Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Hum Brain Mapp. Author manuscript; available in PMC 2010 December 1.
Published in final edited form as:
PMCID: PMC2790012

Mapping Brain Abnormalities in Boys with Autism


Children with autism spectrum disorder (ASD) exhibit characteristic cognitive and behavioral differences, but no systematic pattern of neuroanatomical differences has been consistently found. Recent neurodevelopmental models posit an abnormal early surge in subcortical white matter growth in at least some autistic children, perhaps normalizing by adulthood, but other studies report subcortical white matter deficits. To investigate the profile of these alterations in 3D, we mapped brain volumetric differences using a relatively new method, tensor-based morphometry (TBM). 3D T1-weighted brain MRIs of 24 male children with ASD (age: 9.5 years ± 3.2 SD) and 26 age-matched healthy controls (age: 10.3 ± 2.4 SD) were fluidly registered to match a common anatomical template. Autistic children had significantly enlarged frontal lobes (by 3.6% on the left and 5.1% on the right), and all other lobes of the brain were enlarged significantly, or at trend level. By analyzing the applied deformations statistically point-by-point, we detected significant gray matter volume deficits in bilateral parietal, left temporal and left occipital lobes (p=0.038, corrected), trend-level cerebral white matter volume excesses, and volume deficits in the cerebellar vermis, adjacent to volume excesses in other cerebellar regions. This profile of excesses and deficits in adjacent regions may (1) indicate impaired neuronal connectivity, resulting from aberrant myelination and/or an inflammatory process, and (2) help to understand inconsistent findings of regional brain tissue excesses and deficits in autism.

Keywords: Autism, TBM, white matter, gray matter, cerebellum, morphometry


Autism is a developmental disorder characterized by social deficits, impaired communication, and restricted and repetitive behavior patterns (American Psychiatric Association, 2000). Postmortem and structural magnetic resonance imaging studies have highlighted the frontal lobes, amygdala and cerebellum as pathological in autism (Amaral et al., 2008), but there has yet to be agreement on the anatomical extent, timing, and consistency across subjects of the biological abnormalities (Williams and Minshew, 2007).

Brain imaging studies of these developmental abnormalities often report an increased total brain volume (Hazlett et al., 2005) and early acceleration in brain growth in autism, but it is not agreed whether this enlargement is restricted to childhood or continues into adulthood (Nicolson and Szatmari, 2003).

Studies examining the differential contributions of gray and white matter to this abnormal growth in autistic patients have not had entirely consistent results, some detecting an increase in only gray matter or only white matter, but others finding it in both tissue types (Nicolson and Szatmari, 2003). The localization of this brain volume increase is also debated (Bonilha et al., 2008): frontal areas may contribute disproportionately to the volume increase (Carper et al., 2002), but some suggest that more posterior brain regions are disproportionately affected (Hazlett et al., 2006). A recent meta-analysis also found an overall increase in cerebellar volume (which may be proportional to the increase in total brain volume) and in the caudate nucleus, but found consistent reductions in the cross-sectional area of the corpus callosum (Stanfield et al., 2007).

Most traditional volumetric analysis have used region of interest analyses, using manual tracing of structures or automated segmentation (Yushkevich et al., 2006). Measures of overall structure volumes may fail to detect subtle or highly localized anatomical differences between groups, and may overlook consistent regional differences in anatomical shape. Recently, computational mapping methods have been used increasingly to examine brain structure. Unlike traditional volumetric methods, statistical maps can detect highly localized group differences in brain morphology without the need for manual tracing or prior specification of regions of interest (Thompson et al., 2004a; 2004b). These methods have detected regional thinning of the corpus callosum (Vidal et al., 2006), subtle hippocampal volume reductions (Nicolson et al., 2006), and ventricular volume reductions (Vidal et al., 2008) in autism, even when significant volume reductions in the brain as a whole were not detectable. In Vidal et al. (2008), surface-based statistical maps of group differences revealed subtle, localized reductions in ventricular size in patients with autism in the left frontal and occipital horns, which may reflect exaggerated brain growth early in life. Ventricular volumes measured using traditional methods did not differ significantly between groups. Other voxel-based anatomical mapping techniques, such as voxel-based morphometry (Ashburner and Friston, 2000), have been used to detect subtle alterations in the corpus callosum in autism (Chung et al., 2004). One study suggested that increases in temporal and parietal cortical thickness (Hardan et al., 2006) may contribute to the volumetric increases in autism and may also relate to anomalies in cortical connectivity. Even so, another voxel-based mapping study had apparently conflicting findings (McAlonan et al., 2005): children with autism had a significant reduction in total gray matter volume and significant increase in CSF volume. They had significant localized gray matter reductions within fronto-striatal and parietal gray matter and additional decreases in ventral and superior temporal gray matter.

To better understand the distribution and direction of these effects, further voxel-based studies are urgently needed.

Tensor-based morphometry (TBM) is a related structural image analysis technique that can reveal profiles of volumetric gains and deficits in patients versus control populations. TBM has not, to our knowledge, been applied to study autism. In TBM, a fluid image warping approach reshapes a set of brain images to match a common anatomical template. From these fluid deformation mappings, relative volume differences are computed between each individual and the anatomical template, and displayed voxel-by-voxel as a map. These maps may be compared across groups to identify regions with systematic volumetric differences. TBM has been used previously to characterize brain differences in various neurological disorders such as Alzheimer’s disease, semantic dementia, HIV/AIDS (Chiang et al., 2005, 2007; Leow et al., 2006; Hua et al., 2008, Leporé, 2008a), and neurodevelopmental disorders such as Fragile X syndrome (Lee et al., 2007) and Williams syndrome (Chiang et al., 2007). A similar approach has been applied to longitudinal scans to study brain changes over time (Thompson et al., 2000; Chung et al., 2001; Aljabar et al., 2008). TBM may also be used to study statistical associations between regional brain volumes and relevant predictors, such as age, sex, or IQ (Chiang et al., 2007).

This study had two goals. First we examined the three dimensional (3D) profile of systematic morphometric differences between patients with autism and controls using TBM. While TBM can reveal differences throughout the brain in 3D without a priori specification of regions of interest, we hypothesized that patients with autism would have diffuse volumetric excesses throughout the brain, particularly in the white matter, based on reports of white matter overgrowth in infancy. In line with prior reports, we anticipated localized gray matter abnormalities (either reductions or excesses, as the direction of the effects is not consistent in the literature) in temporal and parietal regions that include classical language processing systems. We also hypothesized that we would detect volume increases in the cerebellum, a region frequently reported as abnormal in autism.

In the original version of TBM, the determinants of the Jacobian matrices are derived from the local deformation field obtained after the nonlinear registration. These encode compressions and expansions, and can be used to map regional volume differences between patients and controls. In this study, we used the more general method described in Pennec (2004) and Leporé et al. (2008a) (summarized in Figure 1), in which the local deformation tensor field is analyzed statistically to detect local volume and local shape differences in tissue.

