Francois Budin

DTIPrep: quality control of diffusion-weighted images

In the last decade, diffusion MRI (dMRI) studies of the human and animal brain have been used to investigate a multitude of pathologies and drug-related effects in neuroscience research. Study after study identifies white matter (WM) degeneration as a crucial biomarker for all these diseases. The tool of choice for studying WM is dMRI. However, dMRI has inherently low signal-to-noise ratio and its acquisition requires a relatively long scan time; in fact, the high loads required occasionally stress scanner hardware past the point of physical failure. As a result, many types of artifacts implicate the quality of diffusion imagery. Using these complex scans containing artifacts without quality control (QC) can result in considerable error and bias in the subsequent analysis, negatively affecting the results of research studies using them. However, dMRI QC remains an under-recognized issue in the dMRI community as there are no user-friendly tools commonly available to comprehensively address the issue of dMRI QC. As a result, current dMRI studies often perform a poor job at dMRI QC. Thorough QC of dMRI will reduce measurement noise and improve reproducibility, and sensitivity in neuroimaging studies; this will allow researchers to more fully exploit the power of the dMRI technique and will ultimately advance neuroscience. Therefore, in this manuscript, we present our open-source software, DTIPrep, as a unified, user friendly platform for thorough QC of dMRI data. These include artifacts caused by eddy-currents, head motion, bed vibration and pulsation, venetian blind artifacts, as well as slice-wise and gradient-wise intensity inconsistencies. This paper summarizes a basic set of features of DTIPrep described earlier and focuses on newly added capabilities related to directional artifacts and bias analysis.

Ethanol-induced face-brain dysmorphology patterns are correlative and exposure-stage dependent.

Prenatal ethanol exposure is the leading preventable cause of congenital mental disability. Whereas a diagnosis of fetal alcohol syndrome (FAS) requires identification of a specific pattern of craniofacial dysmorphology, most individuals with behavioral and neurological sequelae of heavy prenatal ethanol exposure do not exhibit these defining facial characteristics. Here, a novel integration of MRI and dense surface modeling-based shape analysis was applied to characterize concurrent face-brain phenotypes in C57Bl/6J fetuses exposed to ethanol on gestational day (GD)7 or GD8.5. The facial phenotype resulting from ethanol exposure depended upon stage of insult and was predictive of unique patterns of corresponding brain abnormalities. Ethanol exposure on GD7 produced a constellation of dysmorphic facial features characteristic of human FAS, including severe midfacial hypoplasia, shortening of the palpebral fissures, an elongated upper lip, and deficient philtrum. In contrast, ethanol exposure on GD8.5 caused mild midfacial hypoplasia and palpebral fissure shortening, a shortened upper lip, and a preserved philtrum. These distinct, stage-specific facial phenotypes were associated with unique volumetric and shape abnormalities of the septal region, pituitary, and olfactory bulbs. By demonstrating that early prenatal ethanol exposure can cause more than one temporally-specific pattern of defects, these findings illustrate the need for an expansion of current diagnostic criteria to better capture the full range of facial and brain dysmorphology in fetal alcohol spectrum disorders.

UNC-Utah NA-MIC framework for DTI fiber tract analysis

Diffusion tensor imaging has become an important modality in the field of neuroimaging to capture changes in micro-organization and to assess white matter integrity or development. While there exists a number of tractography toolsets, these usually lack tools for preprocessing or to analyze diffusion properties along the fiber tracts. Currently, the field is in critical need of a coherent end-to-end toolset for performing an along-fiber tract analysis, accessible to non-technical neuroimaging researchers. The UNC-Utah NA-MIC DTI framework represents a coherent, open source, end-to-end toolset for atlas fiber tract based DTI analysis encompassing DICOM data conversion, quality control, atlas building, fiber tractography, fiber parameterization, and statistical analysis of diffusion properties. Most steps utilize graphical user interfaces (GUI) to simplify interaction and provide an extensive DTI analysis framework for non-technical researchers/investigators. We illustrate the use of our framework on a small sample, cross sectional neuroimaging study of eight healthy 1-year-old children from the Infant Brain Imaging Study (IBIS) Network. In this limited test study, we illustrate the power of our method by quantifying the diffusion properties at 1 year of age on the genu and splenium fiber tracts.

Peri-adolescent ethanol vapor exposure produces reductions in hippocampal volume that are correlated with deficits in prepulse inhibition of the startle.

