Duke Center for In-Vivo Microscopy (CIVM) High Resolution MRI Images
All datasets for each study are available for review and download from CIVMSpace, the Duke Center for In-Vivo Microscopy's Web portal.
System Requirements:
CIVMSpace is designed to work on the Microsoft Windows and Mac OS X platforms.
Supported browsers on Windows are Internet Explorer versions 6.0 and above, and Mozilla Firefox versions 1.0 and above.
On the Mac, Safari versions 1.3 and above and Mozilla Firefox versions 1.0 and above are supported.
VoxStation requires a working Java installation. It has been tested with Java 1.4, 1.5 and 1.6 on Windows, and Java 1.4 and 1.5 on Mac OS X.
Common Specimen
A collection of 8 brains imaged with two protocols: a T1 and a T2 weighted protocol with 43x43x90 microns resolution.The project efforts are focused on the development and application of correlated imaging approaches (confocal and electron microscopy, microscopic MRI) to Parkinson's Disease (PD) - applied first to recently generated transgenic animal models of PD (alpha-synuclein). These MRI scans will allow the examination and comparison volumes of brain structures quantitatively, as well as qualitative changes in tissue content, in both the a-SYN transgenic and age-matched control animals.
DAT-KO
A collection of Male DAT-KO (n = 4) and WT mice (n = 4) between 4 and 8 months of age were used in this study. Images were acquired with 3 dimensional rf refocused spin warp encoding on 3 D (256 × 256 × 512) image arrays covering an 11 × 11 × 22 mm field of view yielding isotropic voxels of 43 × 43 × 43 μm (8 × 10−5 mm3) T1 weighted images were acquired with TR = 100 ms, TE = 5 ms, NEX = 4. Perfusion with contrast agent reduces the mean T1 to <200 ms allowing acquisition of a T2 weighted image with a much shorter TR (200 ms) and TE (15 ms) than might be used with unstained (formalin fixed) tissues. NEX was reduced to 2.-
Neuroimage. 2005 May 15;26(1):83-90.
-
Magnetic resonance imaging at microscopic resolution reveals subtle morphological changes in a mouse model of dopaminergic hyperfunction.
Department of Cell Biology, Center for Models of Human Disease, Duke University Medical Center, Durham, NC 27710, USA.
Structural abnormalities of the basal ganglia have been documented in several neuropsychiatric conditions associated with dysregulation of the dopamine system. However, the histological nature underlying these changes is largely unknown. Using magnetic resonance imaging at microscopic resolution (MRI, 9.4 T with 43 microm isotropic spatial resolution) and stereological techniques, we have investigated the effect of increased dopamine neurotransmission on brain morphology in mice with elevated extracellular dopamine, the dopamine transporter knockout (DAT-KO) mice. We first demonstrate the usefulness of MRI at microscopic resolution for the accurate identification and measurement of volumes of specific subregions, accounting for less than 0.03% (0.16 mm(3)) of the volume of a mouse brain. Furthermore, the MRI analysis reveals a significantly lower volume (-9%) of the anterior striatum of DAT-KO mice, while the volume of other dopamine-related structures such as the posterior striatum and the substantia nigra pars reticulata is unchanged in comparison to wild type littermates. Stereological analysis performed in the same brains reveals that one important structural factor accounting for this selective change in volume is a reduction of 18% in the absolute number of neuronal cell bodies. The feasibility of assessing accurately small morphological alterations in mouse models, where the molecular and histological pathologies can be easily compared in a controlled manner, provides a paradigm to examine the relevance of selective brain volumetric changes associated with a number of neuropathological conditions.
Unstained mouse brains
A collection of 6 brains imaged with a multispectral protocol (5 constrast). The three-dimensional (3D) MR datasets acquired at 90-microm isotropic resolution. 21 labeles structures and the whole brain were identified in those brains.
Reference:
-
Neuroimage. 2005 Aug 15;27(2):425-35.
-
-
Automated segmentation of neuroanatomical structures in multispectral MR microscopy of the mouse brain.
Center for In Vivo Microscopy, Box 3302, Duke University Medical Center, Durham, NC 27710, USA. anjum.ali@duke.edu
We present the automated segmentation of magnetic resonance microscopy (MRM) images of the C57BL/6J mouse brain into 21 neuroanatomical structures, including the ventricular system, corpus callosum, hippocampus, caudate putamen, inferior colliculus, internal capsule, globus pallidus, and substantia nigra. The segmentation algorithm operates on multispectral, three-dimensional (3D) MR data acquired at 90-microm isotropic resolution. Probabilistic information used in the segmentation is extracted from training datasets of T2-weighted, proton density-weighted, and diffusion-weighted acquisitions. Spatial information is employed in the form of prior probabilities of occurrence of a structure at a location (location priors) and the pairwise probabilities between structures (contextual priors). Validation using standard morphometry indices shows good consistency between automatically segmented and manually traced data. Results achieved in the mouse brain are comparable with those achieved in human brain studies using similar techniques. The segmentation algorithm shows excellent potential for routine morphological phenotyping of mouse models.
Stained Mouse Brain
- A collection of 6 stained brains imaged witin the skull using two protocols: a T1 weighted protocol and a T2 image (MEFIC enhanced 3D CPMG). Automated segmentations of the brains in 33 structures are provided, as well as as an atlas brain.
