Search for a command to run...
The gold standard for clinical assessment of Autism Spectrum Disorders (ASD) relies on assessing behavior via semi-structured play-based interviews and parent interviews. Although these methods show good sensitivity and specificity in diagnosing ASD cases, behavioral assessments alone may hinder the identification of asymptomatic at-risk group. Resting-state functional magnetic resonance imaging (rs-fMRI) could be an appropriate approach to produce objective neural markers to supplement behavioral assessments due to its non-invasive and task-free nature. Previous neuroimaging studies reported inconsistent resting-state abnormalities in ASD, which may be explained by small sample sizes and phenotypic heterogeneity in ASD subjects, and/or the use of different analytical methods across studies. The current study aims to investigate the local resting-state abnormalities of ASD regardless of subject age, IQ, gender, disease severity and methodological differences, using activation likelihood estimation (ALE). MEDLINE/PubMed databases were searched for whole-brain rs-fMRI studies on ASD published until Feb 2018. Eight experiments involving 424 subjects were included in the ALE meta-analysis. We demonstrate two ASD-related resting-state findings: local underconnectivity in the dorsal posterior cingulate cortex (PCC) and in the right medial paracentral lobule. This study contributes to uncovering a consistent pattern of resting-state local abnormalities that may serve as potential neurobiological markers for ASD.