Brain functional magnetic resonance imaging lateralization and diagnostic classification in Autism

Systematic differences in functional connectivity magnetic resonance imaging metrics have been consistently observed in autism. I attempted to predict group membership using data provided by the Autism Brain Imaging Data Exchange, including resting state functional magnetic resonance imaging data obtained from 964 subjects and 16 separate international sites. For each of 964 subjects, I obtained pairwise functional connectivity measurements from a lattice of 7266 regions of interest covering the gray matter and attempted to classify the subjects using a leave-one-out classifier with the 26.4 million connections as features. Classification accuracy significantly outperformed chance but was much lower for multisite prediction than for previous single site results. As high as 60% accuracy was obtained for whole brain classification. Classification accuracy was significantly higher for sites with longer blood oxygen-level dependent imaging times. Attempts to use multisite classifiers will likely require improved classification algorithms, longer blood oxygen-level dependent imaging times, and standardized acquisition parameters for possible future clinical utility. Lateralization of brain structure and function occurs in typical development and subserves functions such as language and visuospatial processing. Abnormal lateralization is present in various neuropsychiatric disorders. It has been conjectured that individuals may be left-brain dominant or right-brain dominant based on personality and cognitive style, but neuroimaging data has not provided clear evidence whether such iv phenotypic differences in the strength of left-dominant or right-dominant networks exist. I evaluated whether strongly lateralized connections covaried within the same typically developing individuals (n = 1011). I also compared lateralization of functional connections in typical development and in autism. In typical development, left- and rightlateralized hubs formed two separable networks of mutually lateralized regions. Connections involving only left- or only right-lateralized hubs showed positive correlation across subjects, but only for connections sharing a node. Our data are not consistent with a whole-brain phenotype of greater "left-brained" or greater "rightbrained" network strength across individuals. The autism group lacked left lateralization in three connections involving language regions and regions from the default mode network. Abnormal language lateralization in autism may be due to abnormal language development rather than a deficit in hemispheric specialization of the entire brain.