Table 7.

DTI-based studying of the ASD at adolescence.

Work	Autistic group			Control group			sMRI	Brain regions	Data analysis	Findings
	Age, y	IQ	Size	Age, y	IQ	Size
Barnea-Goraly et al., 2004	11–18	89–113	7 HFA	10–16	98–116	9	3 T; 5 mm; 900 b; 6 GD	Right motor and premotor areas; temporoparietal junction; superior and middle temporal gyrus.	VBA; t-test SPM99	ASD: Decreased FA in ventromedial prefrontal cortices, anterior cingulate gyri, temporoparietal junctions, superior temporal sulcus bilaterally, temporal lobes approaching amygdala bilaterally, occipitotemporal tracts, and CC
Keller et al., 2007	12–26	87–117	34 HFA	13–25	101–119	31	3 T; 3 mm; 850 b; 6 GD	Posteriormidbody/isthmus of CC; left and right anterior corona radiata near CC genu	VBA; REVWMRA; SPM2	ASD: Decreased FA in areas within and near CC and in right retrolenticular IC portion Positively correlated group and linear age effect on FA resulted in posterior right IC limb
Alexander et al., 2007	9–24	95–121	43 AD, ASP, NPDD	10–22	101–125	34	3 T; 2 mm; 1000 b; 12 GD	CC genu, body and splenium	ROI; ANOVA; BCA; AIR; FSL	ASD subgroup: Decreased FA and increased MD and λ⊥ in CC; significantly lower performance IQ than for either ASP+NPDD or controls; positively correlated λ⊥ and processing speed during the performance IQ tests (ASD subjects with low FA and high MD seem to have the highest λ⊥ and slowest processing speeds)
Lee et al., 2007	9–23	95–121	43 AD, NPDD	10–22	101–125	34	3 T; 2 mm; 1000 b; 12 GD	ASD: Superior temporal gyrus; temporal stem	ROI; ANCOVA; FSL; SPSS14.0	ASD: In all regions, significantly decreased FA and significantly increased both MD and λ⊥; very little age-related FA changes Controls: Increased FA with age Controls+ASD: Decreased MD and λ⊥ with age
Lee et al., 2009	9–23	94–121	43 HFA, NPDD	10–22	101–125	34	3 T; 2.5 mm; 1000 b; 12 GD	CC; bilateral superior temporal gyrus; anterior cingulum bundle	VBA; T-SPOON; ANCOVA; AIR; FSL; SPSS 14.0	ASD: T-SPOON VBA is more consistent with ROI measurements in CC and temporal lobe regions; T-SPOON VBA found unobserved before in Lee et al. (2007) and significant focal CC body regions during FA vs. processing speed test
Adluru et al., 2009	9–23	–	41 HFA	10–22	–	32	2 mm	WM tracts through CC splenium	Classification; VBA; FSL	ASD vs. controls classification by WM tracts shapes: 75.3% accuracy, 71.9% specificity, and 0.7645 AUC
Noriuchi et al., 2010	11–17	86–100	7 HFA, NPDD	10–16	106–126	7	3 T; 2 mm; 800 b; 32 GD	Anterior cingulate cortex; left dorsolateral prefrontal cortex; right temporal pole; amygdala, superior longitudinal fasc.; occipitofrontal fasc.; mid- and left anterior CC	VBA, MVR; SPM2; SPSS 17.0	ASD: Reduced FA and λ|| in left dorsolateral prefrontal cortex, posterior superior temporal sulcus/temporo-parietal junction, right temporal pole, amygdala, superior longitudinal fasc. and occipitofrontal fasc.; negatively correlated FA of left dorsolateral prefrontal cortex and SRS scores
Lange et al., 2010	10–22	93–127	30 HFA	10–22	102–128	30	3 T; 2 mm; 1000 b; 12 GD	Superior temporal gyrus; temporal stem	Classification; ROI; QDA	ASD: Reversed hemispheric DT skewness asymmetry in superior temporal gyrus; increased DT skewness on right and decreased FA in left superior temporal stem; increased MD, λ||, and λ⊥ in right temporal stem; positively correlated FA of left superior temporal gyrus and composite Vineland score; negatively correlated FA of right temporal stem and performance IQ/language functioning ASD vs. controls classification by six WM measurements: 92% accuracy.; 94% sensitivity, and 90% specificity
Knaus et al., 2010	11–19	VIQ: 104	14 AD	11–19	VIQ: 119	20	3 T; 2 mm; 1000 b; 15 GD	Arcuate fasc	PT; ROI; ANOVA; FreeSurfer; FSL; BET	ASD vs. controls: More prevalent atypical language laterality; no significant group differences in FA, but subjects with more typical leftward lateralization had greater FA in arcuate fasc. across both groups
Fletcher et al., 2010	12–16	85–123	10 HFA	12–14	93–113	10	3 T; 2.5 mm; 1000 b; 12 GD	Arcuate fasc	ROI; LME models; ITK-SNAP	ASD: Increased MD and λ⊥ in the left arcuate fasc,; less lateralized MD and FA
Sahyoun et al., 2010b	11–15	89–113	12 HFA	11–16	97–115	12	3 T; 2 mm; 700 b; 60 GD	Inferior parietal and frontal sulcus; medial temporal gyrus; superior temporal sulcus; fusiform gyrus	PT; ROI; FSL	ASD: Decreased FA in pathways between inferior frontal sulcus and fusiform gyrus and in both hemispheres; secreased FA in pathways between Inferior frontal sulcus and medial temporal gyrus and in right hemisphere
Shukla et al., 2010	12–13	VIQ: 107–113	26 HFA	12–14	VIQ; 107–113	24	3 T; 5 mm; 2000 b; 15 GD	CC genu, body, and splenium; IC genu and anterior and posterior limbs; middle cerebellar peduncle	ROI; VBA; ANOVA; FSL; SPM5; SPSS 16.0	Decreased FA and Increased λ⊥ for whole-brain WM and all three segments of CC and IC; increased MD for whole brain and for IC anterior and posterior limbs; decreased λ|| in CC body and decreased FA in middle cerebellar peduncle
Sahyoun et al., 2010a	11–15	89–113	9 HFA	11–16	97–115	12	3 T; 2 mm; 700 b; 60 GD	Forceps minor; bilateral superior longitudinal fasc. and arcuate fasc.; bilateral inferior fronto-occipital fasc.; right cerebellar WM; bilateral uncinated fasc	TBSS; ANOVA; FSL; BET; IRTK	Controls: Increased FA in WM tracts connecting with frontal lobe: bilaterally within forceps minor, in left inferior fronto-occipital fasc adjacent to middle and inferior frontal gyri, left superior longitudinal fasc., and right posterior ASD: Increased FA bilaterally within uncinate fasc. in temporal lobe and in right superior longitudinal fasc. peripherally near middle frontal gyrus Group differences in correlations between task performance and FA of different brain areas
