Automatically parcellating the human cerebral cortex

Geometry driven volumetric registration

In this paper, we propose a novel method for the registration of volumetric images of the brain that attempts to maximize the overlap of cortical folds. In order to achieve this, relevant geometrical information is extracted from a surface-based morph and is diffused throughout the volume using the Navier operator of elasticity. The result is a volumetric warp that aligns the folding patterns.

Effects of registration regularization and atlas sharpness on segmentation accuracy

In this paper, we propose a unified framework for computing atlases from manually labeled data at various degrees of "sharpness" and the joint registration-segmentation of a new brain with these atlases. In non-rigid registration, the tradeoff between warp regularization and image fidelity is typically set empirically. In segmentation, this leads to a probabilistic atlas of arbitrary "sharpness": weak regularization results in well-aligned training images and a "sharp" atlas; strong regularization yields a "blurry" atlas. We study the effects of this tradeoff in the context of cortical surface parcellation by comparing three special cases of our framework, namely: progressive registration-segmentation of a new brain to increasingly "sharp" atlases with increasingly flexible warps; secondly, progressive registration to a single atlas with increasingly flexible warps; and thirdly, registration to a single atlas with fixed constrained warps. The optimal parcellation in all three cases corresponds to a unique balance of atlas "sharpness" and warp regularization that yield statistically significant improvements over the previously demonstrated parcellation results.

Active mean fields: solving the mean field approximation in the level set framework

We describe a new approach for estimating the posterior probability of tissue labels. Conventional likelihood models are combined with a curve length prior on boundaries, and an approximate posterior distribution on labels is sought via the Mean Field approach. Optimizing the resulting estimator by gradient descent leads to a level set style algorithm where the level set functions are the logarithm-of-odds encoding of the posterior label probabilities in an unconstrained linear vector space. Applications with more than two labels are easily accommodated. The label assignment is accomplished by the Maximum A Posteriori rule, so there are no problems of "overlap" or "vacuum". We test the method on synthetic images with additive noise. In addition, we segment a magnetic resonance scan into the major brain compartments and subcortical structures.

Statistical neuroanatomy of the human inferior frontal gyrus and probabilistic atlas in a standard stereotaxic space

We manually defined the inferior frontal gyrus (IFG) on high-resolution MRIs in native space in 30 healthy subjects (15 female, median age 31 years; 15 male, median age 30 years), resulting in 30 individual atlases. Using standard software (SPM99), these were spatially transformed to a widely used stereotaxic space (MNI/ICBM 152) to create probabilistic maps. In native space, the total IFG volume was on average 5%, and the gray matter (GM) portion 12% larger in women (not significant). Expressed as a percentage of ipsilateral frontal lobe volume (i.e., correcting for brain size), the IFG was an average of 20%, and the GM portion of the IFG 27%, larger in women (P < 0.005). Correcting for total lobar volume yielded the same result. No asymmetry was found in IFG volumes. There were significant positional differences between the right and left IFGs, with the right IFG being further lateral in both native and stereotaxic space. Variability was similar on the left and right, but more pronounced anteriorly and superiorly. We show differences in IFG volume, composition, and position between sexes and between hemispheres. Applications include probabilistic determination of location in group studies, automatic labeling of new scans, and detection of anatomical abnormalities in patients.

Spatial dynamics of masked picture repetition effects

The aim of this study was to further elucidate the mechanisms of early and automatic object processing using a masked picture priming paradigm with both identity and exemplar repetition in functional MRI (fMRI). Masked repetition priming has been commonly used with words to isolate automatic, rapidly occurring mechanisms involved in visual word recognition; however, studies using the technique of masked priming with rapid presentation of pictures have been limited. This study demonstrates how the masked priming technique can be used to study early, automatic processing of rapidly presented complex objects. Temporal-occipital regions previously found to be sensitive to repetition priming in both masked word and unmasked picture studies were found to show repetition suppression for the identity primes only. Most notably, when divided into anterior and posterior divisions, the fusiform gyrus showed anatomically specific repetition suppression only in the posterior portion. This finding is comparable to that found in a previous study of masked word priming, and the similarity may suggest an overlap in the early identification processes for visual word form and visual object processing in this region. Finally, masked repetition of different exemplar objects did not result in reliable neural effects, suggesting that the underlying mechanisms of the more semantic-based, different exemplar priming may occur later or require the intervention of conscious processes.