2. Materials and Methods

2.1 Subjects

Study participants included 24 males with autism (age: 9.5 ± 3.2 years; range: 6 to 16 years) diagnosed using the Autism Diagnostic Interview-Revised (ADI-R) (Lord et al., 1994), the Autism Diagnostic Observation Schedule (ADOS-R) (Lord et al., 2000), and by clinical observation. All patients met DSM-IV-TR criteria for autism (American Psychiatric Association, 2000) as well as ADI-R and ADOS algorithm criteria. Patients were also assessed using the Wechsler Intelligence Scale for Children, 3rd Edition (WISC-III) or the Leiter International Performance Scale. Socioeconomic status was determined for each patient (Hollingshead, 1975). Patients with a non-verbal IQ below 70 were excluded. All patients had a physical examination prior to the study; subjects with a seizure disorder or other neurological condition or a cytogenetic abnormality or genetic syndrome (such as Fragile X syndrome) were excluded. At scan time, eight patients were medication naïve; four others had discontinued their previous medications prior to the scan. Among the remainder, five were being treated with dopamine antagonists, eight were taking stimulants, four were receiving SSRIs (selective serotonin re-uptake inhibitors), and one was being treated with a cholinesterase inhibitor.

Twenty-six healthy males (age: 10.3 ± 2.4 years; range: 6 to 16 years), drawn from the local community through advertisement and word of mouth, participated as control subjects. They were assessed with the Schedule for Affective Disorders and Schizophrenia-Childhood Version (K-SADS) (Kaufman et al., 1997) to ensure that none had a major psychiatric disorder. None had a personal history of neurological disorders or learning disorders or a family history of autism, mental retardation, language disorders, or learning disorders. Controls were also assessed with the WISC-III or the Wechsler Abbreviated Scale of Intelligence; a full-scale IQ of less than 70 was exclusionary. Age, race, handedness, height, and intelligence were compared between the two groups using t-tests or chi-squared analyses (Table 1).

Table 1
Demographic and clinical characteristics of patients with autism and control subjects.

This study was approved by the Health Sciences Research Ethics Board at the University of Western Ontario. The parents or legal guardians of all subjects provided written consent for participation in this study, while the subjects provided written assent.

2.2 Scanning procedure

All subjects were scanned on a 3-Tesla scanner (IMRIs, Winnipeg, Canada). Sixteen of the subjects with autism required sedation with oral midazolam, to complete their scans. Standard T1-weighted localizer images were acquired initially. Images used for volumetric analysis were then acquired using a T1-weighted 3-D MP-RAGE (Magnetization Prepared Rapid Gradient Echo) sequence (TI=200ms, TR=11ms, TE=5 ms, flip-angle=12 degrees, total scan time: 8 minutes) with 1.2mm isotropic voxels.

2.3 Preprocessing

Image distortions due to radiofrequency field inhomogeneities were corrected using a nonparametric method (Sled et al., 1998). Extra-cerebral tissues were removed, assisted by manual editing, in the BrainSuite software package (Shattuck and Leahy, 2002). MRI brain scans were first globally aligned to the International Consortium of Brain Mapping brain template (ICBM-53; Mazziotta et al., 2001) using a 9-parameter registration (3 translations, 3 rotations and 3 orthogonal scales) with the ANIMAL software (Collins et al., 1994). The cerebellum was manually traced in each subject using the program Multitracer (Woods, 2003; available at and registered using a 9-parameter registration (with the ANIMAL software). It was delineated from the most posterior section in the coronal view, where the fissure separating the cerebellum from the cerebrum becomes visible. The brain stem was carefully excluded from the mask in the regions where it begins to merge with the cerebellum (which is more visible in sagittal slices; triaxial views were used to make this easier to identify). The middle cerebellar peduncle and the brachium conjunctivum were excluded at the point where it completely fuses with the middle cerebellar peduncle. Lobar regions of interest were delineated on the control subject used as a target (see paragraph 2.4) according to the criteria used to define lobar boundaries in the ICBM-53 atlas. Gray and white matter segmentations were also created for this same volume, using the BrainSuite software package (Shattuck and Leahy, 2002).

2.4 Fluid image registration

Some TBM studies generate a minimal deformation target (MDT) from the scans, with a mathematically-defined mean geometry for a population (Christensen et al, 1996; Good et al., 2002; Joshi et al., 2004; Kochunov et al., 2001, 2002; Leporé et al., 2007b; Lorenzen et al., 2004; Hua et al., 2007). As in other TBM studies (e.g., Davatzikos et al, 2001), we preferred using registration to a single control subject’s image versus a multi-subject average intensity atlas as it had higher contrast, better spatial resolution and sharper features. Template optimization for TBM is the subject of further on-going study by us and others (Kochunov et al., 2002; Leporé et al., 2007b). All scans were non-linearly registered to this subject’s scan using a fluid image registration algorithm (Bro-Nielsen and Gramkow, 1996; Leporé et al., 2008b; Brun et al., 2007), which was accelerated using a fast filter (described in Gramkow et al., 1996). Details of the method are described in Leporé et al. (2008b).

2.5 Statistical analysis

Non-rigid registration of each individual brain image to the common anatomical template gave a 3D displacement vector field from which we computed Jacobian matrices J of the deformation. Determinants of these Jacobian matrices, det(J), are commonly used in TBM studies and interpreted as “local expansion factors” (Leporé et al., 2008a). They quantify local expansions (where det(J) > 1) or local contractions (where det(J) < 1), and reflect regional volumetric differences between each subject and the corresponding anatomical regions in the template. Another recently developed approach is to retain the full information in the transformation by computing symmetric definite-positive matrices S=JTJ, at each point, called deformation matrices. Multivariate statistics are then computed on these matrices (Leporé et al., 2008a) using the log-Euclidean framework to account for the curvature of the space of positive-definite symmetric matrices to which S belongs (Arsigny et al, 2005). The intuitive meaning of this approach is to detect anatomical regions where structures may be locally enlarged or compressed along certain directions (as explained in Figure 1). As brain growth is anisotropic, i.e., not uniform in all directions (Thompson et al., 2000), some structures may become enlarged in disease, relative to the control average, along certain directions. Multivariate TBM is designed to pick up on these anisotropic changes. Past studies found that multivariate TBM can identify regional abnormalities that are overlooked by the analysis of the determinant (local volume difference) only (Leporé et al., 2008a).

Before statistical analysis, we also covaried the computed deformation matrices and determinants at each voxel with age, to adjust for possible age effects. We computed Scov,ij with ij one component of the matrix and (det J)cov, according to the regression equation


with regression coefficients βi, diagnosis coded using a dummy binary variable, and Scov,ij = SijSij,predicted the resulting adjusted measure. Once adjusted for age effects, the determinants of these Jacobian matrices were used to compute the lobar volumes for each subject after delineating each lobe in the registration target image (as explained in paragraph 2.3). Lobe volumes were averaged within the two groups and compared. The data was also analyzed with two types of statistics: (1) a univariate Student’s t-test on the age-adjusted volumes, after logarithmic transformation, i.e., log10((det J)cov and (2) a multivariate Hotelling’s T2-test on log(Scov). To avoid assuming that our random variables are normally distributed, we used voxelwise permutation tests to establish a null distribution at each voxel (Nichols and Holmes, 2002), using the suprathreshold volume. We permuted the assignments of subjects to groups 5000 times. This number of permutations N was determined according to Edgington (1995), to control the standard error SEp, of the omnibus probability p, which follows a binomial distribution B(N,p) with SEp = p(p1)/N. The overall significance of the observed pattern of effects in the statistical maps is assessed by computing this omnibus probability, p (that we will call corrected pcorrected) which is a way of correcting for the multiple spatial comparisons implicit in computing maps of statistics. The general method is further detailed in Nichols and Holmes (2002). pcorrected values were computed for the overall gray and white matter in each lobe, using the lobar volumes and the classified gray and white matter segmentations (see paragraph 2.3) to the Jacobian determinant maps obtained in each subject.

