Search

Prior neuroimaging studies have reported white matter network underconnectivity as a potential mechanism for autism spectrum disorder (ASD). In this study, we examined the structural connectome of children with ASD using edge density imaging (EDI), and then applied machine-learning algorithms to identify children with ASD based on tract-based connectivity metrics. Boys aged 8–12 years were included: 14 with ASD and 33 typically developing children. The edge density (ED) maps were computed from probabilistic streamline tractography applied to high angular resolution diffusion imaging. Tract-based spatial statistics was used for voxel-wise comparison and coregistration of ED maps in addition to conventional diffusion tensor imaging (DTI) metrics of fractional anisotropy (FA), mean diffusivity (MD), and radial diffusivity (RD). Tract-based average DTI/connectome metrics were calculated and used as input for different machine-learning models: naïve Bayes, random forest, support vector machines (SVMs), and neural networks. For these models, cross-validation was performed with stratified random sampling ( × 1,000 permutations). The average accuracy among validation samples was calculated. In voxel-wise analysis, the body and splenium of corpus callosum, bilateral superior and posterior corona radiata, and left superior longitudinal fasciculus showed significantly lower ED in children with ASD; whereas, we could not find significant difference in FA, MD, and RD maps between the two study groups. Overall, machine-learning models using tract-based ED metrics had better performance in identification of children with ASD compared with those using FA, MD, and RD. The EDI-based random forest models had greater average accuracy (75.3%), specificity (97.0%), and positive predictive value (81.5%), whereas EDI-based polynomial SVM had greater sensitivity (51.4%) and negative predictive values (77.7%). In conclusion, we found reduced density of connectome edges in the posterior white matter tracts of children with ASD, and demonstrated the feasibility of connectome-based machine-learning algorithms in identification of children with ASD.

The edges of the structural connectome traverse the white matter to connect cortical and subcortical nodes, although the anatomic embedding of these edges is generally overlooked in the literature. Characterization of the geometry of the structural connectome could provide an improved understanding of the relative importance of various white matter regions to the network architecture of the human brain in normal development and aging, as well as in white matter diseases with regionally specific patterns of vulnerability. Edge density imaging (EDI) has previously been used to show that the posterior periventricular white matter contains a disproportionately large number of connectome edges. In this study, the regional distribution of connectome edges within cerebral white matter, including the importance of posterior periventricular white matter, is further investigated and demonstrated to be invariant to different gray matter parcellations and different diffusion MRI acquisition and postprocessing/tractography methods. An examination of the highest k-core edges and a virtual lesion analysis illuminate hemispheric asymmetries (leftright) in the embedding of connectome edges. Therefore, EDI reveals specific areas of vulnerability within the white matter connectivity of the human brain, especially in the periventricular white matter. The idea of a periventricular nexus fits with the known neurobiology of brain development and may result from simple geometrical considerations in minimizing wiring cost in structural brain connectivity.