP's process: 13/10498-6 - Machine learning in neuroimaging: development of methods and clinical applications in psychiatric disorders
• Grantee: João Ricardo Sato
• Support Opportunities: Regular Research Grants