Sally L. Gewalt

Cochlear fluid space dimensions for six species derived from reconstructions of three-dimensional magnetic resonance images

Objectives: To establish the dimensions and volumes of the cochlear fluid spaces.

Magnetic resonance histology for morphologic phenotyping.

Magnetic resonance histology (MRH) images of the whole mouse have been acquired at 100‐micron isotropic resolution at 2.0T with image arrays of 256 × 256 × 1024. Higher resolution (50 × 50 × 50 microns) of limited volumes has been acquired at 7.1T with image arrays of 512 × 512 × 512. Even higher resolution images (20 × 20 × 20 microns) of isolated organs have been acquired at 9.4T. The volume resolution represents an increase of 625,000× over conventional clinical MRI. The technological basis is summarized that will allow basic scientists to begin using MRH as a routine method for morphologcic phenotyping of the mouse. MRH promises four unique attributes over conventional histology: 1) MRH is non‐destructive; 2) MRH exploits the unique contrast mechanisms that have made MRI so successful clinically; 3) MRH is 3‐dimensional; and 4) the data are inherently digital. We demonstrate the utility in morphologic phenotyping a whole C57BL/6J mouse. J. Magn. Reson. Imaging 2002;16:423–429. Published 2002 Wiley‐Liss, Inc.

Nuclear magnetic resonance imaging at microscopic resolution

Abstract Resolution limits in NMR imaging are imposed by bandwidth considerations, available magnetic gradients for spatial encoding, and signal to noise. This work reports modification of a clinical NMR imaging device with picture elements of 500 × 500 × 5000 μm to yield picture elements of 50 × 50 × 1000 μm. Resolution has been increased by using smaller gradient coils permitting gradient fields >0.4 mT/cm. Significant improvements in signal to noise are achieved with smaller rf coils, close attention to choice of bandwidth, and signal averaging. These improvements permit visualization of anatomical structures in the rat brain with an effective diameter of 1 cm with the same definition as is seen in human imaging. The techniques and instrumentation should open a number of basic sciences such as embryology, plant sciences, and teratology to the potentials of NMR imaging.

High-throughput morphologic phenotyping of the mouse brain with magnetic resonance histology

The Mouse Biomedical Informatics Research Network (MBIRN) has been established to integrate imaging studies of the mouse brain ranging from three-dimensional (3D) studies of the whole brain to focused regions at a sub-cellular scale. Magnetic resonance (MR) histology provides the entry point for many morphologic comparisons of the whole brain. We describe a standardized protocol that allows acquisition of 3D MR histology (43-microm resolution) images of the fixed, stained mouse brain with acquisition times <30 min. A higher resolution protocol with isotropic spatial resolution of 21.5 microm can be executed in 2 h. A third acquisition protocol provides an alternative image contrast (at 43-microm isotropic resolution), which is exploited in a statistically driven algorithm that segments 33 of the most critical structures in the brain. The entire process, from specimen perfusion, fixation and staining, image acquisition and reconstruction, post-processing, segmentation, archiving, and analysis, is integrated through a structured workflow. This yields a searchable database for archive and query of the very large (1.2 GB) images acquired with this standardized protocol. These methods have been applied to a collection of both male and female adult murine brains ranging over 4 strains and 6 neurologic knockout models. These collection and acquisition methods are now available to the neuroscience community as a standard web-deliverable service.

Detection and quantification of endolymphatic hydrops in the guinea pig cochlea by magnetic resonance microscopy

Three-dimensional magnetic resonance microscopy (MRM) was used to study normal and hydropic cochleae of the guinea pig. With this technique consecutive serial slices representing the entire volume of isolated, fixed cochleae were obtained. The voxels (volume elements) making up the contiguous slices were isotropic (25 microns 3) and in each slice the boundaries of scala media, including the position of Reissners membrane, were clearly delineated. Three-dimensional reconstructions of the endolymphatic and perilymphatic scale were generated. Custom software was developed to quantify cross-sectional area (CSA) of all scalae. In the normal cochlea all 3 scalae, including scala media, showed a gradual decrease in CSA from base to apex. Marked differences existed between our findings and previously reported cochlear dimensions, especially for the perilymphatic scalae in the basal turn. In hydropic cochleae the scala media was enlarged to a varying extent in different turns and marked changes in the degree of distension of Reissners membrane occurred along the cochlea. MRM and subsequent computer analysis of the isotropic data provide excellent methods for imaging and quantifying the fluid spaces of normal and hydropic cochleae.

Quantitative anatomy of the round window and cochlear aqueduct in guinea pigs

In order to analyze the entry of solutes through the round window membrane, a quantitative description of round window anatomy in relationship to scala tympani is required. High-resolution magnetic resonance microscopy was used to visualize the fluid spaces and tissues of the inner ear in three dimensions in isolated, fixed specimens from guinea pigs. Each specimen was represented as consecutive serial slices, with a voxel size of approximately 25 microm(3). The round window membrane, and its relationship to the terminal portion of scala tympani in the basal turn, was quantified in six specimens. In each image slice, the round window membrane and scala tympani were identified and segmented. The total surface area of the round window membrane averaged 1.18 mm(2) (S.D. 0.08, n=6). The length and variation of cross-sectional area as a function of distance for the cochlear aqueduct was determined in five specimens. The cochlear aqueduct was shown to enter scala tympani at the medial limit of the round window membrane, which corresponded to a distance of approximately 1 mm from the end of the scala when measured along its mid-point. These data are of value in simulating drug and other solute movements in the cochlear fluids and have been incorporated into a public-domain simulation program available at http://oto.wustl.edu/cochlea/.

