White matter Fractional Anisotropy measures in Autism Spectrum Disorder. Implications of differences in structural correlation with performance IQ measures and age. Results from EU- Autism Interventions data.
Robert Anthony Dallyn1,2, Pedro Angel Luque Laguna1,2,3, Cate Davidson1,2,3, EU-AIMS2 TRIALS Consortium1,2, Declan Murphy1, and Flavio Dell'Acqua1,2

1Forensic and Developmental Neuroscience, King's College London, London, United Kingdom, 2Natbrainlab, King's College London, London, United Kingdom, 3Neuroimaging, King's College London, London, United Kingdom


We present voxel-wise statistics on Fractional Anisotropy from EU-AIMS diffusion imaging data on Autism Spectrum Disorder. We validate previous findings of structural white matter abnormalities in a younger cohort. Correlation analysis of white matter development with behavioural tasks points to altered functioning in ASD individuals in visuospatial reasoning tasks, consistent with the central coherence theory of autism. Cohort differences between ASD and control of white matter integrity correlation with age hint toward altered developmental trajectory in Autism Spectrum individuals.


European Autism Interventions is a multicentre initiative aiming to stratify Autism Spectrum Disorder (ASD) through acquisition and integration of large quantities of data from different imaging modalities and behavioural measures. We perform voxel-wise statistical analysis on Diffusion Tensor Imaging data of a large (N=187) subset of ASD and Typically Developing (TD) subjects imaged at King’s College London. We are able to reproduce regional findings of reduced Fractional Anisotropy (FA) from previous studies1 in adults and identify disparities between ASD and TD in brain network correlations of FA with performance IQ measures and subject age.


Data analysed is taken from the King’s College London cohort of EU-AIMS study. Following exclusion for anatomical/genetic abnormality we have a cohort of 187 subjects (77 Typically Developing, 110 Autism Spectrum Disorder). We de-noise2 the data and correct for Gibb’s ringing3, eddy-current correction using exploreDTI4 and outlier detection & replacement using FSL Eddy5. For the TBSS results presented a further 32 subjects categorised with intellectual disability are excluded, and we also exclude subjects with an average RMS movement per DTI volume above 0.6mm6. Behavioural data for each subject includes Performance IQ, evaluated using the Wechsler Abbreviated Scale of Intelligence (WASI).

Voxel-wise statistical analysis of Fractional Anisotropywas carried out using TBSS (Tract-Based Spatial Statistics7) part of FSL8. We examine group differences in FA between ASD and TD, and correlation of FA with performance IQ and Age. For all statistical analysis we also covariate with Gender and correct for multiple comparisons (except where used as a correlate).


TBSS group comparison (TD vs ASD): We find collections of significant (p<0.05)voxels across the body of the Corpus Callosum, the Cerebellar peduncles and the Uncinate fasciculus (fig .1). We are able to replicate the results of MRC-AIMS study1 (cohort age 18-40) with the EU-AIMS dataset (cohort age 6-30, mean=18.2 STD=5.8).

TBSS (Correlation with performance IQ): We find regions of voxels consistent with U-shaped fibres connecting medially to caudal inferior parietal lobule (fig .2) in the left hemisphere which show as significant in the ASD group but not in the whole cohort (TD group did not have sufficient power to produce significant results).

TBSS (Correlation with age): Correlation of FA with age produces understandably a surplus of highly significant (p<0.01) voxels, as our cohort age ranges from 6-30 (fig .3). We perform an image subtraction of correlation for age in ASD population with age in TD population to examine regions of the brain where the statistically significant effects of age on white matter development (p<0.02) are different between the populations. Clusters of voxels in the genu and the splenium of the corpus callosum survive this procedure (fig .3).


Replication of MRC-AIMS structural diffusion findings indicates that results are robust over a wide range of ages above 6 years old. The fact that the Genu and Splenium emerge as highly correlated with age only in ASD and not in TD supports that brain development may follow a delayed trajectory in autism9, with these areas continuing to develop over our age bracket only in the ASD cohort.

We find tentative evidence to support the central coherence10 theory of autism. The WASI used to administer performance IQ test features matrix reasoning and block design tests, the latter in particular have been shown to be implicated in central coherence10. We find that performance on these tasks correlates highly with short-connecting fibres in the inferior parietal lobule associated with visuospatial reasoningonly in the ASD group, supporting the hypothesis that autistic individuals are approaching tasks intuitively from a more localised perspective.


AIMS-2-TRIALS – the work leading to these results has received funding from the Innovative Medicines Initiative Joint Undertaking (IMI). We would like to acknowledge The Sackler Institue, EU-AIMS funding, EU-AIMS2 TRIALS funding, MRC-AIMS funding and all recruiting, scanning and clinical teams across all centres.