Regional gray matter volumetric changes in autism associated with social and repetitive behavior symptoms

BACKGROUND

Although differences in brain anatomy in autism have been difficult to replicate using manual tracing methods, automated whole brain analyses have begun to find consistent differences in regions of the brain associated with the social cognitive processes that are often impaired in autism. We attempted to replicate these whole brain studies and to correlate regional volume changes with several autism symptom measures.

METHODS

We performed MRI scans on 24 individuals diagnosed with DSM-IV autistic disorder and compared those to scans from 23 healthy comparison subjects matched on age. All participants were male. Whole brain, voxel-wise analyses of regional gray matter volume were conducted using voxel-based morphometry (VBM).

RESULTS

Controlling for age and total gray matter volume, the volumes of the medial frontal gyri, left pre-central gyrus, right post-central gyrus, right fusiform gyrus, caudate nuclei and the left hippocampus were larger in the autism group relative to controls. Regions exhibiting smaller volumes in the autism group were observed exclusively in the cerebellum. Significant partial correlations were found between the volumes of the caudate nuclei, multiple frontal and temporal regions, the cerebellum and a measure of repetitive behaviors, controlling for total gray matter volume. Social and communication deficits in autism were also associated with caudate, cerebellar, and precuneus volumes, as well as with frontal and temporal lobe regional volumes.

CONCLUSION

Gray matter enlargement was observed in areas that have been functionally identified as important in social-cognitive processes, such as the medial frontal gyri, sensorimotor cortex and middle temporal gyrus. Additionally, we have shown that VBM is sensitive to associations between social and repetitive behaviors and regional brain volumes in autism.

Task-dependent changes in brain activation following therapy for nonfluent aphasia: discussion of two individual cases

The complex process of cortical reorganization of language-related brain regions during recovery from aphasia and the effects of therapeutic interventions on brain systems are poorly understood. We studied two patients with chronic aphasia and compared their functional neuroanatomical responses to a younger control group on two tasks, an oral-reading task involving overt speech and a "passive" audiovisual story-comprehension task. Following identical therapy, we re-examined behavioral (language) and functional neuroanatomical changes using the same functional magnetic resonance imaging (fMRI) tasks. We hypothesized that better recovery would be associated with brain activation patterns more closely resembling healthy controls, whereas positive responses to language treatment would be associated with increased activity in undamaged left perisylvian areas and/or right-hemisphere areas homologous to the damaged regions. For the participant with a frontal lesion who was most responsive to therapy, brain activation increased in the right hemisphere during oral-reading, but decreased bilaterally in most regions on story-comprehension. The other participant with a temporal-parietal lesion showed decreased activation, particularly in the right hemisphere, during oral-reading but increased activation bilaterally on story-comprehension. Results highlight individual variability following language therapy, with brain activation changes depending on lesion site and size, language skill, type of intervention, and the nature of the fMRI task.

Dealing with the shortcomings of spatial normalization: multi-subject parcellation of fMRI datasets

The analysis of functional magnetic resonance imaging (fMRI) data recorded on several subjects resorts to the so-called spatial normalization in a common reference space. This normalization is usually carried out on a voxel-by-voxel basis, assuming that after coregistration of the functional images with an anatomical template image in the Talairach reference system, a correct voxel-based inference can be carried out across subjects. Shortcomings of such approaches are often dealt with by spatially smoothing the data to increase the overlap between subject-specific activated regions. This procedure, however, cannot adapt to each anatomo-functional subject configuration. We introduce a novel technique for intra-subject parcellation based on spectral clustering that delineates homogeneous and connected regions. We also propose a hierarchical method to derive group parcels that are spatially coherent across subjects and functionally homogeneous. We show that we can obtain groups (or cliques) of parcels that well summarize inter-subject activations. We also show that the spatial relaxation embedded in our procedure improves the sensitivity of random-effect analysis.