Imaging the cochlea by magnetic resonance microscopy

The isolated, fixed cochlea of the mustached bat was studied with three dimensional magnetic resonance (MR) microscopy. The cochlea of this animal is about 4 mm in diameter and its entire volume was imaged. With the field of view and matrix size used, the volume elements (voxels) making up the volume data set were isotropic 25 x 25 x 25 micron cubes. Three dimensional (3D) MR microscopy based on isotropic voxels has many advantages over commonly used light microscopy: 1) it is non destructive; 2) it is much less time consuming; 3) no dehydration is required and shrinkage is minimized; 4) the data set can be used to create sections in any desired plane; 5) the proper alignment of sections is inherent in the 3D acquisition so that no reference points are required; 6) the entire data set can be viewed from any point of view in a volume rendered image; 7) the data is digital and features can be enhanced by computer image processing; and 8) the isotropic dimensions of the voxels make the data well-suited for structural reconstructions and measurements. Good images of the osseous spiral lamina, spiral ligament, scala tympani, scala vestibuli, and nerve bundles were obtained. The vestibular (Reissners) membrane was easily identified in the mustached bat and it appears to bulge into the scala vestibuli. The visibility of this structure suggests that MR microscopy would be well-suited for studies of endolymphatic hydrops.

Quantitative mouse brain phenotyping based on single and multispectral MR protocols.

Sophisticated image analysis methods have been developed for the human brain, but such tools still need to be adapted and optimized for quantitative small animal imaging. We propose a framework for quantitative anatomical phenotyping in mouse models of neurological and psychiatric conditions. The framework encompasses an atlas space, image acquisition protocols, and software tools to register images into this space. We show that a suite of segmentation tools (Avants, Epstein et al., 2008) designed for human neuroimaging can be incorporated into a pipeline for segmenting mouse brain images acquired with multispectral magnetic resonance imaging (MR) protocols. We present a flexible approach for segmenting such hyperimages, optimizing registration, and identifying optimal combinations of image channels for particular structures. Brain imaging with T1, T2* and T2 contrasts yielded accuracy in the range of 83% for hippocampus and caudate putamen (Hc and CPu), but only 54% in white matter tracts, and 44% for the ventricles. The addition of diffusion tensor parameter images improved accuracy for large gray matter structures (by >5%), white matter (10%), and ventricles (15%). The use of Markov random field segmentation further improved overall accuracy in the C57BL/6 strain by 6%; so Dice coefficients for Hc and CPu reached 93%, for white matter 79%, for ventricles 68%, and for substantia nigra 80%. We demonstrate the segmentation pipeline for the widely used C57BL/6 strain, and two test strains (BXD29, APP/TTA). This approach appears promising for characterizing temporal changes in mouse models of human neurological and psychiatric conditions, and may provide anatomical constraints for other preclinical imaging, e.g. fMRI and molecular imaging. This is the first demonstration that multiple MR imaging modalities combined with multivariate segmentation methods lead to significant improvements in anatomical segmentation in the mouse brain.

Quantitative anatomy of the guinea pig endolymphatic sac

The endolymphatic sac is believed to represent one of the primary loci for endolymph volume regulation in the inner ear. Quantitative analysis of physiologic measurements from the endolymphatic sac requires knowledge of the anatomy of the structure, specifically the luminal volume and the variation of cross-sectional area with distance along the sac. Recently techniques have become available to make these measurements. In the present study, fixed, isolated specimens of the guinea pig endolymphatic sac were imaged by high-resolution magnetic resonance microscopy (MRM) or by histological serial sections. Structures were reconstructed and quantified using image analysis software. In specimens imaged by MRM the endolymphatic sac volume, including tissue and lumen, was 359 nl for the intraosseous region and 106 nl for the extraosseous region, totaling 465 nl for the entire structure. The luminal volumes were 131 nl for the intraosseous region and 13 nl for the extraosseous region, totaling 144 nl. In histological specimens the volume, including tissue and lumen, was 414 nl for the intraosseous region and 121 nl for the extraosseous region, totaling 535 nl for the entire structure. The luminal volumes were 152 nl for the intraosseous region and 26 nl for the extraosseous region, totaling 179 nl. Differences in volume estimates obtained by the two methods were not statistically significant and variation was dominated by inter-specimen variation. Pooling the data, the total volume of the endolymphatic sac in the guinea pig including tissue and lumen was 506 nl (S.D. 100, n=17) and the volume of the lumen was 169 nl (S.D. 48, n=14).

An engineering approach to image-based phenotyping

The explosive growth of rodent models for basic biomedical research has created a compelling need for new methods of phenotyping. Modern imaging technologies offer a number of solutions. But, simply scaling clinical systems to use on small animal models is inappropriate. We present here a method to develop imaging tools for both structural and functional phenotyping using an interdisciplinary approach. Two examples are presented: (1) in vivo microscopy of the rat lung using hyperpolarized /sup 3/He, and (2) morphologic phenotyping of the perfusion fixed mouse.