Three shades of grey: detecting brain abnormalities in children with autism using source‐, voxel‐ and surface‐based morphometry

Abstract

Autistic spectrum disorder (ASD) is a neurodevelopmental disorder characterized by deficits in social interactions, communication and stereotyped behaviour. Recent evidence from neuroimaging supports the hypothesis that ASD deficits in adults may be related to abnormalities in a specific frontal-temporal network [Autism-specific Structural Network (ASN)]. To see whether these results extend to younger children and to better characterize these abnormalities, we applied three morphometric methods on brain grey matter (GM) of children with and without ASD. We selected 39 sMRI images of male children with ASD and 42 typically developing (TD) from the Autism Brain Imaging Data Exchange database. We used source-based morphometry (SoBM), a whole-brain multivariate approach to identify GM networks, voxel-based morphometry (VBM), a voxel-wise comparison of the local GM concentration and surface-based morphometry (SuBM) for the estimation of the cortical parameters. SoBM showed a bilateral frontal-parietal-temporal network different between groups, including the inferior-middle temporal gyrus, the inferior parietal lobule and the postcentral gyrus; VBM returned differences only in the right temporal lobe; SuBM returned a thinning in the right inferior temporal lobe thinner in ASD, a higher gyrification in the right superior parietal lobule in TD and in the middle frontal gyrus in ASD. For the first time, we investigated the brain abnormalities in children with ASD using three morphometric techniques. The results were relatively consistent between methods, stressing the role of an Autism-specific Structural Network in ASD individuals. We also make methodological speculations on the relevance of using multivariate and whole-brain neuroimaging analysis to capture ASD complexity.

A systematic review of structural MRI biomarkers in autism spectrum disorder: A machine learning perspective

“…Furthermore, alterations in the frontal lobe are consistent with the known role of frontal striatal circuitry with repetitive behaviours (Kohls et al, 2014), making it a potentially useful endophenotype of these symptoms. A further nine studies reported CT changes in the temporal lobes, however the direction of these findings was inconsistent with four reporting regional increases in temporal lobe CT (Andrews et al, 2017; Haar et al, 2016; Hyde et al, 2009; Sharda et al, 2016b), and five reporting regional decreases in temporal lobe CT (Andrews et al, 2017; Doyle‐Thomas et al, 2013a; Ecker et al, 2013a; Pappaianni et al, 2018; Shi et al, 2013). As with the temporal lobe, the remaining findings showed a comparable number of studies reporting increases and decreases in the CT of other brain regions.…”

“…From the reviewed studies outlined in Supplementary Table 1, this increase in ICV is largely a result of regional increases in the grey matter volume (GMV), with 20 studies reporting regional increases in GMV, compared to 10 studies reporting reduced regional GMV. One of the most implicated regions of autism was a reported increase in GMV in the inferior, superior and middle temporal gyri in 13 reviewed studies (Brun et al, 2009; Ecker et al, 2012, 2010b, Lai et al, 2015, 2013; Lim et al, 2015; Pappaianni et al, 2018; Riddle et al, 2017; Wang et al, 2017; Xiao et al, 2014). The involvement of the temporal lobe in autism has been discovered previously (Hetzler and Griffin, 1981), and can be involved with the socioemotional abnormalities that are typical of ASD (Bachevalier, 1994; Blakemore, 2008).…”

“…Of the two reviewed studies quantifying sulcal morphology, specifically sulcal depth (SD), one reported altered SD in children with autism (Dierker et al, 2015) and the other reported no changes in SD (Pappaianni et al, 2018). Despite increased SD (Ecker et al, 2010a; Nordahl et al, 2007), sulcal length (Shokouhi et al, 2012) and sulcal depth and length combined (Auzias et al, 2014) being observed in patients with ASD in several 1.5 T MRI studies, specifically in frontal and parietal regions, as well as sulcal morphology being correlating with several measures of severity including CARS and ADOS (Auzias et al, 2014; Watanabe et al, 2014), there is limited evidence to establish a specific measure of sulcal morphology as a biomarker of autism.…”

Sex differences in brain structure: An autism twin study on restricted and repetitive behaviors

“…However, in that study, the sex specific effects were found only for white, not gray matter. Furthermore, abnormal intraparietal sulci (Auzias et al 2014) and reduced gyrification of an area near the intraparietal sulcus (Pappaianni et al 2017) were observed in male children with ASD. Functional activation studies have found differential activation in males with ASD in the intraparietal sulcus during motor learning (Travers et al 2015) response shifting (Shafritz et al 2008) tasks, and in the orbital gyri during motor-inhibition 1 5 (Schmitz et al 2006) and temporal delay decision making (Murphy et al 2017), and these differences correlated in most cases with RRBI symptoms (Shafritz et al 2008;Travers et al 2015;Murphy et al 2017).…”

Explorations and perspectives on the neurobiological bases of autism spectrum disorder

“…Our authors wade into the ever thorny issue of phenotyping, classification systems and putative relationships between the autism and schizophrenia phenotypes (Aggernaes, ). The issue showcases new work that very effectively leverages the publicly available consortial neuroimaging databases to interrogate potential structural and functional differences between ASD and neurotypical individuals (Pappaianni et al ., ; Subbaraju et al ., ). Structural brain imaging approaches are also used to assess potential white matter tract differences in both the superior longitudinal fasciculus (Fitzgerald et al ., ) and the oft‐implicated corpus callosum (Giuliano et al ., ).…”