Making sense of discourse: An fMRI study of causal inferencing across sentences

To build up coherence between sentences (comprehend discourse), we must draw inferences, i.e. activate and integrate information that is not actually stated. We used event-related fMRI to determine the localization and extent of brain activity mediating causal inferencing across short, three-sentence scenarios. Participants read and made causal coherence judgments to sentences that were highly causally related, intermediately related or unrelated to their preceding two-sentence contexts. The highly related and intermediately related scenarios were matched in terms of semantic similarities between their individual component words. A pre-rating study established that causal inferences were generated to the intermediately related but not to the highly related or unrelated scenarios. In the scanner, sentences that were intermediately related (relative to highly related or unrelated) to their preceding contexts were associated with longer judgment reaction times and sustained increases in hemodynamic activity within left lateral temporal/inferior parietal/prefrontal cortices, the right inferior prefrontal gyrus and bilateral superior medial prefrontal cortices. In contrast, sentences that were unrelated (relative to highly related) to their preceding contexts were associated with only transient increases in activity (at, but not after, the peak of the hemodynamic response) within the right lateral temporal cortex and the right inferior prefrontal gyrus. These data suggest that, to make sense of discourse, we activate a large bilateral cortical network in response to what is not explicitly stated. We suggest that this network reflects the activation, retrieval and integration of information from long-term semantic memory into incoming discourse structure during causal inferencing.

Automatic anatomical brain MRI segmentation combining label propagation and decision fusion

Regions in three-dimensional magnetic resonance (MR) brain images can be classified using protocols for manually segmenting and labeling structures. For large cohorts, time and expertise requirements make this approach impractical. To achieve automation, an individual segmentation can be propagated to another individual using an anatomical correspondence estimate relating the atlas image to the target image. The accuracy of the resulting target labeling has been limited but can potentially be improved by combining multiple segmentations using decision fusion. We studied segmentation propagation and decision fusion on 30 normal brain MR images, which had been manually segmented into 67 structures. Correspondence estimates were established by nonrigid registration using free-form deformations. Both direct label propagation and an indirect approach were tested. Individual propagations showed an average similarity index (SI) of 0.754+/-0.016 against manual segmentations. Decision fusion using 29 input segmentations increased SI to 0.836+/-0.009. For indirect propagation of a single source via 27 intermediate images, SI was 0.779+/-0.013. We also studied the effect of the decision fusion procedure using a numerical simulation with synthetic input data. The results helped to formulate a model that predicts the quality improvement of fused brain segmentations based on the number of individual propagated segmentations combined. We demonstrate a practicable procedure that exceeds the accuracy of previous automatic methods and can compete with manual delineations.

Vector-based spatial–temporal minimum L1-norm solution for MEG

Minimum L1-norm solutions have been used by many investigators to analyze MEG responses because they provide high spatial resolution images. However, conventional minimum L1-norm approaches suffer from instability in spatial construction, and poor smoothness of the reconstructed source time-courses. Activity commonly "jumps" from one grid point to (usually) the neighboring grid points. Equivalently, the time-course of one specific grid point can show substantial "spiky-looking" discontinuity. In the present study, we present a new vector-based spatial-temporal analysis using a L1-minimum-norm (VESTAL). This approach is based on a principle of MEG physics: the magnetic waveforms in sensor-space are linear functions of the source time-courses in the imaging-space. Our computer simulations showed that VESTAL provides good reconstruction of the source amplitude and orientation, with high stability and resolution in both the spatial and temporal domains. "Spiky-looking" discontinuity was not observed in the source time-courses. Importantly, the simulations also showed that VESTAL can resolve sources that are 100% correlated. We then examined the performance of VESTAL in the analysis of human median-nerve MEG responses. The results demonstrated that this method easily distinguishes sources very spatially close to each other, including individual primary somatosensory areas (BA 1, 2, 3b), primary motor area (BA 4), and other regions in the somatosensory system (e.g., BA 5, 7, SII, SMA, and temporal-parietal junction) with high temporal stability and resolution. VESTAL's potential for obtaining information on source extent was also examined.