1. Marco Catani,Flavio Dell’Acqua, Sanja Budisavljevic, Henrietta Howells, Michel Thiebaut de Schotten, Seán Froudist-Walsh, Lucio D’Anna, Abigail Thompson, Stefano Sandrone, Edward T. Bullmore, John Suckling, Simon Baron-Cohen, Michael V. Lombardo, Sally J. Wheelwright, Bhismadev Chakrabarti, Meng-Chuan Lai, Amber N. V. Ruigrok, Alexander Leemans, Christine Ecker, Michael C. Craig, and Declan G. M. MurphyBrain. 2016 Feb; 139(2): 616–630. Frontal networks in adults with autism spectrum disorder

2. Veraart, J.; Fieremans, E. & Novikov, D.S. Diffusion MRI noise mapping using random matrix theory Magn. Res. Med., 2016, early view, doi: 10.1002/mrm.26059

3. Kellner, E; Dhital, B; Kiselev, V.G & Reisert, M. Gibbs-ringing artifact removal based on local subvoxel-shifts. Magnetic Resonance in Medicine, 2016, 76, 157

4. Jesper L. R. Andersson, Mark S. Graham, Eniko Zsoldos and Stamatios N. Sotiropoulos. Incorporating outlier detection and replacement into a non-parametric framework for movement and distortion correction of diffusion MR images. NeuroImage, 141:556-572, 2016.

5. Leemans A. and Jones D.K., Magnetic Resonance in Medicine 61(6):1336-1349, 2009 (http://www.ncbi.nlm.nih.gov/pubmed/19319973)

6.Yendiki A, Koldewyn K, Kakunoori S, Kanwisher N, Fischl B. Spurious group differences due to head motion in a diffusion MRI study. Neuroimage. 2014 Mar;88:79-90. doi: 10.1016/j.neuroimage.2013.11.027.

7. S.M. Smith, M. Jenkinson, H. Johansen-Berg, D. Rueckert, T.E. Nichols, C.E. Mackay, K.E. Watkins, O. Ciccarelli, M.Z. Cader, P.M. Matthews, and T.E.J. Behrens. Tract-based spatial statistics: Voxelwise analysis of multi-subject diffusion data. NeuroImage, 31:1487-1505, 2006.

8. S.M. Smith, M. Jenkinson, M.W. Woolrich, C.F. Beckmann, T.E.J. Behrens, H. Johansen-Berg, P.R. Bannister, M. De Luca, I. Drobnjak, D.E. Flitney, R. Niazy, J. Saunders, J. Vickers, Y. Zhang, N. De Stefano, J.M. Brady, and P.M. Matthews. Advances in functional and structural MR image analysis and implementation as FSL. NeuroImage, 23(S1):208-219, 2004.

9. Jason J. Wolff, Ph.D. Hongbin Gu, Ph.D.
Guido Gerig, Ph.D.
Jed T. Elison, Ph.D.
Martin Styner, Ph.D.
Sylvain Gouttard, M.S.
Kelly N. Botteron, M.D. Stephen R. Dager, M.D. Geraldine Dawson, Ph.D. Annette M. Estes, Ph.D.
Alan C. Evans, Ph.D.
Heather C. Hazlett, Ph.D. Penelope Kostopoulos, Ph.D. Differences in White Matter Fiber tract Development Present From 6 to 24 Months in Infants With Autism. (Am J Psychiatry 2012; 169:589–600)

10. Happé, F., & Frith, U. (2006, January). The weak coherence account: Detail-focused cognitive style in autism spectrum disorders.


Fig 1. A) Group comparison of FA between TD and ASD cohort from EU-AIMS KCL cohort. Significant groups of voxels corresponding to the Corpus Callosum, Superior Cerebellar Peduncles and the Uncinate fasciculus (present bilaterally) can be seen. B) Similar results from the previous MRC-AIMS diffusion imaging project1. Group comparison shows significant regions are reproducible in the present study despite large cohort age differences.

Fig 2. A) Voxels shown to have significant correlation of FA with WASI Performance IQ evaluation tasks in whole cohort. B) Voxels shown to correlate only in the ASD cohort. C) Highlighted voxels in inferior parietal lobule, consistent with short connecting fibres in visuospatial region. Highly significant only in ASD cohort. Both images show left hemisphere

Fig 3. A) Significant (p<0.01) voxels correlating increasing FA with age in TD cohort (Blue) and ASD cohort (Red). B) We perform an image subtraction of B from A and show the voxels which survive at a significance of (p<0.02) to show clusters which correlate highly with age in ASD but not in TD. Groups of voxels corresponding to fibres of the Genu and Splenium of the Corpus Callosum are present.

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)