BACKGROUND Epidemiological studies suggest that excessive alcohol consumption is prevalent among adolescents and may have lasting neurobehavioral consequences. The use of animal models allows for the separation of the effects of adolescent ethanol (EtOH) exposure from genetic background and other environmental insults. In this study, the effects of moderate EtOH vapor exposure, during adolescence, on structural diffusion tensor imaging (DTI) and behavioral measures were evaluated in adulthood. METHODS A total of 53 Wistar rats were received at postnatal day (PD) 21 and were randomly assigned to EtOH vapor (14 hours on/10 hours off/day) or air exposure for 35 days from PD 23 to 58 (average blood ethanol concentration: 169 mg%). Animals were received in 2 groups that were subsequently sacrificed at 2 time points following withdrawal from EtOH vapor: (i) at 72 days of age, 2 weeks following withdrawal or (ii) at day 128, 10 weeks following withdrawal. In the second group, behavior in the light/dark box and prepulse inhibition (PPI) of the startle was also evaluated. Fifteen animals in each group were scanned, postmortem, for structural DTI. RESULTS There were no significant differences in body weight between EtOH and control animals. Volumetric data demonstrated that total brain, hippocampal, corpus callosum but not ventricular volume were significantly larger in the 128-day-sacrificed animals as compared to the 72 day animals. The hippocampus was smaller and the ventricles larger at 128 days as compared to 72 days, in the EtOH-exposed animals, leading to a significant group × time effect. EtOH-exposed animals sacrificed at 128 days also had diminished PPI, and more rears in the light box were significantly correlated with hippocampal size. CONCLUSIONS These studies demonstrate that DTI volumetric measures of hippocampus are significantly impacted by age and peri-adolescent EtOH exposure and withdrawal in Wistar rats.

Magnetic resonance microscopy-based analyses of the neuroanatomical effects of gestational day 9 ethanol exposure in mice

Animal model-based studies have shown that ethanol exposure during early gestation induces developmental stage-specific abnormalities of the face and brain. The exposure time-dependent variability in ethanols teratogenic outcomes is expected to contribute significantly to the wide spectrum of effects observed in humans with fetal alcohol spectrum disorder (FASD). The work presented here employs a mouse FASD model and magnetic resonance microscopy (MRM; high resolution magnetic resonance imaging) in studies designed to further our understanding of the developmental stage-specific defects of the brain that are induced by ethanol. At neurulation stages, i.e. at the beginning of gestational day (GD) 9 and again 4 hours later, time-mated C57Bl/6J dams were intraperitoneally administered 2.9 g/kg ethanol or vehicle. Ethanol-exposed fetuses were collected on GD 17, processed for MRM analysis, and results compared to comparably staged controls. Linear and volume measurements as well as shape changes for numerous individual brain regions were determined. GD 9 ethanol exposure resulted in significantly increased septal region width, reduction of cerebellar volume, and enlargement of all of the ventricles. Additionally, the results of shape analyses showed that many areas of the ethanol-exposed brains including the cerebral cortex, hippocampus and right striatum were significantly misshapen. These data demonstrate that ethanol can induce dysmorphology that may not be obvious based on volumetric analyses alone, highlight the asymmetric aspects of ethanol-induced defects, and add to our understanding of ethanols developmental stage-dependent neuroteratogenesis.

Automatic skull-stripping of rat MRI/DTI scans and atlas building

3D Magnetic Resonance (MR) and Diffusion Tensor Imaging (DTI) have become important noninvasive tools for the study of animal models of brain development and neuropathologies. Fully automated analysis methods adapted to rodent scale for these images will allow highthroughput studies. A fundamental first step for most quantitative analysis algorithms is skullstripping, which refers to the segmentation of the image into two tissue categories, brain and non-brain. In this manuscript, we present a fully automatic skull-stripping algorithm in an atlasbased manner. We also demonstrate how to either modify an external atlas or to build an atlas from the population itself to present a self-contained approach. We applied our method to three datasets of rat brain scans, at different ages (PND5, PND14 and adult), different study groups (control, ethanol exposed, intrauterine cocaine exposed), as well as different image acquisition parameters. We validated our method by comparing the automated skull-strip results to manual delineations performed by our expert, which showed a discrepancy of less than a single voxel on average. We thus demonstrate that our algorithm can robustly and accurately perform the skull-stripping within one voxel of the manual delineation, and in a fraction of the time it takes a human expert.

Fully automated rodent brain MR image processing pipeline on a Midas server: from acquired images to region-based statistics

Magnetic resonance imaging (MRI) of rodent brains enables study of the development and the integrity of the brain under certain conditions (alcohol, drugs etc.). However, these images are difficult to analyze for biomedical researchers with limited image processing experience. In this paper we present an image processing pipeline running on a Midas server, a web-based data storage system. It is composed of the following steps: rigid registration, skull-stripping, average computation, average parcellation, parcellation propagation to individual subjects, and computation of region-based statistics on each image. The pipeline is easy to configure and requires very little image processing knowledge. We present results obtained by processing a data set using this pipeline and demonstrate how this pipeline can be used to find differences between populations.