An fMRI investigation of syllable sequence production

Fluent speech comprises sequences that are composed from a finite alphabet of learned words, syllables, and phonemes. The sequencing of discrete motor behaviors has received much attention in the motor control literature, but relatively little has been focused directly on speech production. In this paper, we investigate the cortical and subcortical regions involved in organizing and enacting sequences of simple speech sounds. Sparse event-triggered functional magnetic resonance imaging (fMRI) was used to measure responses to preparation and overt production of non-lexical three-syllable utterances, parameterized by two factors: syllable complexity and sequence complexity. The comparison of overt production trials to preparation only trials revealed a network related to the initiation of a speech plan, control of the articulators, and to hearing one's own voice. This network included the primary motor and somatosensory cortices, auditory cortical areas, supplementary motor area (SMA), the precentral gyrus of the insula, and portions of the thalamus, basal ganglia, and cerebellum. Additional stimulus complexity led to increased engagement of the basic speech network and recruitment of additional areas known to be involved in sequencing non-speech motor acts. In particular, the left hemisphere inferior frontal sulcus and posterior parietal cortex, and bilateral regions at the junction of the anterior insula and frontal operculum, the SMA and pre-SMA, the basal ganglia, anterior thalamus, and the cerebellum showed increased activity for more complex stimuli. We hypothesize mechanistic roles for the extended speech production network in the organization and execution of sequences of speech sounds.

Reliability of MRI-derived measurements of human cerebral cortical thickness: the effects of field strength, scanner upgrade and manufacturer

In vivo MRI-derived measurements of human cerebral cortex thickness are providing novel insights into normal and abnormal neuroanatomy, but little is known about their reliability. We investigated how the reliability of cortical thickness measurements is affected by MRI instrument-related factors, including scanner field strength, manufacturer, upgrade and pulse sequence. Several data processing factors were also studied. Two test-retest data sets were analyzed: 1) 15 healthy older subjects scanned four times at 2-week intervals on three scanners; 2) 5 subjects scanned before and after a major scanner upgrade. Within-scanner variability of global cortical thickness measurements was <0.03 mm, and the point-wise standard deviation of measurement error was approximately 0.12 mm. Variability was 0.15 mm and 0.17 mm in average, respectively, for cross-scanner (Siemens/GE) and cross-field strength (1.5 T/3 T) comparisons. Scanner upgrade did not increase variability nor introduce bias. Measurements across field strength, however, were slightly biased (thicker at 3 T). The number of (single vs. multiple averaged) acquisitions had a negligible effect on reliability, but the use of a different pulse sequence had a larger impact, as did different parameters employed in data processing. Sample size estimates indicate that regional cortical thickness difference of 0.2 mm between two different groups could be identified with as few as 7 subjects per group, and a difference of 0.1 mm could be detected with 26 subjects per group. These results demonstrate that MRI-derived cortical thickness measures are highly reliable when MRI instrument and data processing factors are controlled but that it is important to consider these factors in the design of multi-site or longitudinal studies, such as clinical drug trials.

An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest

In this study, we have assessed the validity and reliability of an automated labeling system that we have developed for subdividing the human cerebral cortex on magnetic resonance images into gyral based regions of interest (ROIs). Using a dataset of 40 MRI scans we manually identified 34 cortical ROIs in each of the individual hemispheres. This information was then encoded in the form of an atlas that was utilized to automatically label ROIs. To examine the validity, as well as the intra- and inter-rater reliability of the automated system, we used both intraclass correlation coefficients (ICC), and a new method known as mean distance maps, to assess the degree of mismatch between the manual and the automated sets of ROIs. When compared with the manual ROIs, the automated ROIs were highly accurate, with an average ICC of 0.835 across all of the ROIs, and a mean distance error of less than 1 mm. Intra- and inter-rater comparisons yielded little to no difference between the sets of ROIs. These findings suggest that the automated method we have developed for subdividing the human cerebral cortex into standard gyral-based neuroanatomical regions is both anatomically valid and reliable. This method may be useful for both morphometric and functional studies of the cerebral cortex as well as for clinical investigations aimed at tracking the evolution of disease-induced changes over time, including clinical trials in which MRI-based measures are used to examine response to treatment.

A Bayesian model for joint segmentation and registration

A statistical model is presented that combines the registration of an atlas with the segmentation of magnetic resonance images. We use an Expectation Maximization-based algorithm to find a solution within the model, which simultaneously estimates image artifacts, anatomical labelmaps, and a structure-dependent hierarchical mapping from the atlas to the image space. The algorithm produces segmentations for brain tissues as well as their substructures. We demonstrate the approach on a set of 22 magnetic resonance images. On this set of images, the new approach performs significantly better than similar methods which sequentially apply registration and segmentation.

