Caroline C. Brun

Brain Structure Changes Visualized in Early- and Late-Onset Blind Subjects

We examined 3D patterns of volume differences in the brain associated with blindness, in subjects grouped according to early and late onset. Using tensor-based morphometry, we mapped volume reductions and gains in 16 early-onset (EB) and 16 late-onset (LB) blind adults (onset <5 and >14 years old, respectively) relative to 16 matched sighted controls. Each subjects structural MRI was fluidly registered to a common template. Anatomical differences between groups were mapped based on statistical analysis of the resulting deformation fields revealing profound deficits in primary and secondary visual cortices for both blind groups. Regions outside the occipital lobe showed significant hypertrophy, suggesting widespread compensatory adaptations. EBs but not LBs showed deficits in the splenium and the isthmus. Gains in the non-occipital white matter were more widespread in the EBs. These differences may reflect regional alterations in late neurodevelopmental processes, such as myelination, that continue into adulthood.

Sex differences in brain structure in auditory and cingulate regions

We applied a new method to visualize the three-dimensional profile of sex differences in brain structure based on MRI scans of 100 young adults. We compared 50 men with 50 women, matched for age and other relevant demographics. As predicted, left hemisphere auditory and language-related regions were proportionally expanded in women versus men, suggesting a possible structural basis for the widely replicated sex differences in language processing. In men, primary visual, and visuo-spatial association areas of the parietal lobes were proportionally expanded, in line with prior reports of relative strengths in visuo-spatial processing in men. We relate these three-dimensional patterns to prior functional and structural studies, and to theoretical predictions based on nonlinear scaling of brain morphometry.

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. 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. Hum Brain Mapp, 2009.

A Tensor-Based Morphometry Study of Genetic Influences on Brain Structure Using a New Fluid Registration Method

We incorporated a new Riemannian fluid registration algorithm into a general MRI analysis method called tensor-based morphometry to map the heritability of brain morphology in MR images from 23 monozygotic and 23 dizygotic twin pairs. All 92 3D scans were fluidly registered to a common template. Voxelwise Jacobian determinants were computed from the deformation fields to assess local volumetric differences across subjects. Heritability maps were computed from the intraclass correlations and their significance was assessed using voxelwise permutation tests. Lobar volume heritability was also studied using the ACE genetic model. The performance of this Riemannian algorithm was compared to a more standard fluid registration algorithm: 3D maps from both registration techniques displayed similar heritability patterns throughout the brain. Power improvements were quantified by comparing the cumulative distribution functions of the p-values generated from both competing methods. The Riemannian algorithm outperformed the standard fluid registration.

3D Mapping of brain differences in native signing congenitally and prelingually deaf subjects

In the prelingual and congenital deaf, functional reorganization is known to occur throughout brain regions normally associated with hearing. However, the anatomical correlates of these changes are not yet well understood. Here, we perform the first tensor‐based morphometric analysis of voxel‐wise volumetric differences in native signing prelingual and congenitally deaf subjects when compared with hearing controls. We obtained T1‐weighted scans for 14 native signing prelingual and congenitally deaf subjects and 16 age‐ and gender‐matched controls. We used linear and fluid registration to align each image to a common template. Using the voxel‐wise determinant of the Jacobian of the fluid deformation, significant volume increases, of up to 20%, were found in frontal lobe white matter regions including Brocas area, and adjacent regions involved in motor control and language production. A similar analysis was performed on hand‐traced corpora callosa. A strong trend for group differences was found in the area of the splenium considered to carry fibers connecting the temporal (and occipital) lobes. These anatomical differences may reflect experience‐mediated developmental differences in myelination and cortical maturation associated with prolonged monomodal sensory deprivation. Hum Brain Mapp, 2010.

Best individual template selection from deformation tensor minimization

We study the influence of the choice of template in tensor- based morphometry. Using 3D brain MR images from 10 monozygotic twin pairs, we defined a tensor-based distance in the log-Euclidean framework [1] between each image pair in the study. Relative to this metric, twin pairs were found to be closer to each other on average than random pairings, consistent with evidence that brain structure is under strong genetic control. We also computed the intraclass correlation and associated permutation p-value at each voxel for the determinant of the Jacobian matrix of the transformation. The cumulative distribution function (cdf) of the p-values was found at each voxel for each of the templates and compared to the null distribution. Surprisingly, there was very little difference between CDFs of statistics computed from analyses using different templates. As the brain with least log-Euclidean deformation cost, the mean template defined here avoids the blurring caused by creating a synthetic image from a population, and when selected from a large population, avoids bias by being geometrically centered, in a metric that is sensitive enough to anatomical similarity that it can even detect genetic affinity among anatomies.