Jou et al., 2011b	9–17	46–116	10 AD, ASP, NPDD	10–18	87–123	10	1.5 T; 4 mm; 1000 b; 6 GD	Anterior radatiata and body of CC/cingulum; left superior longitudinal fasc./accurate fasc.; left inferior fronto-occipital fasc.; bilateral inferior longitudinal fasc	VBA; VOI; DT; BioImage Suite	ASD: Decreased FA in inferior longitudinal fasc./inferior fronto-occipital fasc., superior longitudinal fasc., and CC/cingulum
Ameis et al., 2011	9–15	79–119	19 AD, NPDD	8–16	86–116	16	3 T; 3 mm; 1250; 12 GD	Bilateral uncintate fasc.; bilateral inferior fronto-occipital fasc.; forceps minor and major	TBSS; VBA; HRM; FSL 4.1; ANALYZE 9.0; SPSS 15.0	ASD: Increased MD and λ⊥ in corticocortical and inter-hemispheric WM tracts, especially in children and within frontal lobe
Verhoeven et al., 2012	12–15	–	19 AD	9–11	12		3 T; 2.2 mm; 800 b; 45 GD	Superior longitudinal fasc	DT; ROI; GLM; ExploreDTI	No significant differences in FA and ADC between ASD-LI and control subjects; decreased FA in ASD-SLI subjects w.r.t. their controls
Shukla et al., 2011b	10–16	106–112	26 AD, ASP	10–16	VIQ: 105–111	24	3 T; 5 mm; 2000 b; 15 GD	Short and long distance WM tracts in frontal, parietal, and temporal lobes in both hemispheres	TBSS; ROI; ANOVA; FSL; FNIRT	ASD: Decreased FA and Increased MD and λ⊥ in short distance tracts in frontal lobe; increased MD and λ⊥ in short distance tracts in temporal and parietal lobes Significant positive correlations between age and FA and negative correlations between age and MD and λ⊥ in short-distance tracts in each lobe in controls, but only in frontal lobe of ASD subjects
Groen et al., 2011	12–16	80–116	17 HFA	14–17	96–114	25	1.5 T; 2.5 mm; 900 b; 30 GD	Superior longitudinal fasc./accurate fasc.; inferior longitudinal fasc.; left corona radiate	VBA; SPM5	ASD: Decreased FA in left and right superior and inferior longitudinal fasc. (but this effect is insignificant after adjusting for age and IQ); increased kurtosis of WM FA probability distribution; increased MD tin all brain regions and shifted GM and WM MD probability distributions; no difference in GM or WM volume
Lo et al., 2011	14–16	101–116	15 HFA	14–16	101–121	15	3 T; 2.7 mm; 4000 b; 102 GD	Bilateral cingulum bundle; arcuate fasc.; uncinate fasc	DSI; ROI; ODF; MARINA; SPM5; SPSS 13.0	ASD: Decreased GFA in the three callossal fiber tracts Consistent leftward asymmetry in three pairs of association fibers in controls, but not in ASD subjects
Shukla et al., 2011a	12–13	106–112	26 AD, ASP	12–14	VIQ: 105–111	24	3 T; 2.7 mm; 2000 b; 15 GD	Inferior longitudinal fasc.; inferior fronto-occipital fasc.; superior longitudinal fasc./accurate fasc.; cingulum	TBSS; TLCBT	ASD: Decreased FA and increased MD in IC anterior and posterior limbs, CC, inferior and superior longitudinal fasc., inferior fronto-occipital fasc., CST, cingulum, and anterior thalamic radiation Age was positively correlated with FA in clusters of significant group differences and negatively correlated with MD and λ⊥ in controls, but not in ASD
Bode et al., 2011	13–15	–	27 HFA	13–16	–	26	3 T; 3.5 mm; 1000 b; 40 GD	Optic radiation; right inferior fronto-occipital fasc	TBSS; BET; FNIRT	ASD: Increased FA and decreased λ⊥ in the area containing clusters of optic radiation and right inferior fronto-occipital fasc.; no age-related correlations
Ameis et al., 2013	9–16	–	19 HFA	8–16	–	16	3 T; 3 mm; 1250; 12 GD	Cingulum bundle	DT; Diffusion Toolkit; SPSS 20	ASD: Significant age-group interaction effects across FA, MD, λ⊥, and λ|| in cingulum bundle; increased MD, λ⊥, and λ|| and decreased FA in cingulum bundle of younger subjects
Nair et al., 2013	11–17	VIQ: 96–127	26	12–16	VIQ: 95–117	27	3 T; 2 mm; 1000 b; 61 GD	Thalamic connections with prefrontal, parietal-occipital, motor, somatosensory and temporal cortices	PT; FSL	ASD: Increased MD in tracts connecting thalamus with motor and somatosensory cortices bilaterally and with prefrontal ROI in right hemisphere; increased λ⊥ in thalamic tracts for motor ROI bilaterally and somatosensory ROI in left and prefrontal ROI in right hemisphere, with a further marginal increase for somatosensory ROI in right hemisphere; negatively correlated FA and social and total ADOS scores in fronto-thalamic tracts and temporo-thalamic connections in left hemisphere ASD and controls: No FA or tract volume differences
Ikuta et al., 2014	15–21	77–117	21 AD, ASP	15–21	98–122	21	3 T; 2.5 mm; 1000 b; 31 GD	Cingulum bundle; anterior thalamic radiation	PT; ROI; ANCOVA; FSL; FLIRT	ASD: Decreased FA within cingulum bundle Significant group-age interaction, such that ASD subjects show no typical age-associated FA increases observed among controls; significant negative correlation between cingulum FA and total BRIEF score in ASD subjects, but not controls; ngatively correlated cingulum FA and shifting subscale in ASD subjects, but not in controls; neither group effect, nor group–age interaction in anterior thalamic radiation FA; insignificantly correlated anterior thalamic radiation FA and BRIEF BRI scores