Structural brain imaging in patients with cognitive impairment in the year 2015

Cognitive impairment, especially in its early stages, is associated with very mild signs and symptoms that are difficult to detect by clinical and neuropsychological assessment. Advanced imaging analysis techniques applied to magnetic resonance images allow the detection of cerebral structural changes in vivo in mildly affected patients, and might be a useful supporting tool in the early diagnosis and treatment of patients with cognitive impairment. The increasing importance of computer science in cognitive neuroscience has led to the dissemination of a new discipline, neuroinformatics, which is crucial for the introduction of research findings into clinical practice. This review describes some advanced imaging analysis techniques aimed at studying brain structural images and how these techniques might benefit clinical practice through image data sharing and remote analysis in order to increase the accuracy of diagnosis in patients with cognitive impairment.

Logarithm odds maps for shape representation

The concept of the Logarithm of the Odds (LogOdds) is frequently used in areas such as artificial neural networks, economics, and biology. Here, we utilize LogOdds for a shape representation that demonstrates desirable properties for medical imaging. For example, the representation encodes the shape of an anatomical structure as well as the variations within that structure. These variations are embedded in a vector space that relates to a probabilistic model. We apply our representation to a voxel based segmentation algorithm. We do so by embedding the manifold of Signed Distance Maps (SDM) into the linear space of LogOdds. The LogOdds variant is superior to the SDM model in an experiment segmenting 20 subjects into subcortical structures. We also use LogOdds in the non-convex interpolation between space conditioned distributions. We apply this model to a longitudinal schizophrenia study using quadratic splines. The resulting time-continuous simulation of the schizophrenic aging process has a higher accuracy then a model based on convex interpolation.

Rostral and dorsal anterior cingulate cortex make dissociable contributions during antisaccade error commission

The anterior cingulate cortex (ACC) participates in both performance optimization and evaluation, with dissociable contributions from dorsal (dACC) and rostral (rACC) regions. Deactivation in rACC and other default-mode regions is important for performance optimization, whereas increased rACC and dACC activation contributes to performance evaluation. Errors activate both rACC and dACC. We propose that this activation reflects differential error-related involvement of rACC and dACC during both performance optimization and evaluation, and that these two processes can be distinguished by the timing of their occurrence within a trial. We compared correct and error antisaccade trials. We expected errors to correlate with an early failure of rACC deactivation and increased activation of both rACC and dACC later in the trial. Eighteen healthy subjects performed a series of prosaccade and antisaccade trials during event-related functional MRI. We estimated the hemodynamic responses for error and correct antisaccades using a finite impulse-response model. We examined ACC activity by comparing error and correct antisaccades with a fixation baseline and error to correct antisaccades directly. Compared with correct antisaccades, errors were characterized by an early bilateral failure of deactivation of rACC and other default-mode regions. This difference was significant in rACC. Errors also were associated with increased activity in both rACC and dACC later in the trial. These results show that accurate performance involves deactivation of the rACC and other default mode regions and suggest that both rACC and dACC contribute to the evaluation of error responses.

Increased intracranial volume in Parkinson's disease

BACKGROUND

Parkinson's disease (PD) and multiple system atrophy (MSA) are neurodegenerative diseases that can be difficult to diagnose and distinguish from each other. STUDY AIMS AND METHODS: Patients with PD and MSA and controls were studied with magnetic resonance imaging (MRI) using tissue segmentation and outlining of regions in order to identify regional volume changes that might be useful in the diagnosis of the two diseases.

RESULTS

Patients with PD had significantly larger intracranial volumes (ICVs) and significantly smaller putaminal and sustantia nigra volumes than controls. MSA patients had significantly smaller substantia nigra and caudate volumes than controls but normal intracranial volume. In both patient groups there was a further trend towards smaller amygdala volumes.

DISCUSSION

Increased ICV in PD patients is a new finding that may be explained by genetic factors or compensatory responses to early CNS damage. Atrophy of the amygdala in MSA patients has not been demonstrated with MR before. It might explain why these patients can have hyposmia. The putaminal atrophy found in the PD group may be a trait of the later stages of PD. Segmentation of the substantia nigra can be a useful aid in the diagnosis of PD and MSA.