Brain Differences Visualized in the Blind Using Tensor Manifold Statistics and Diffusion Tensor Imaging

Diffusion tensor magnetic resonance imaging (DTI) reveals the local orientation and integrity of white matter fiber structure based on imaging multidirectional water diffusion. Group differences in DTI images are often computed from single scalar measures, e.g., the Fractional Anisotropy (FA), discarding much of the information in the 6-parameter symmetric diffusion tensor. Here, we compute multivariate 6D tensor statistics to detect brain morphological changes in 12 blind subjects versus 14 sighted controls. After Log-Euclidean tensor de- noising, images were fluidly registered to a common template. Fluidly-convected tensor signals were re-oriented by applying the local rotational and translational component of the deformation. Since symmetric, positive- definite matrices form a non-Euclidean manifold, we applied a Riemannian manifold version of the Hotellings T2 test to the logarithms of the tensors, using a log- Euclidean metric. Statistics on the full 6D tensor-valued images outperformed univariate analysis of scalar images, such as the FA and the geodesic anisotropy (GA).

A new registration method based on Log-Euclidean Tensor metrics and its application to genetic studies

In structural brain MRI, group differences or changes in brain structures can be detected using tensor-based morphometry (TBM). This method consists of two steps: (1) a non-linear registration step, that aligns all of the images to a common template, and (2) a subsequent statistical analysis. The numerous registration methods that have recently been developed differ in their detection sensitivity when used for TBM, and detection power is paramount in epidemological studies or drug trials. We therefore developed a new fluid registration method that computes the mappings and performs statistics on them in a consistent way, providing a bridge between TBM registration and statistics. We used the log-Euclidean framework to define a new regularizer that is a fluid extension of the Riemannian elasticity, which assures diffeomorphic transformations. This regularizer constrains the symmetrized Jacobian matrix, also called the deformation tensor. We applied our method to an MRI dataset from 40 fraternal and identical twins, to revealed voxelwise measures of average volumetric differences in brain structure for subjects with different degrees of genetic resemblance.

Multivariate variance-components analysis in DTI

Twin studies are a major research direction in imaging genetics, a new field, which combines algorithms from quantitative genetics and neuroimaging to assess genetic effects on the brain. In twin imaging studies, it is common to estimate the intraclass correlation (ICC), which measures the resemblance between twin pairs for a given phenotype. In this paper, we extend the commonly used Pearson correlation to a more appropriate definition, which uses restricted maximum likelihood methods (REML). We computed proportion of phenotypic variance due to additive (A) genetic factors, common (C) and unique (E) environmental factors using a new definition of the variance components in the diffusion tensor-valued signals. We applied our analysis to a dataset of Diffusion Tensor Images (DTI) from 25 identical and 25 fraternal twin pairs. Differences between the REML and Pearson estimators were plotted for different sample sizes, showing that the REML approach avoids severe biases when samples are smaller. Measures of genetic effects were computed for scalar and multivariate diffusion tensor derived measures including the geodesic anisotropy (tGA) and the full diffusion tensors (DT), revealing voxel-wise genetic contributions to brain fiber microstructure.

A Lagrangian formulation for statistical fluid registration

We defined a new statistical fluid registration method with Lagrangian mechanics. Although several authors have suggested that empirical statistics on brain variation should be incorporated into the registration problem, few algorithms have included this information and instead use regularizers that guarantee diffeomorphic mappings. Here we combine the advantages of a large-deformation fluid matching approach with empirical statistics on population variability in anatomy. We reformulated the Riemannian fluid algorithmdeveloped in [4], and used a Lagrangian framework to incorporate 0th and 1st order statistics in the regularization process. 92 2D midline corpus callosum traces from a twin MRI database were fluidly registered using the non-statistical version of the algorithm (algorithm 0), giving initial vector fields and deformation tensors. Covariance matrices were computed for both distributions and incorporated either separately (algorithm 1 and algorithm 2) or together (algorithm 3) in the registration. We computed heritability maps and two vector and tensorbased distances to compare the power and the robustness of the algorithms.