Additional abbreviations: b, b value unit (s/mm²); f, females; fasc., fascilus; m, males; y, year;

vs., versus; w.r.t., with respect to; REVWMRA, Random-effects voxel-wise multiple regression analysis.

ADC, apparent diffusion coefficient; ADOS, autism diagnostic observation schedule;

AIR, automated image registration; ANCOVA, analysis of covariances; T, MRI magnetic field strength (tesla);

ANOVA, analysis of variances; AUC, area under ROC curve;

BCA, bivariate correlation analysis; BRIEF, behavior rating inventory of executive function;

CC, corpus callossum; CST, cortico-spinal tract; DSI, diffusion spectrum imaging;

DT, deterministic tractography; GD, gradient directions;

GFA, generalized FA; GLM, general linear models; HRM, hierarchical regression model;

IC, internal caplsule; LME, linear mixed effect;

MANOVA, multivariate analysis of variances;

MVR, multivariate regression; ODF, orientation distribution function; PT, probabilistic tractography;

QDA, quadratic discriminant analysis; ROC, receiver operating characteristic;

SLI, specific LI; SRS, social responsiveness scale;

TBSS, tract based spatial statistics; TLCBT, tract labeling cluster-based thresholding;

3D Slicer, ANALYZE, BET, BioimageSuite, Diffusion Toolkit,

DSI Studio, DTIprep, ExploreDTI, FiberViewer, FLIRT, FNIRT, FreeSurfer, FSL, IRTK, ITK-SNAP, MARINA, PSL,

RESTORE, SAS, SPM, SPSS, T-SPOON, data processing packages.