Mindboggle: automated brain labeling with multiple atlases

Background

To make inferences about brain structures or activity across multiple individuals, one first needs to determine the structural correspondences across their image data. We have recently developed Mindboggle as a fully automated, feature-matching approach to assign anatomical labels to cortical structures and activity in human brain MRI data. Label assignment is based on structural correspondences between labeled atlases and unlabeled image data, where an atlas consists of a set of labels manually assigned to a single brain image. In the present work, we study the influence of using variable numbers of individual atlases to nonlinearly label human brain image data.

Methods

Each brain image voxel of each of 20 human subjects is assigned a label by each of the remaining 19 atlases using Mindboggle. The most common label is selected and is given a confidence rating based on the number of atlases that assigned that label. The automatically assigned labels for each subject brain are compared with the manual labels for that subject (its atlas). Unlike recent approaches that transform subject data to a labeled, probabilistic atlas space (constructed from a database of atlases), Mindboggle labels a subject by each atlas in a database independently.

Results

When Mindboggle labels a human subject's brain image with at least four atlases, the resulting label agreement with coregistered manual labels is significantly higher than when only a single atlas is used. Different numbers of atlases provide significantly higher label agreements for individual brain regions.

Conclusion

Increasing the number of reference brains used to automatically label a human subject brain improves labeling accuracy with respect to manually assigned labels. Mindboggle software can provide confidence measures for labels based on probabilistic assignment of labels and could be applied to large databases of brain images.

Shape Based Segmentation of Anatomical Structures in Magnetic Resonance Images

Standard image based segmentation approaches perform poorly when there is little or no contrast along boundaries of different regions. In such cases, segmentation is largely performed manually using prior knowledge of the shape and relative location of the underlying structures combined with partially discernible boundaries. We present an automated approach guided by covariant shape deformations of neighboring structures, which is an additional source of prior information. Captured by a shape atlas, these deformations are transformed into a statistical model using the logistic function. Structure boundaries, anatomical labels, and image inhomogeneities are estimated simultaneously within an Expectation-Maximization formulation of the maximum a posteriori probability estimation problem. We demonstrate the approach on 20 brain magnetic resonance images showing superior performance, particularly in cases where purely image based methods fail.

Neuroimaging biomarkers for clinical trials of disease-modifying therapies in Alzheimer's disease

The pathophysiologic process leading to neurodegeneration in Alzheimer's disease (AD) is thought to begin long before clinical symptoms develop. Existing therapeutics for AD improve symptoms, but increasing efforts are being directed toward the development of therapies to impede the pathologic progression of the disease. Although these medications must ultimately demonstrate efficacy in slowing clinical decline, there is a critical need for biomarkers that will indicate whether a candidate disease-modifying therapeutic agent is actually altering the underlying degenerative process. A number of in vivo neuroimaging techniques, which can reliably and noninvasively assess aspects of neuroanatomy, chemistry, physiology, and pathology, hold promise as biomarkers. These neuroimaging measures appear to relate closely to neuropathological and clinical data, such as rate of cognitive decline and risk of future decline. As this work has matured, it has become clear that neuroimaging measures may serve a variety of potential roles in clinical trials of candidate neurotherapeutic agents for AD, depending in part on the question of interest and phase of drug development. In this article, we review data related to the range of neuroimaging biomarkers of Alzheimer's disease and consider potential applications of these techniques to clinical trials, particularly with respect to the monitoring of disease progression in trials of disease-modifying therapies.

Program for Assisted Labeling of Sulcal Regions (PALS): description and reliability

With the improvements in techniques for generating surface models from magnetic resonance (MR) images, it has recently become feasible to study the morphological characteristics of the human brain cortex in vivo. Studies of the entire surface are important for measuring global features, but analysis of specific cortical regions of interest provides a more detailed understanding of structure. We have previously developed a method for automatically segmenting regions of interest from the cortical surface using a watershed transform. Each segmented region corresponds to a cortical sulcus and is thus termed a "sulcal region." In this work, we describe two important augmentations of this methodology. First, we describe a user interface that allows for the efficient labeling of the segmented sulcal regions called the Program for Assisted Labeling of Sulcal Regions (PALS). An additional augmentation allows for even finer divisions on the cortex with a methodology that employs the fast marching technique to track a curve on the cortical surface that is then used to separate segmented regions. After regions of interest have been identified, we compute both the cortical surface area and gray matter volume. Reliability experiments are performed to assess both the long-term stability and short-term repeatability of the proposed techniques. These experiments indicate the proposed methodology gives both highly stable and repeatable results.