Badea A, Ali-Sharief AA, Johnson GA. Morphometric analysis of the C57BL/6J mouse brain. Neuroimage. 2007 Sep 1;37(3):683-93. Epub 2007 Jun 7.
Johnson GA, Ali-Sharief A, Badea A, Brandenburg J, Cofer G, Fubara B, Gewalt S, Hedlund LW, Upchurch L. High-throughput morphologic phenotyping of the mouse brain with magnetic resonance histology.
Neuroimage. 2007 Aug 1;37(1):82-9. Epub 2007 May 18.
Reeler mouse model
- A collection of 12 high resolution three-dimensional (3D) MR data acquired at 21.5-micron isotropic resolution.
-
Neuroimage. 2007 Feb 15;34(4):1363-74. Epub 2006 Dec 20.
-
Neuroanatomical phenotypes in the reeler mouse.
Center for In Vivo Microscopy, Box 3302, Duke University Medical Center, Durham, NC 27710, USA.
The reeler mouse (Reln) has been proposed as a neurodevelopmental model for certain neurological and psychiatric conditions and has been studied by qualitative histochemistry and electron microscopy. Using magnetic resonance microscopy (MRM), we have quantitated for the first time the neuromorphology of Reln mice at a resolution of 21.5 microm. The neuroanatomical phenotypes of heterozygous and homozygous mutant Reln mice were compared to those of wild type (WT) littermates using morphometry and texture analysis. The cortical, hippocampal, and cerebellar phenotypes of the heterozygous and homozygous mutant Reln mice were confirmed, and new features were revealed. The Reln(rl/rl) mice possessed a smaller brain, and both Reln(rl/+) and Reln(rl/rl) mice had increased ventricles compared to WT controls. Shape differences were found between WT and Reln(rl/rl) brains, specifically in cerebellum, olfactory bulbs, dorsomedial frontal and parietal cortex, certain regions of temporal and occipital lobes, as well as in the lateral ventricles and ventral hippocampus. These findings suggest that certain brain regions may be more severely impacted by the Reln mutation than others. Gadolinium-based active staining demonstrated that layers of the hippocampus were disorganized in Reln(rl/rl) mice and differences in thickness of these layers were identified between WT and Reln(rl/rl) mice. The intensity distributions characteristic to the dorsal, middle, and ventral hippocampus were altered in the Reln(rl/rl), especially in the ventral hippocampus. These differences were quantified using skewness and modeling the intensity distributions with a Gaussian mixture. Our results suggest that structural features of Reln(rl/rl) brain most closely phenocopy those of patients with Norman-Roberts lissencephaly.
Genetic reference population
A collection of 24 brains : C57BL/6, DBA2 and 10 BXD strains selected for large hippocampal volume variations. The images are high resolution three-dimensional (3D) MR data acquired at 21.5-micron isotropic resolution (T1) and matched 43-micron resolution MEFIC processed scans (T2). The results of automated segmentation into 33 structures are also provided.
The panel of isogenic BXD strains of mice provides a superb resource to study the genetic basis of differences in brain structure and function. In previous work several research groups have discovered significant heritable differences in the size of the hippocampus, striatum, cerebellum, thalamus, olfactory bulb, neocortex, and amygdala. In many cases this variation has been linked to gene loci. However, it has not yet been practical to systematically quantify genetic covariance across multiple brain regions using a single "coherent" set of animals. In this study we addressed this problem by exploiting high-resolution MR microscopy and automated segmentation. We segmented 33 brain regions in a subset of BXD strains with maximal differences in hippocampal weight. We describe between strain differences in the volumes of these 33 brain structures. These differences in volume range for example from 20.42 mm3 to 30.42 mm3, with a coefficient of variation of 12.99 % for hippocampus, from 14.51 mm3 to 24.66 mm3, with a coefficient of variation of 12.58% for striatum, and from 43.95 mm3 to 62.69 mm3, with a coefficient of variation of 8.68% for cerebellum. Data on the volume variability across these BXD strains are accessible online at www.genenetwork.org.
alpha-synuclein Parkinson disease model
A collection of 10 high resolution three-dimensional (3D) MR data acquired at 21.5-micron isotropic resolution (T1) and 10 matched 43-micron resolution MEFIC processed scans (T2).
Correlated Imaging Approaches and Multi-scale Databases for Research in Parkinson's Disease. The goal of this project is to examine the brains of a transgenic mouse model of Parkinsonism using rodent MRI imaging. Specimens will include age matched non-transgenic controls. All efforts will be made to have a sufficient number of animals per group (with equal numbers of male and female animals). This study is designed to clarify the results of an earlier pilot study (UCSD-CIVM) in which a low number of subjects precluded our ability to ascertain the presence of gross volumetric and regional signal intensity differences in alpha-synuclein transgenic animals in comparison with non-transgenic control animals. The changes in structural volumes should coincide with areas known to be predisposed to either neuronal cell death or reactive gliosis. Regional changes in signal intensity may be associated with protein aggregations, neuronal cell death, or reactive gliosis. The MRI scans can then be reconciled with large scale images of protein distributions in similar animals. We hypothesize that there are such differences in the transgenic animals and that this new MRI study (with higher group N's and higher scanning resolutions) will provide us with the necessary information to determine which scenario is most likely.