3 Results

3.1 Subjects

The groups did not differ significantly in age, socioeconomic status, race, or height (see Table 1), although there were proportionally more left-handed subjects in the patient group. While there was no significant difference in non-verbal IQ between the two groups, patients did have a significantly lower verbal and full-scale IQ.

3.2 Analysis of the cerebrum

Volumetric summaries are shown in Figure 2.A. and 2.B. In the raw (unscaled) data, the autistic group had significantly higher frontal lobe volumes in the left (+3.6%; p=0.049) and right hemispheres (+5.1%; p=0.011). The occipital, temporal, and limbic regions were also abnormally enlarged in the autism group on the left, and at trend-level on the right (see Table for significance levels). After adjusting these data for overall differences in brain scale across subjects, none of the lobes showed evidence for reduction or excess in patients with autism compared to controls; even so, group differences in gray matter and trend-level differences in white matter were still detected in the group difference maps (Figure 3).

We expected univariate and multivariate maps to offer more power than lobar averages to detect relative regional volume differences throughout the brain. Figure 3 shows volumetric excesses in the white matter (Figure 3.A) and excesses and losses in the gray matter (Figure 3.B) in the autism group in terms of the volume ratio (i.e., mean autism volume divided by mean control volume). Statistical maps are also shown, based on the univariate (DET) and multivariate tests (LOG). The volume ratio (RATIO) and the univariate ((det J)cov) analyses (DET) show a complex pattern of volume differences (mostly reductions) in the gray matter and excesses in the white matter. Although the differences seem quite prominent in the white matter in the ratio maps, their significance is only at trend level after multiple comparisons correction (pcorrected = 0.1 for DET and pcorrected = 0.09 for LOG); the gray matter volume reductions, however are significant overall when the gray matter alone is assessed (pcorrected = 0.038 for DET and pcorrected = 0.05 for LOG). When using the unscaled data, the difference is not significant either in the white matter, or in the gray matter, which is expected, given the wide variations in brain volume across subjects.

As differences were hypothesized independently in the white matter (based on earlier reports in independent samples Herbert et al., 2004), and in the gray matter (McAlonan et al., 2005) we also conducted analysis restricted to these two regions.

We performed two-tailed t-tests at each voxel to assess the hypothesis of local gray matter abnormalities. Given the significance of these tests (pcorrected = 0.038 for DET and pcorrected =0.05 for LOG, see previous paragraph), we conducted additional one-tailed t-tests in the same region to determine the direction of the changes; as these were post-hoc tests, the significance of the gray matter deficits described should be based on the 2-tailed test only. Two one-tailed t-tests were similarly performed at each voxel in the white matter to assess the alternative hypotheses of white matter excess or deficits in the patient group.

For completeness, as a post hoc test, we also subsequently examined the results of the opposite contrasts (designed to detect white matter loss), and confirmed that there were no effects in those directions.

In Figure 3.A, regions with white matter excesses in the RATIO maps (shown in red, equivalent to +15–20%) correspond to regions of trend-level volume differences in the DET maps, whereas volume deficits in the RATIO maps (light blue, -5 %) do not correspond to any signal in the DET maps (dark blue corresponding to a p-value = 1). These maps indicate that there was only weak evidence for distributed white matter volume excesses in the patients, and no evidence for deficits. When computing one-tailed t-tests over the entire image, there was there was no significant overall white matter excess in autism compared to controls; it was only a trend (this was also the case when using the unscaled data). Although there were excesses in some white matter regions, it is not logically implied that the white matter is enlarged overall, as there may also be subtle but nonsignificant reductions in other white matter regions, i.e., a redistribution, with no net overall gain. Tests were also performed in the different lobes in the gray matter (Figure 3.B), where we found an overall significant omnibus probability pcorrected = 0.038, a finding that was not replicated in the unscaled case. Whereas the RATIO maps exhibit a complex pattern of gray matter excesses and losses, one-tailed t-tests showed evidence for overall gray matter losses, especially in the left and right parietal lobes, the left temporal lobe and the left occipital lobe (Table 2). It may therefore be too simplistic to expect a generalized gray matter deficit in patients; it may be that, as in Williams syndrome (Thompson et al., 2005) some areas show deficits while others show excesses or no systematic difference.

Table 2
Significance of the gray matter excesses and losses computed using one-tailed t-tests. Note that there was no clear directional hypothesis regarding these differences, as the literature is inconsistent and different studies have reported losses and excesses. ...

These results are consistent with the localization of significant p-values in the DET maps in Figure 3.B.

Anisotropic changes, as assessed with the multivariate analysis, were also significant, with pcorrected =0.05 for the Hotelling’s T2 statistic computed in the gray matter and pcorrected = 0.09 (i.e., a trend level effect) in the white matter. The multivariate method is sensitive to both anisotropic and uniform volumetric differences. Gray matter regions with significant group differences at the voxel level were mostly found in the left hemisphere including the supramarginal and superior temporal gyri, and around the anterior part of the central sulcus (see Figure 3.B). Generally, these morphometric differences are consistent with a complex pattern of local volume deficits and excesses in cortical areas (including both gray and white matter).

3.3 Analysis of the cerebellum

Figure 4 shows horizontal sections through the cerebellum in the rostral to caudal direction. The first column shows the volume ratio (i.e., mean autism divided by mean control volume). Both univariate and multivariate analysis implicate the same regions at the voxel level. Cerebellar volume excesses were found to be significant after multiple comparisons correction in the univariate analysis (pcorrected = 0.006). These were detected only at trend level by the multivariate analysis, with pcorrected =0.07. The spatial distribution of effects was similar, suggesting that the univariate test has greater signal-to-noise ratio for detecting differences in this case. Significant cerebellar volume excesses in autism were observed primarily in the vermis; systematic structural differences were also found in lobes III and VIII and in the corpus medullare (volume reduction) and in lobes Vc, Vd, VIIa, VIIb, and IX (volume excess). Corrected p-values for gain (pcorrected = 0.03) and deficits (pcorrected =0.02) were computed from two one-tailed t-tests and were both significant in the cerebellum, suggesting that volume loss and volume gain may occur in different lobules of the cerebellum as shown in the maps. Even so, to take into account the multiple testing for effects of gain and loss, it would be conventional to use a Bonferroni correction to double these significance values. Therefore, we should regard the gain as a trend (pcorrected = 0.06) but the loss as statistically significant (pcorrected = 0.04). This is equivalent to performing a two-tailed test followed by inspection of the gain and loss effects separately.