Computational anatomy: shape, growth, and atrophy comparison via diffeomorphisms

Computational anatomy (CA) is the mathematical study of anatomy I in I = I(alpha) o G, an orbit under groups of diffeomorphisms (i.e., smooth invertible mappings) g in G of anatomical exemplars I(alpha) in I. The observable images are the output of medical imaging devices. There are three components that CA examines: (i) constructions of the anatomical submanifolds, (ii) comparison of the anatomical manifolds via estimation of the underlying diffeomorphisms g in G defining the shape or geometry of the anatomical manifolds, and (iii) generation of probability laws of anatomical variation P(.) on the images I for inference and disease testing within anatomical models. This paper reviews recent advances in these three areas applied to shape, growth, and atrophy.

Bilateral hippocampal volume reduction in adults with post-traumatic stress disorder: A meta-analysis of structural MRI studies

Over the last decade a significant number of studies have reported smaller hippocampal volume in individuals with symptoms of post-traumatic stress disorder (PTSD) relative to control groups, and in some cases hemispheric asymmetries in this effect have been noted. However these reported asymmetries have not been in a consistent direction, and other well-controlled studies have failed to observe any hippocampal volume difference. This paper reports a systematic review and meta-analysis of studies in which hippocampal volume was estimated from magnetic resonance images in adult patients with PTSD. After applying a variety of selection criteria intended to minimize potential confounds in pooled effect-size estimates, the meta-analysis included 13 studies of adult patients with PTSD that compared the patients to well-matched control groups, for a total of 215 patients and 325 control subjects. The studies varied with respect to participant age, gender distribution, source of trauma, severity of symptoms, duration of disorder, the nature of the control groups, and the methods employed for volumetric quantification. Despite these differences, pooled effect size calculations across the studies indicated significant volume differences in both hemispheres. On average PTSD patients had a 6.9% smaller left hippocampal volume and a 6.6% smaller right hippocampal volume compared with control subjects. These volume differences were smaller when comparing PTSD patients with control subjects exposed to similar levels of trauma, and larger when comparing PTSD patients to control subjects without significant trauma exposure. Such differences are consistent with the notion that exposure to stressful experiences can lead to hippocampal atrophy, although prospective studies would be necessary to unambiguously establish such a relationship.

Mindboggle: a scatterbrained approach to automate brain labeling

Mindboggle (http://www.binarybottle.com/mindboggle.html) is a fully automated, feature matching approach to label cortical structures and activity anatomically in human brain MRI data. This approach does not assume that the existence of component structures and their relative spatial relationship is preserved from brain to brain, but instead disassembles a labeled atlas and reassembles its pieces to match corresponding pieces in an unlabeled subject brain before labeling. Mindboggle: (1) converts linearly coregistered subject and atlas MRI data into sulcus pieces, (2) matches each atlas piece with a combination of subject pieces by minimizing a cost function, (3) transforms atlas label boundaries to the matching subject pieces, (4) warps atlas labels to their transformed boundaries, and (5) propagates labels to fill remaining gaps in a mask derived from the subject brain. We compared Mindboggle with four registration methods: linear registration, and nonlinear registration using SPM2, AIR, and ANIMAL. Automated labeling by all of the nonlinear methods was found to be at least comparable with linear registration. Mindboggle outperformed every other method, as measured by the agreement between overlapping atlas labels and manually assigned subject labels, with respect to the union or the intersection of voxels. After applying the same procedure that Mindboggle uses to fill a subject's segmented gray matter mask with labels (step 5), the results of the other methods improved. However, after performing a one-way ANOVA (and Tukey's honestly significant difference criterion) in a multiple comparison between the results obtained by the different methods, Mindboggle was still found to be the only nonlinear method whose labeling performance was significantly better than that of linear registration or SPM2. Further advantages to Mindboggle include a high degree of robustness against image artifacts, poor image quality, and incomplete brain data. We tested the latter hypothesis by conducting all of the tests again, this time registering the atlas to an artificially lesioned version of itself, and found that Mindboggle was the only method whose performance did not degrade significantly as the lesion size increased.