Comparison of Magnetic Resonance Imaging in Live vs. Post Mortem Rat Brains

Magnetic Resonance Imaging (MRI) is an increasingly popular technique for examining neurobiology in rodents because it is both noninvasive and nondestructive. MRI scans can be acquired from either live or post mortem specimens. In vivo scans have a key advantage in that subjects can be scanned at multiple time-points in longitudinal studies. However, repeated exposure to anesthesia and stress may confound studies. In contrast, post mortem scans offer improved image quality and increased signal-to-noise ratio (SNR) due to several key advantages: First, the images are not disrupted by motion and pulsation artifacts. Second, they allow the brain tissue to be perfused with contrast agents, enhancing tissue contrast. Third, they allow longer image acquisition times, yielding higher resolution and/or improved SNR. Fourth, they allow assessment of groups of animals at the same age without scheduling complications. Despite these advantages, researchers are often skeptical of post mortem MRI scans because of uncertainty about whether the fixation process alters the MRI measurements. To address these concerns, we present a thorough comparative study of in vivo and post mortem MRI scans in healthy male Wistar rats at three age points throughout adolescence (postnatal days 28 through 80). For each subject, an in vivo scan was acquired, followed by perfusion and two post mortem scans at two different MRI facilities. The goal was to assess robustness of measurements, to detect any changes in volumetric measurements after fixation, and to investigate any differential bias that may exist between image acquisition techniques. We present this volumetric analysis for comparison of 22 anatomical structures between in vivo and post mortem scans. No significant changes in volumetric measurements were detected; however, as hypothesized, the image quality is dramatically improved in post mortem scans. These findings illustrate the validity and utility of using post mortem scans in volumetric neurobiological studies.

Characterization of subtle brain abnormalities in a mouse model of Hedgehog pathway antagonist-induced cleft lip and palate.

Subtle behavioral and cognitive deficits have been documented in patient cohorts with orofacial clefts (OFCs). Recent neuroimaging studies argue that these traits are associated with structural brain abnormalities but have been limited to adolescent and adult populations where brain plasticity during infancy and childhood may be a confounding factor. Here, we employed high resolution magnetic resonance microscopy to examine primary brain morphology in a mouse model of OFCs. Transient in utero exposure to the Hedgehog (Hh) signaling pathway antagonist cyclopamine resulted in a spectrum of facial dysmorphology, including unilateral and bilateral cleft lip and palate, cleft of the secondary palate only, and a non-cleft phenotype marked by midfacial hypoplasia. Relative to controls, cyclopamine-exposed fetuses exhibited volumetric differences in several brain regions, including hypoplasia of the pituitary gland and olfactory bulbs, hyperplasia of the forebrain septal region, and expansion of the third ventricle. However, in affected fetuses the corpus callosum was intact and normal division of the forebrain was observed. This argues that temporally-specific Hh signaling perturbation can result in typical appearing OFCs in the absence of holoprosencephaly—a condition classically associated with Hh pathway inhibition and frequently co-occurring with OFCs. Supporting the premise that some forms of OFCs co-occur with subtle brain malformations, these results provide a possible ontological basis for traits identified in clinical populations. They also argue in favor of future investigations into genetic and/or environmental modulation of the Hh pathway in the etiopathogenesis of orofacial clefting.

Automatic cortical thickness analysis on rodent brain

Localized difference in the cortex is one of the most useful morphometric traits in human and animal brain studies. There are many tools and methods already developed to automatically measure and analyze cortical thickness for the human brain. However, these tools cannot be directly applied to rodent brains due to the different scales; even adult rodent brains are 50 to 100 times smaller than humans. This paper describes an algorithm for automatically measuring the cortical thickness of mouse and rat brains. The algorithm consists of three steps: segmentation, thickness measurement, and statistical analysis among experimental groups. The segmentation step provides the neocortex separation from other brain structures and thus is a preprocessing step for the thickness measurement. In the thickness measurement step, the thickness is computed by solving a Laplacian PDE and a transport equation. The Laplacian PDE first creates streamlines as an analogy of cortical columns; the transport equation computes the length of the streamlines. The result is stored as a thickness map over the neocortex surface. For the statistical analysis, it is important to sample thickness at corresponding points. This is achieved by the particle correspondence algorithm which minimizes entropy between dynamically moving sample points called particles. Since the computational cost of the correspondence algorithm may limit the number of corresponding points, we use thin-plate spline based interpolation to increase the number of corresponding sample points. As a driving application, we measured the thickness difference to assess the effects of adolescent intermittent ethanol exposure that persist into adulthood and performed t-test between the control and exposed rat groups. We found significantly differing regions in both hemispheres.