4 Discussion

Neurobiological findings

In this study, autistic children had significantly enlarged frontal lobes (by 3.6% on the left and 5.1% on the right), and all other lobes of the brain were enlarged significantly, or at trend level. By analyzing the applied deformations statistically point-by-point, we detected significant gray matter volume deficits in bilateral parietal, left temporal and left occipital lobes (p=0.038, corrected), trend-level cerebral white matter volume excesses, and volume deficits in the cerebellar vermis, adjacent to volume excesses in other cerebellar regions. Our maps also suggested trend-level excesses in central white matter volume among these subjects and gray matter losses mainly in the parietal and left temporal and occipital lobes. These results were found using both univariate and multivariate mapping methods and related volumetric regions of interest. Patients with autism also had regional excesses in white matter volume as well as deficits in the volume of lobes III and VIII of the cerebellar vermis and an increase in volume of vermal lobes Vc, Vd, VIIa, VIIb, and IX.

The cerebellum has been studied extensively in autism, since the early work of Courchesne et al. (1988). That paper suggested that for non-adjusted data (i.e., data at its original scale), cerebellar vermal lobules VI–VII were smaller in patients than in controls, suggesting a developmental hypoplasia, while the lobules I–V were normal. Those findings led to a controversy as they were not replicated by some other investigators; Filipek (1995b) stated that a definitive conclusion on the vermis pathology was premature. Indeed, differences in the non-scaled cerebellar vermis volume have often been found (Haas et al., 1996), but not in all studies. Courchesne (1999) found an increased volume in the vermis while Levitt et al. (1999) found that lobules VIII–X were smaller. In our data, which was scaled to adjust for differences in overall brain volume across subjects, we found both volume excesses and deficits in the cerebellum between autistic patients and controls.

We found significant lobar enlargement in autism (in the raw, unscaled data), consistent with prior reports. The autistic group had significantly enlarged frontal lobes (by 3.6% on the left and 5.1% on the right; p<0.049, p<0.011), and all other lobes of the brain were enlarged either significantly, or at trend level, with average enlargements in different lobes ranging from +3.2 to +6.7%. Prior studies of young children (Courchesne et al., 2001;Cody et al., 2006) as well as older children and adolescents (Herbert et al., 2003) have accordingly reported enlargement of white matter volumes in autism. Here we also found gray matter deficits in the bilateral parietal lobes, as well as the left temporal and left occipital lobes. These differences are also relatively subtle and may not be universally found; in other studies of the gray matter in autism, some groups have found increased or decreased volume and others detected no difference.

White matter excesses in autism have been interpreted in functional neuroanatomical terms as suggesting that brain connectivity may be impaired in regions showing volume excesses. The abnormally rapid growth over the brain overall, observed during early infancy, may be the result of an abnormal myelination process during childhood. We recently found excess subcortical gray matter in Fragile X syndrome, a neurodevelopmental disorder whose mechanism is thought to be a genetically mediated impairment in dendritic pruning (Lee et al., 2007). Although the mechanism is different, we also found an excess in cortical thickness in Williams syndrome, another genetically-mediated neurodevelopmental disorder, and this thickening may reflect a failure in cortical neuronal packing, due to deficiencies in the elastin gene (Thompson et al., 2005). As such, it is plausible that excesses in white matter, observed here, may be attributable to an over-production of myelin in infancy, an abnormality in myelin packing, or anomalies in the production or anatomical distribution of oligodendrocytes that produce myelin throughout the white matter.

The presence of neuroinflammation is another factor implicated in white matter volume enlargement. In patients with autism, an active neuroinflammatory process has been shown to exist in the white matter, cortex and cerebellum (Vargas et al., 2005), a finding that may contribute to the volumetric increase in white matter in this and other studies. Although it is not clear what the cause of neuroinflammation might be in autism, there are other situations in which an inflammatory hypothesis has been invoked to explain white matter excesses. In two independent studies of methamphetamine abusers, we (Thompson et al., 2004) and others (Jernigan et al., 2005) found white matter excess in methamphetamine abusers versus controls, in conjunction with gray matter deficits, and we suggested that, in line with the animal literature, inflammatory processes may contribute to the white matter hypertrophy.

In autism there is no drug-induced change (or known pathogen) to trigger an inflammatory process, but it remains a candidate explanation for the enlarged white matter.

If these white matter alterations indicate impaired axonal conduction velocity or impaired neuronal connectivity, this may also lead to a delayed or incomplete development of cortical gray matter structures, in line with the gray matter deficits seen here. Studies with diffusion tensor and functional imaging are required to better evaluate this possibility. We also found gray matter abnormalities in regions that include the left posterior temporal lobes in autistic patients compared to controls, a deficit that may be implicated in the characteristic difficulties in vocabulary and language processing in autism. The left occipital lobe and parietal lobes showed regional gray matter excesses and reductions, which may relate to recent findings demonstrating an abnormal magnocellular pathway in children with autism, which may affect visual processing and sensory integration (Milne et al., 2002).


No well-replicated pattern of characteristic brain abnormalities has yet been found in autism, although some review papers suggest evidence for white matter hypertrophy in at least a subset of autistic patients (Herbert, 2005). In this paper, multivariate tests - minor variants of the standard TBM - were used in addition to the commonly used univariate methods. In principle, they include a larger amount of information on brain morphology, as they analyze the Jacobian matrix J which is derived from the vector fields after fluid registration and not just the determinant of this matrix. For this reason, one might expect these statistics to be consistently more powerful, as they are sensitive to both volume and anisotropic volume differences. However, in our study, multivariate tests did not give greater effect sizes. The overall corrected significance values were not substantially different using the methods that assess volume difference alone (p=0.038, corrected for gray matter, p=0.1 corrected in white matter) versus those that assess potential stretching or compression along a given direction (where p=0.05, corrected, for gray matter, p=0.09, corrected, in white matter), even so the analyses support each other to some extent. The noise in each of the multivariate parameters must be taken into account, and it may generally require a larger sample to estimate them reliably. In other analyses with the same method (Leporé et al., 2008a) we found that the anisotropy statistics detected brain atrophy in HIV-AIDS with genuinely better power than standard volumetric assessments, but in the current autism study, the anisotropy statistics essentially agreed with the volumetric assessments. It is therefore plausible that the difference in autism is better represented as a simple volume difference (with no directional preference), whereas the neurodegeneration in HIV/AIDS may occur preferentially in a certain direction (e.g., radially along corticothalamic tracts in the brain).


As noted earlier, prior MRI studies of regional gray and white matter volumes in autism, using traditional analysis methods, have not always been consistent in their results. TBM may be beneficial in this population as it can reveal systematic differences in brain structure even in situations where overall lobar volume measures cannot. In particular, this situation is possible when one selective subregion belonging to a structure shows systematic gain and a second one shows deficit, as in the present study. In this case, the power to detect the effect is depleted when using overall volumes to summarize differences over lobar regions, which motivated our use of TBM. In other types of voxel-based studies, such as voxel-based morphometry (VBM; Ashburner and Friston, 2000), a question sometimes arises as to whether the findings may be attributable to imperfect registration. This question arises because in VBM, smoothed maps of classified gray matter, derived from an explicit tissue classification of the image into gray and white matter and CSF, are automatically aligned across subjects and smoothed, and then statistical inferences are made regarding group differences, by voxel-by-voxel subtraction of the group-averaged images. As such it is possible that a difference detected at any one location is due to imperfect registration (Thacker, 2005).