Image processing: global and regional changes with age

Our knowledge about the process of aging has continued to evolve as the methods used to study this process become more sophisticated. As more becomes known about the diagnostic criteria for dementia, the population of subjects taking part in aging studies has become more carefully screened minimizing the role of dementia as a confounding variable. Furthermore, advances in imaging techniques now allow us to view the anatomy of the brain in vivo better than any time in the past paving the way for longitudinal studies of the brain. It should not be surprising given the changes seen in studies of aging that our conventional wisdom of the aging process is being called into question.

A hybrid approach to the skull stripping problem in MRI

We present a novel skull-stripping algorithm based on a hybrid approach that combines watershed algorithms and deformable surface models. Our method takes advantage of the robustness of the former as well as the surface information available to the latter. The algorithm first localizes a single white matter voxel in a T1-weighted MRI image, and uses it to create a global minimum in the white matter before applying a watershed algorithm with a preflooding height. The watershed algorithm builds an initial estimate of the brain volume based on the three-dimensional connectivity of the white matter. This first step is robust, and performs well in the presence of intensity nonuniformities and noise, but may erode parts of the cortex that abut bright nonbrain structures such as the eye sockets, or may remove parts of the cerebellum. To correct these inaccuracies, a surface deformation process fits a smooth surface to the masked volume, allowing the incorporation of geometric constraints into the skull-stripping procedure. A statistical atlas, generated from a set of accurately segmented brains, is used to validate and potentially correct the segmentation, and the MRI intensity values are locally re-estimated at the boundary of the brain. Finally, a high-resolution surface deformation is performed that accurately matches the outer boundary of the brain, resulting in a robust and automated procedure. Studies by our group and others outperform other publicly available skull-stripping tools.

Activation of the fusiform gyrus when individuals with autism spectrum disorder view faces

Prior imaging studies have failed to show activation of the fusiform gyrus in response to emotionally neutral faces in individuals with autism spectrum disorder (ASD) [Critchley et al., Brain 124 (2001) 2059; Schultz et al., Arch. Gen. Psychiatry 57 (2000) 331]. However, individuals with ASD do not typically exhibit the striking behavioral deficits that might be expected to result from fusiform gyrus damage, such as those seen in prosopagnosia, and their deficits appear to extend well beyond face identification to include a wide range of impairments in social perceptual processing. In this study, our goal was to further assess the question of whether individuals with ASD have abnormal fusiform gyrus activation to faces. We used high-field (3 T) functional magnetic resonance imaging to study face perception in 11 adult individuals with autism spectrum disorder (ASD) and 10 normal controls. We used face stimuli, object stimuli, and sensory control stimuli (Fourier scrambled versions of the face and object stimuli) containing a fixation point in the center to ensure that participants were looking at and attending to the images as they were presented. We found that individuals with ASD activated the fusiform face area and other brain areas normally involved in face processing when they viewed faces as compared to non-face stimuli. These data indicate that the face-processing deficits encountered in ASD are not due to a simple dysfunction of the fusiform area, but to more complex anomalies in the distributed network of brain areas involved in social perception and cognition.

Object-based morphometry of the cerebral cortex

Most of the approaches dedicated to automatic morphometry rely on a point-by-point strategy based on warping each brain toward a reference coordinate system. In this paper, we describe an alternative object-based strategy dedicated to the cortex. This strategy relies on an artificial neuroanatomist performing automatic recognition of the main cortical sulci and parcellation of the cortical surface into gyral patches. A set of shape descriptors, which can be compared across subjects, is then attached to the sulcus and gyrus related objects segmented by this process. The framework is used to perform a study of 142 brains of the International Consortium for Brain Mapping (ICBM) database. This study reveals some correlates of handedness on the size of the sulci located in motor areas, which was not detected previously using standard voxel based morphometry.