Diagnostic classification of intrinsic functional connectivity highlights somatosensory, default mode, and visual regions in autism

Abstract

Despite consensus on the neurological nature of autism spectrum disorders (ASD), brain biomarkers remain unknown and diagnosis continues to be based on behavioral criteria. Growing evidence suggests that brain abnormalities in ASD occur at the level of interconnected networks; however, previous attempts using functional connectivity data for diagnostic classification have reached only moderate accuracy. We selected 252 low-motion resting-state functional MRI (rs-fMRI) scans from the Autism Brain Imaging Data Exchange (ABIDE) including typically developing (TD) and ASD participants (n = 126 each), matched for age, non-verbal IQ, and head motion. A matrix of functional connectivities between 220 functionally defined regions of interest was used for diagnostic classification, implementing several machine learning tools. While support vector machines in combination with particle swarm optimization and recursive feature elimination performed modestly (with accuracies for validation datasets <70%), diagnostic classification reached a high accuracy of 91% with random forest (RF), a nonparametric ensemble learning method. Among the 100 most informative features (connectivities), for which this peak accuracy was achieved, participation of somatosensory, default mode, visual, and subcortical regions stood out. Whereas some of these findings were expected, given previous findings of default mode abnormalities and atypical visual functioning in ASD, the prominent role of somatosensory regions was remarkable. The finding of peak accuracy for 100 interregional functional connectivities further suggests that brain biomarkers of ASD may be regionally complex and distributed, rather than localized.

Alterations in resting state connectivity along the autism trait continuum: a twin study

“…[1][2][3] Notably, a large multi-site classification study including ASD and typically developing individuals (N = 252, 6-36 years) showed that of the top 100 biomarkers of brain connectivity achieving together 90.8% accuracy, 45% were related to hyperconnectivity and the remaining 55% to hypo-connectivity. 4 It is likely that the observed inconsistencies in classical brain connectivity investigations are limited by methodological issues, such as insufficient control for head motion or the inclusion of global signal regression which can modulate or even invert the direction of functional connectivity differences. [5][6][7] Moreover, gender has recently been shown to modulate the direction of connectivity differences in ASD.…”

“…4 It is likely that the observed inconsistencies in classical brain connectivity investigations are limited by methodological issues, such as insufficient control for head motion or the inclusion of global signal regression which can modulate or even invert the direction of functional connectivity differences. [5][6][7] Moreover, gender has recently been shown to modulate the direction of connectivity differences in ASD.…”

Electrophysiological signatures of atypical intrinsic brain connectivity networks in autism

“…Most of these studies focused on either global network efficiency (e.g., the whole brain network) or functional connectivity of specific networks of interest. Studies of the whole brain network using graph theory (Di Martino et al, 2013; Keown et al, 2013), regional homogeneity analysis (Di Martino et al, 2014), or other analytic techniques (Anderson et al, 2011; Chen et al, 2015; Supekar et al, 2013), mostly found enhanced local connectivity over multiple brain regions in ASD (Rudie and Dapretto, 2013), especially in the frontal lobe (Courchesne and Pierce, 2005). In contrast to global network patterns, recent studies on networks of interest indicated region-specific reduced long-range connectivity (Muller et al, 2011; Vissers et al, 2012), mainly on networks that may underlie behavioral abnormalities in ASD, e.g., default mode network (DMN) (Abbott et al, 2015; Assaf et al, 2010; Lynch et al, 2013; Washington et al, 2014), salience network (SN) (Uddin et al, 2013a), and executive control network (ECN) (Abbott et al, 2015).…”

Functional SPECT neuroimaging using machine learning algorithms distinguishes autism spectrum disorder from healthy subjects

“…This work has not previously been performed with brain SPECT imaging data, so these results contribute to the growing body of literature in the neuroimaging field that have utilized pattern recognition with machine learning classification to improve distinguishing autism from typical developing subjects. Other methods include functional connectivity MRI [10,36,[53][54][55][56][57][58], voxel based morphology [59], EEG [19,20,60,61] and diffusion tensor imaging studies [18]. Among the top most informative features identified using LASSO, we found involvement of regions implicated in ASD pathology, including regions of the cerebellum and vermis, anterior cingulate gyrus, amygdala, thalamus, frontal, and temporal lobes.…”

“…Reported ASD abnormalities have been identified in the cerebellum [21] and cerebellar vermis [22], anterior cingulate gyrus [23], amygdala [24][25][26], hippocampus [24,27,28], and areas of the frontal [29][30][31], temporal [29,32] parietal lobes [30], caudate and putamen [29,33]. Additional imaging abnormalities in autism include impaired brain growth [24], cortical thickness [12,34], alterations in white matter architecture [18,35], and aberrant connectivity within the somatosensory, visual and default mode network [36].…”