In TBM, however, the signals analyzed are based only on the registrations of the images and not the aligned gray matter classifications, so it is not required that the gray matter be perfectly registered across subjects as the gray matter density is not analyzed at each stereotaxic location. As such, false positive findings due to systematic group differences in registration errors are less likely. Even so, there may be false negative findings, because the power to detect morphometric differences depends on the scale at which anatomic data can be matched by the warping algorithm.

When using voxel-based methods such as VBM or TBM, the difficulty in matching cortical regions across subjects may mean that subtle regional differences in cortical structure may go undetected. In TBM, all morphometric differences are inferred from deformation fields based on automated matching of intensities in the images, and the spatial smoothness of these fields makes it difficult to register the entire cortical mantle across subjects, as would be required to gauge the level of systematic atrophy in cortical gray matter. Alternative approaches can compute cortical thickness at each point, but these are typically more time-consuming as they generally extract explicit models of the cortical surface as geometric meshes, prior to computing the cortical thickness directly from the meshes (Lerch and Evans, 2005), or by tissue classification of the images and voxel-coding (Thompson et al., 2004b; Aganj et al., 2008). Even so, there are at least two possible solutions to better sensitize our TBM approach for detecting cortical gray matter loss. The first is to use voxel-based morphometry (VBM; Ashburner and Friston, 2000) or a related approach termed RAVENS (Davatzikos et al., 2001). A second method to identify cortical gray matter atrophy with TBM was developed by Studholme et al. (2003), in which deformation-based compression signals at each point are smoothed adaptively depending on the amount of gray matter lying under the filter kernel. This is a way to avoid incorrect assignment of gray matter differences to the white matter, when both tissues are partial volumed within a voxel. A third solution is to run the deformation maps at a very high spatial resolution and with less spatial regularization, or with a regularizer (smoothness term) that enforces continuity but not smoothness. We plan to investigate these methods in the future, to quantify the gray matter reductions more precisely with independent but more time-consuming methods.

Limitation and future work

Our results should be interpreted taking into account certain specific limitations. As there are known gender differences in the prevalence and severity of autism (Fombonne, 2003), and to some extent in normal brain development (Lenroot et al., 2007), the inclusion of males only in this study may have highlighted group differences by removing gender variables affecting neurodevelopment, and prevents the applicability of the conclusions to girls with autism. The lack of girls in this sample limits what can be said about females with autism, but this is not necessarily a limitation in terms of what can be said about group differences in males.

Furthermore, some patients in the present study were taking psychotropic medications, which may potentially have influenced the results, as we have previously shown for some types of drug treatment in psychiatric cohorts (e.g., in bipolar patients taking lithium: Bearden et al., 2007). In particular, it has been shown that dopamine antagonists, such as risperidone (and other atypical antipsychotics), may influence the extrapyramidal system (e.g., medulla, pons, cerebellum), which plays a role in motor coordination (Chevreuil et al., 2008; Baghdadli et al., 2002).

None of the other medications have been reported to affect brain structures, although there is a lack of studies examining this. There is no conclusive evidence of a possible adverse effect of stimulants, such as ritalin commonly used in ASD, on brain function and development (Ghanizadeh, 2009). SSRIs, such as citalopram, have been shown to affect brain development during prenatal exposure (van der Veere et al., 2007). Even so, to our knowledge, use of SSRIs and cholinesterase inhibitors, has not been shown to be associated with detectable differences in brain structure in children. However, future studies in a larger cohort are required to assess modulatory effects of psychotropic medications. In principle, differences in handedness in the population (Table 1) may also lead to confounding effects (Sun et al., 2006). Even so, some very large studies of normal subjects with voxel-based morphometry (VBM) (N=465; Good et al., 2001) have failed to detect effects of handedness, suggesting that effects of handedness on brain structure might be relatively minimal.

Further studies are required to confirm the differences found here in larger samples. As emphasized by Thompson et al. (2005a) and Shaw et al. (2006, 2008), cortical development is associated with an increase and then a decrease in gray matter. In psychiatric populations (e.g., bipolar illness; Gogtay et al., 2007), or in normal children with above-average intellectual ability (IQ; Shaw et al., 2006), cortical maturation may be accelerated or delayed versus the normal time-course, leading to time-points in which excesses in certain tissue types are detected and other time-points in which deficits are detected, even in the same brain regions (Gogtay et al., 2004). As such, longitudinal data is needed to determine whether these gray matter deficits and lobar volume excesses persist into adulthood, or when they are first detectable. One limitation of this study is cross-sectional design, which deals with age by covarying it out prior to other analysis. Given that the previously demonstrated ability of this method to capture longitudinal changes (Thompson et al., 2000; Hua et al., 2007; Gogtay et al., 2008), we hope to apply this method, in the future, in a longitudinal study design. In the future, TBM may be used within such a design to better understand apparently conflicting voxel-based studies of tissue deficits and excesses (Hardan et al., 2006, McAlonan et al., 2005). Once the developmental trajectory of these structural brain changes is better established, the anatomy of autism and its developmental time-course will be better understood.


This work was funded by grants from the National Institute of Aging, the National Institute for Biomedical Imaging and Bioengineering, and the National Center for Research Resources (AG016570, EB01651, RR019771 to PT). Other financial contributions came from the Child and Parent Resource Institute, the London Health Science Foundation, the Ontario Mental Health Foundation, the Hospital for Sick Children Foundation, and the Human Brain Mapping Project, funded by NIMH and NIDA (MH/DA52176), RR13642, MH655166 to AWT).


