Multivariate characterization of white matter heterogeneity in autism spectrum disorder

Abstract

The complexity and heterogeneity of neuroimaging findings in individuals with autism spectrum disorder has suggested that many of the underlying alterations are subtle and involve many brain regions and networks. The ability to account for multivariate brain features and identify neuroimaging measures that can be used to characterize individual variation have thus become increasingly important for interpreting and understanding the neurobiological mechanisms of autism. In the present study, we utilize the Mahalanobis distance, a multidimensional counterpart of the Euclidean distance, as an informative index to characterize individual brain variation and deviation in autism. Longitudinal diffusion tensor imaging data from 149 participants (92 diagnosed with autism spectrum disorder and 57 typically developing controls) between 3.1 and 36.83 years of age were acquired over a roughly 10-year period and used to construct the Mahalanobis distance from regional measures of white matter microstructure. Mahalanobis distances were significantly greater and more variable in the autistic individuals as compared to control participants, demonstrating increased atypicalities and variation in the group of individuals diagnosed with autism spectrum disorder. Distributions of multivariate measures were also found to provide greater discrimination and more sensitive delineation between autistic and typically developing individuals than conventional univariate measures, while also being significantly associated with observed traits of the autism group. These results help substantiate autism as a truly heterogeneous neurodevelopmental disorder, while also suggesting that collectively considering neuroimaging measures from multiple brain regions provides improved insight into the diversity of brain measures in autism that is not observed when considering the same regions separately. Distinguishing multidimensional brain relationships may thus be informative for identifying neuroimaging-based phenotypes, as well as help elucidate underlying neural mechanisms of brain variation in autism spectrum disorders.

Multivariate white matter alterations are associated with epilepsy duration

“…We hypothesised that patients with a longer epilepsy duration would be associated with greater abnormalities ipsilateral to the epileptic focus. This approach has been fruitful in studies of autism and traumatic brain injury (Dean et al, 2017;Taylor et al, 2020). Applications of the Mahalanobis distance include analysing individual tracts by integrating multiple diffusion metrics into a single measure, or by pooling numerous metrics from a number of different modalities.…”

Developmental changes of cortical white–gray contrast as predictors of autism diagnosis and severity

“…Such behavioural abnormalities are found to have certain brain-structural and physiological correlates on the level of the cortical gray matter (GM) 2 – 4 . It has also been found that ASD features not only GM aberrations, but also abnormal white-matter (WM) microstructure 5 – 7 , specifically evident as age-related difference between ASD and typical development (TD) in adolescent developmental trajectories 8 . Combining cortical features of GM and WM in one analysis might provide a more complete understanding of the disorder.…”

Detecting microstructural deviations in individuals with deep diffusion MRI tractometry

“…Those tract-segment are then statistically tested in an univariate manner, and as such the correction for multiple comparisons -required by the typical high dimensionality of dMRI data -will hamper the discriminating power of the analysis 20 . Recently, PCA was employed to acknowledge the multivariate nature of dMRI data [20][21][22] , but this approach still relies on linear assumptions thereby ignoring possible complex interactions between the features. We believe strongly that the proposed deep autoencoder approach goes hand-in-hand with existing browser-based dMRI analysis frameworks in encouraging reproducible research and data-driven discoveries.…”

“…Second, when analyzing multiple measures (even when derived within the same tract), statistical analysis is hampered by: (i) the multiple comparisons problem; and (ii) any covariance between measurements 12,20 . Here multidimensional approaches can increase statistical power by combining the sensitivity profiles of independent modalities [20][21][22] .…”