  • Aganj I, Sapiro G, Parikshak N, Madsen SK, Thompson PM. Measurement of Cortical Thickness from MRI by Minimum Line Integrals on Soft-Classified Tissue. submitted to Hum Brain Map. [PMC free article] [PubMed]
  • Aljabar P, Bhatia KK, Murgasova M, Hajnal JV, Boardman JP, Srivivasan L, Rutherford MA, Dyet LE, Edwards AD, Ruecket D. Assessment of brain growth in early childhood using deformation based morphometry. NeuroImage. 2008;39(1):348–358. [PubMed]
  • Amaral DG, Schumann CM, Nordahl CW. Neuroanatomy of autism. Trends Neurosci. 2008;31(3):137–145. [PubMed]
  • American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4 text revision. Washington, DC: American Psychiatric Association; 2000.
  • Arsigny V, Fillard P, Pennec X, Ayache N. Fast and Simple Computations on Tensors with Log-Euclidean Metric. INRIA Research Report. 2005:5584.
  • Autism Genome Project Consortium. Mapping autism risk loci using genetic linkage and chromosomal rearrangements. Nat Genet. 2007;39(3):319–328. [PMC free article] [PubMed]
  • Ashburner J, Friston KJ. Voxel-based Morphometry: the methods. NeuroImage. 2000;11(6):805–821. [PubMed]
  • Baghdadli A, Gonnier V, Aussillous C. Review of psychopharmacological treatments in adolescents and adults with autistic disorders. Encephale. 2002;28(3):248–254. [PubMed]
  • Bearden CE, Thompson PM, Dalwani M, Hayashi KM, Lee AD, Nicoletti M, Trakhtenbroit M, Glahn DC, Brambilla P, Sassi RB, Mallinger AG, Frank E, Kupfer DJ, Soares JC. Greater cortical gray matter density in lithium-treated patients with bipolar disorder. Biol Psychiatry. 2007;62(1):7–16. [PMC free article] [PubMed]
  • Bonilha L, Cendes F, Rorden C, Eckert M, Dalgalarrondo P, Li LM, Steiner CE. Gray and white matter imbalance - Typical structural abnormality underlying classic autism? Brain Dev. 2008;30(6):396–401. [PubMed]
  • Bro-Nielsen M, Gramkow C. Fast Fluid Registration of Medical Images. Proceedings of the 4th International Conference on Visualization in Biomedical Computing; Hamburg, Germany. September 22–25; 1996. pp. 267–327.
  • Brun C, Leporé N, Pennec X, Chou Y-Y, Lopez OL, Aizenstein HJ, Becker JT, Toga AW, Thompson PM. Comparison of Standard and Riemannian Fluid Registration for Tensor-Based Morphometry in HIV/AIDS. 10th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI); Brisbane, Australia. Oct 29–Nov 2; 2007.
  • Carper RA, Moses P, Tigue ZD, Courchesne E. Cerebral lobes in autism: early hyperplasia and abnormal age effects. Neuroimage. 2002;16(4):1038–1051. [PubMed]
  • Chevreuil C, Reymann JM, Fremaux T, Polard E, Seveno T, Bentue-Ferrer D. Risperidone use in child and adolescent psychiatric patients. Therapie. 2008;63(5) [PubMed]
  • Chiang M-C, Reiss AL, Eckert MA, Dutton RA, Lee AD, Alaghband Y, Bellugi U, Galaburda A, Korenberg J, Mills DK, Toga AW, Thompson PM. Tensor-Based Morphometry of the Corpus Callosum in Williams Syndrome. 11th Annual Meeting of Society for Human Brain Mapping (HBM); Toronto, Canada. June 12–16; 2005.
  • Chiang MC, Dutton RA, Hayashi KM, Toga AW, Lopez OL, Aizenstein HJ, Becker JT, Thompson PM. 3D Pattern of Brain Atrophy in HIV/AIDS Mapping using Tensor-Based Morphometry. NeuroImage. 2007;34(1):44–60. [PMC free article] [PubMed]
  • Christensen EG, Rabbitt RD, Miller MI. Deformable templates using large deformation kinematics. IEEE Trans Image Process. 1996;5(10):1435–1447. [PubMed]
  • Chung MK, Worsley KJ, Paus T, Cherif C, Collins DL, Giedd JN, Rapoport JL, Evans AC. A unified statistical approach to deformation-based morphometry. NeuroImage. 2001;14(3):595–606. [PubMed]
  • Chung MK, Dalton KM, Alexander AL, Davidson RJ. Less white matter concentration in autism: 2D voxel-based morphometry. NeuroImage. 2004;23(1):242–251. [PubMed]
  • Collins DL, Neelin P, Peters TM, Evans AC. Automatic 3D intersubject registration of MR volumetric data in standardized Talairach space. J Comput Assist Tomogr. 1994;18(2):192–205. [PubMed]
  • Courchesne E, Yeung-Courchesne R, Press GA, Hesselink JR, Jernigan TL. Hypoplasia of the cerebellar vermal lobules VI and VII in autism. N Engl J Med. 1988;318(21):1349–1354. [PubMed]
  • Courchesne E. An MRI study of autism: the cerebellum revisited. Neurology. 1999;52(5):1106–1007. [PubMed]
  • Courchesne E, Karns CM, David HR, Ziccardi R, Carper RA, Tigue ZD, Chisum HJ, Moses P, Pierce K, Lord C, Lincoln AJ, Pizzo S, Schreibman L, Haas RH, Akshoomoff NA, Courchesne RY. Unusual growth patterns in early life in patients with autistic disorder: an MRI study. Neurology. 2001;57(2):245–254. [PubMed]
  • Davatzikos C, Genc A, Xu D, Resnick SM. Voxel-based morphometry using the RAVENS maps: methods and validation using simulated longitudinal atrophy. Neuroimage. 2001;14(6):1361–1369. [PubMed]
  • Edgington ES. Randomization Tests. 3. Marcel Dekker; 1995.
  • Filipek P. Quantitative magnetic resonance imaging in autism: the cerebellar vermis. Curr Opin Neurol. 1995a;8(2):134–138. [PubMed]
  • Filipek P. Neuroimaging in Autism: the state of the science. J Autism Dev Disord. 1995b;26(2):211–215. [PubMed]
  • Fombonne E. Epidemiological surveys of autism and other pervasive developmental disorders: an update. J Aut Dev Dis. 2003;33:365–382. [PubMed]
  • Ghanizadeh A. Methylphenidate improves response inhibition but not reflection-impulsivity in children with Attention deficit hyperactivity disorder (AHDH) Psychopharmacology. 2009 [PubMed]
  • Gogtay N, Giedd JN, Lusk L, Hayashi KM, Greenstein D, Vaituzis AC, Nugent TF, 3rd, Herman DH, Clasen LS, Toga AW, Rapoport JL, Thompson PM. Dynamic mapping of human cortical development during childhood through early adulthood. Proc Natl Acad Sci. 2004;101(21):8174–8179. [PubMed]
  • Gogtay N, Ordonez A, Herman DH, Hayashi KM, Greenstein D, Vaituzis C, Lenane M, Clasen L, Sharp W, Giedd JN, Jung D, Nugent TF, 3rd, Toga AW, Leibenluft E, Thompson PM, Rapoport JL. Dynamic mapping of cortical development before and after the onset of pediatric bipolar illness. J Child Psychol Psychiatry. 2007;48(9):852–62. [PubMed]
  • Gogtay N, Lu A, Leow AD, Klunder AD, Lee AD, Chavez A, Greenstein D, Giedd JN, Toga AW, Rapoport JL, Thompson PM. 3D Growth pattern abnormalities visualized in childhood-onset schizophrenia using Tensor-Based Morphometry. Proc Nat Acad Sci. 2008 in press. [PubMed]
  • Good CD, Johnsrude IS, Ashburner J, Henson RNA, Friston KJ, Frackowiak RSJ. Cerebral asymmetry and the effects of sex and handedness on brain structure: a voxel-based morphometric analysis of 465 normal adult human brains. NeuroImage. 2001;14(3):685–700. [PubMed]
  • Good CD, Johnsrude IS, Ashburner J, Henson RNA, Friston KJ, Frackowiak RSJ. A voxel-based morphometric study of aging in 465 normal adult human brains. NeuroImage. 2002;14(1):21–36. [PubMed]
  • Gramkow C. Registration of 2D and 3D medical images. Danish Technical University; Copenhagen, Denmark: 1996.
  • Haas RH, Townsend J, Courchesne E, Lincoln AJ, Schreibman L, Yeung-Courchesne R. Neurologic Abnormalities in infantile Autism. Journal of child neurology. 1996;11(2):84–92. [PubMed]
  • Hardan AY, Muddasani S, Vemulapalli M, Keshavan MS, Minshew NJ. An MRI study of increased cortical thickness in autism. Am J Psychiatry. 2006;163(7):1290–1292. [PMC free article] [PubMed]
  • Hazlett HC, Poe MD, Gerig G, Smith RG, Provenzale J, Ross A, Gilmore J, Piven J. Magnetic resonance imaging and head circumference study of brain size in autism: birth through age 2 years. Arch Gen Psychiatry. 2005;62(12):1366–1376. [PubMed]
  • Hazlett HC, Poe MD, Gerig G, Smith RG, Piven J. Cortical gray matter and white tissue volume in adolescent and adults with autism. Biol Psychiatry. 2006;59(1):1–6. [PubMed]
  • Herbert MR, Ziegler DA, Deustch CK, O’Brien LM, Lange N, Bakardjiev A, Hodgson J, Adrien KT, Steele S, Makris N, Kennedy D, Harris GJ, Caviness VS., Jr Dissociations of cerebral cortex, subcortical and cerebral white matter volumes in autistic boys. Brain. 2003;126(5):1182–1192. [PubMed]
  • Herbert MR, Ziegler DA, Makris N, Filipek PA, Kemper TL, Normandin JJ, Sanders HA, Kennedy DN, Caviness VS., Jr Localization of white matter volume increase in autism and developmental language disorder. Ann Neurol. 2004;55(4):530–540. [PubMed]
  • Herbert MR. Large brain in autism: the challenge of pervasive abnormality. Neuroscientist. 2005;11(5):417–440. [PubMed]
  • Hollingshead AB. Four Factor Analysis of Social Status. New Haven, CT: Yale University Department of Sociology; 1975.
  • Hua X, Leow AD, Lee S, Klunder AD, Toga AW, Leporé N, Chou YY, Chiang M-C, Barysheva M, Jack CR, Jr, Bernstein MA, Britson PJ, Gunter JL, Ward CP, Whitwell JL, Borowski B, Fleisher AS, Fox NC, Boyes R, Barnes J, Harvey D, Kornak J, Schuff N, Boreta L, Studholme C, Alexander GE, Weiner MW, Thompson PM. 3D Characterization of Brain Atrophy in Alzheimer’s disease and mild cognitive impairment using Tensor-Based Morphometry. NeuroImage. 2008;41(1):19–34. [PMC free article] [PubMed]
  • Hua X, Leow AD, Levitt JG, Caplan R, Thompson PM, Toga AW. Detecting Brain Growth Patterns in Normal Children using Tensor-Based Morphometry. Human Brain Mapping. 2007 December 6 [PMC free article] [PubMed]
  • Jernigan TL, Gamst AC, Archibald SL, Fennema-Notestine C, Mindt MR, Marcotte TD, Heaton RK, Ellis RJ, Grant I. Effects of methamphetamine dependence and HIV infection on cerebral morphology. Am J Psychiatry. 2005;162(8):1461–1472. [PubMed]
  • Joshi S, Davis B, Jomier M, Gerig G. Unbiased diffeomorphic atlas construction for computational anatomy. NeuroImage. 2004;23(1):151–160. [PubMed]
  • Kaufman J, Birmnaher B, Brent D, Rao U, Flynn C, Moreci P, Williamson D, Ryan N. Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version (K-SADS-PL): Initial reliability and validity data. J Am Acad Child Adolesc Psychiatry. 1997;36(7):980–988. [PubMed]
  • Kanner L. Follow-up study of eleven autistic children originally reported in 1943. 1971, La Psychiatrie de l’Enfant. 1995;38(2):421–461. [PubMed]
  • Kochunov P, Lancaster JL, Thompson PM, Woods R, Mazziotta JC, Hardies J, Fox P. Regional spatial normalization: toward an optimal target. J Comput Assist Tomogr. 2001;25(5):805–816. [PubMed]
  • Kochunov P, Lancaster JL, Thompson PM, Toga AW, Brewer P, Hardies J, Fox P. An optimized individual target brain in the Talairach coordinate system. NeuroImage. 2002;17(2):922–927. [PubMed]
  • Lee AD, Leow AD, Lu A, Reiss AL, Hall S, Chiang MC, Toga AW, Thompson PM. 3D Pattern of Brian Abnormalities in Fragile X Syndrome visualized using Tensor-Based Morphometry. NeuroImage. 2007;34(3):924–938. [PMC free article] [PubMed]
  • Lenroot RK, Gogtay N, Greenstein DK, Wells EM, Wallace GL, Clasen LS, Blumenthal JD, Lerch J, Zijdenbos AP, Evans AC, Thompson PM, Giedd JN. Sexual dimorphism of brain developmental trajectories during childhood and adolescence. NeuroImage. 2007;36(4):1065–1073. [PMC free article] [PubMed]
  • Leow AD, Klunder AD, Jack CR, Jr, Toga AW, Dale AM, Bernstein MA, Britson PJ, Gunter JL, Ward CP, Whitwell JL, Borowski BJ, Fleisher AS, Fox NC, Harvey D, Kornak J, Schuff N, Studholme C, Alexander GE, Weiner MW, Thompson PM. Longitudinal stability of MRI for mapping brain change using tensor-based morphometry. NeuroImage. 2006;31(2):627–640. [PMC free article] [PubMed]
  • Leporé N, Brun C, Pennec X, Chou Y-Y, Lopez OL, Aizenstein HJ, Toga AW, Becker JT, Thompson PM. Mean Template for Tensor-Based Morphometry using Deformation Tensors. Proceedings, 10th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI); Brisbane, Australia. Oct 29–Nov 2; 2007. [PubMed]
  • Leporé N, Brun C, Chou Y-Y, Chiang M-C, Dutton RA, Hayashi KM, Lu A, Lopez OL, Aizenstein HJ, Toga AW, Becker JT, Thompson PM. Generalized Tensor-Based Morphometry, of HIV/AIDS Using Multivariate Statistics on Deformation Tensors . Trans Med Im. 2008a;27(1):129–141. [PMC free article] [PubMed]
  • Leporé N, Chou Y-Y, Lopez OL, Aizenstein HJ, Becker JT, Toga AW, Thompson PM. Fast 3D fluid registration of brain magnetic resonance images. SPIE symposium on Medical Imaging; San Diego, CA, USA. Feb 16–21; 2008b.
  • Lerch JP, Evans AC. Cortical thickness analysis examined through power analysis and a population simulation. Neuroimage. 24(1):163–173. [PubMed]
  • Levitt JG, Blanton R, Capetillo-Cunliffe L, Guthrie D, Toga AW, McCracken JT. Cerebellar vermis lobules VIII–X in autism. Prog Neuropsychopharmacol Biol Psychiatry. 1999;23(4):625–633. [PubMed]
  • Lord C, Rutter M, Le Cuteur A. Autism Diagnostic Interview-Revised: a revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. J Autism Dev Dis. 1994;24:659–685. [PubMed]
  • Lord C, Risi S, Lambrecht L, Cook EH, Jr, Leventhal BL, DiLavore PC, Pickles A, Rutter M. The autism diagnostic observation schedule-generic: A standard measure of social and communication deficits associated with the spectrum of autism. J Autism Dev Dis. 2000;30:205–223. [PubMed]
  • Lorenzen P, Davis BM, Gerig G, Bullitt E, Joshi S. Multi-class posterior atlas formation via unbiased Kullback-Leibler template estimation. Proceedings, 8th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI); Rennes Saint-Malo, France. Sept 26–30; 2004.
  • McAlonan GM, Cheung V, Cheung C, Suckling J, Lam GY, Tai KS, Yip L, Murphy DG, Chua SE. Mapping the brain in autism. A voxel-based MRI study of volumetric differences and intercorrelations in autism. Brain. 2005;128(2):268–276. [PubMed]
  • Milne E, Swettenham J, Hansen P, Campbell R, Jeffries H, Plaisted K. High motion coherence thresholds in children with autism. J Child Psychol Psychiatry. 2002;43(2):255–263. [PubMed]
  • Nichols TE, Holmes AP. Non parametric permutation tests for functional neuroimaging: a primer with examples. Hum Brain Mapp. 2002;15(1):1–25. [PubMed]
  • Nicolson R, Szatmari P. Genetic and neurodevelopmental influence in Autistic Disorder. Can J Psychiatry. 2003;48(8):27–38. [PubMed]
  • Nicolson R, DeVito TJ, Vidal CN, Sui Y, Hayashi KM, Drost DJ, Williamson PC, Rajakumar N, Toga AW, Thompson PM. Detection and Mapping of Hippocampal Abnormalities in Autism. Psychiatry Res. 2006;148(1):11–21. [PubMed]
  • Pennec X. Probabilities and Statistics on Riemannian Manifolds: A Geometric Approach. Research Report INRIA. 2004:5093.
  • Redcay E, Courchesne E. When is the brain enlarged in autism? A meta analysis of all brain size reports. Biol Pyschiatry. 2005;58(1):1–9. [PubMed]
  • Shattuck DW, Leahy RM. BrainSuite: an automated cortical surface identification tool. Med Image Anal. 2002;6(2):129–142. [PubMed]
  • Shaw P, Greenstein D, Lerch J, Clasen L, Lenroot R, Gogtay N, Evans A, Rapoport J, Giedd J. Intellectual ability and cortical development in children and adolescents. Nature. 2006;440(7084):676–679. [PubMed]
  • Shaw P, Kabani NJ, Lerch JP, Eckstrand K, Lenroot R, Gogtay N, Greenstein D, Clasen L, Evans A, Rapoport JL, Giedd JN, Wise SP. Neurodevelopmental trajectories of the human cerebral cortex. J Neurosci. 2008;28(4):3586–3594. [PubMed]
  • Sled JG, Zijdenbos AP, Evans AC. A nonparametric method for automatic correction of intensity, nonuniformity in MRI data. IEEE Trans Med Im. 1998;17(1):87–97. [PubMed]
  • Sowell ER, Thompson PM, Holmes CJ, Batth R, Jernigan TL, Toga AW. Localizing age-related changes in brain structure between childhood and adolscence using statistical parametric, mapping. NeuroImage. 1999;9(6):987–997. [PubMed]
  • Stanfield AC, McIntosh AM, Spencer MD, Philip R, Gaur S, Lawrie SM. European Psychiatry. Towards a neuroanatomy of autism: A systematic review and meta-analysis of structural magnetic resonance imaging studies. in press. [PubMed]
  • Studholme C, Cardenas V, Maudsley A, Weiner M. An intensity consistent filtering approach to the analysis of deformation tensor derived maps of brain shape. NeuroImage. 2003;19(4):1638–1649. [PubMed]
  • Sun T, Walsh CA. Molecular approaches to brain asymmetry and handedness. Nat Rev Neurosci. 2006;7(8):655–662. [PubMed]
  • Thacker NA. Tutorial: A critical analysis of Voxel Based Morphometry (VBM) Internal Memo. 2005
  • Thompson PM, Giedd JN, Woods RP, MacDonald D, Evans AC, Toga AW. Growth Patterns in the Developing Brain Detected By Using Continuum-Mechanical Tensor Maps. Nature. 2000;404(6774):190–193. [PubMed]
  • Thompson PM, Hayashi KM, Sowell ER, Gogtay N, Giedd JN, Rapoport JL, de Zubicaray GI, Janke AL, Rose SE, Semple J, Doddrell DM, Wang YL, van Erp TGM, Cannon TD, Toga AW. Mapping Cortical Change in Alzheimer’s Disease, Brain Development, and Schizophrenia, Special Issue on Mathematics in Brain Imaging. NeuroImage. 2004a;23(1):2–18. [PubMed]
  • Thompson PM, Hayashi KM, de Zubicaray GI, Janke AL, Rose SE, Semple J, Hong MS, Herman D, Gravano D, Doddrell DM, Toga AW. Mapping Hippocampal and Ventricular Change in Alzheimer’s Disease. NeuroImage. 2004b;22(4):1754–66. [PubMed]
  • Thompson PM, Sowell ER, Gogtay N, Giedd JN, Vidal CN, Hayashi KM, Leow A, Nicolson R, Rapoport JL, Toga AW. Structural MRI and brain development. Int Rev Neurobiol. 2005a;67:285–323. [PubMed]
  • Thompson PM, Lee AD, Dtton RA, Geaga JA, Hayashi KM, Eckert MA, Bellugi U, Galaburda AM, Korenberg JR, Mills DL, Toga AW, Reiss AL. Abnormal cortical complexity and thickness profiles mapped in Williams Syndrome. J Neurosci. 2005b;25(16):4146–4158. [PubMed]
  • Van der Veere CN, Bos AS. The use of SSRIs during pregnancy and possible consequences for the development of the child. Ne Tijdschr Geneeskd. 2007;151(52):2873–2874. [PubMed]
  • Vargas DL, Nascimbene C, Krishnan C, Zimmerman AW, Pardo CA. Neuroglial activation and neuroinflammation in the brain of patients with autism. Ann Neurol. 2005;57(1):67–81. [PubMed]
  • Vidal CN, Nicolson R, DeVito TJ, Hayashi KM, Geaga JA, Drost DJ, Williamson PC, Rajakumar N, Sui Y, Dutton RA, Toga AW, Thompson PM. Mapping Corpus Callosum Deficits in autism: an index of aberrant cortical connectivity. Biol Psychiatry. 2006;60(3):218–225. [PubMed]
  • Vidal CN, Nicolson R, Boire JY, Barra V, DeVito TJ, Hayashi KM, Geaga JA, Drost DJ, Williamson PC, Rajakumar N, Toga AW, Thompson PM. Three-Dimensional Mapping, of the Lateral Ventricles in Autism. Psychiatry Res. 2008;163(2):106–115. [PMC free article] [PubMed]
  • Yushkevich PA, Piven J, Hazlett HC, Smith RG, Ho S, Gee JC, Gerig G. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. NeuroImage. 2006;31(3):1116–1128. [PubMed]
  • Williams DL, Minshew NJ. Understanding autism and related disorders: what has imaging taught us? Neuroimaging Clin N Am. 2007;17(4):495–509. [PMC free article] [PubMed]
  • Woods RP. Multitracer: A Java-based tool for anatomic delineation of grayscale volumetric images. NeuroImage. 2003;19(4):1829–1834. [PubMed]