Computer-assisted alignment of standard serial sections without use of artificial landmarks. A practical approach to the utilization of incomplete information in 3-D reconstruction of the hippocampal region

Multi-Scale Investigation of Human Renal Tissue in Three Dimensions

Histopathology as a diagnostic mainstay for tissue evaluation is strictly a 2D technology. Combining and supplementing this technology with 3D imaging has been proposed as one future avenue towards refining comprehensive tissue analysis. To this end, we have developed a laboratory-based X-ray method allowing for the investigation of tissue samples in three dimensions with isotropic volume information. To assess the potential of our method for micro-morphology evaluation, we selected several kidney regions from three patients with cystic kidney disease, obstructive nephropathy and diabetic glomerulopathy. Tissue specimens were processed using our in-house-developed X-ray eosin stain and investigated with a commercial microCT and our in-house-built NanoCT. The microCT system provided overview scans with voxel sizes of [Formula: see text] and the NanoCT was employed for higher resolutions including voxel sizes from [Formula: see text] to 210 nm. We present a methodology allowing for a precise micro-morphologic investigation in three dimensions which is compatible with conventional histology. Advantages of our methodology are its versatility with respect to multi-scale investigations, being laboratory-based, allowing for non-destructive imaging and providing isotropic volume information. We believe, that after future developmental work this method might contribute to advanced multi-modal tissue diagnostics.

3D-Non-destructive Imaging through Heavy-Metal Eosin Salt Contrast Agents

Conventional histology is a destructive technique based on the evaluation of 2D slices of a 3D biopsy. By using 3D X‐ray histology these obstacles can be overcome, but their application is still restricted due to the inherently low attenuation properties of soft tissue. In order to solve this problem, the tissue can be stained before X‐ray computed tomography imaging (CT) to enhance the soft tissue X‐ray contrast. Evaluation of brominated fluorescein salts revealed a mutual influence of the number of bromine atoms and the cations applied on the achieved contrast enhancement. The dibromo fluorescein barium salt turned out to be the ideal X‐ray contrast agent, allowing for 3D imaging and subsequent complementing counterstaining applying standard histological techniques.

Revealing the Microscopic Structure of Human Renal Cell Carcinoma in Three Dimensions

For fully characterizing renal cell carcinoma (RCC), information about the 3D tissue microstructure is essential. Histopathology, which represents the current diagnostic gold standard, is destructive and only provides 2D information. 3D X-ray histology endeavors to overcome these limitations by generating 3D data. In a laboratory environment, most techniques struggle with limited resolution and the weak X-ray attenuation contrast of soft tissue. We recently developed a laboratory-based method combining nanoscopic X-ray CT with a cytoplasm-specific X-ray stain. Here, we present the application of this method to human RCC biopsies. The NanoCT slices enable pathological characterization of crucial structures by reproducing tissue morphology with a similar detail level as corresponding histological light microscopy images. Beyond that, our data offer deeper insights into the 3D configuration of the tumor. By demonstrating the compatibility of the X-ray stain with standard pathological stains, we highlight the feasibility of integrating staining based NanoCT into the pathological routine.

High-resolution episcopic microscopy enables three-dimensional visualization of plant morphology and development

The study of plant anatomy, which can be traced back to the seventeenth century, advanced hand in hand with light microscopy technology and relies on traditional histologic techniques, which are based on serial two-dimensional (2D) sections. However, these valuable techniques lack spatial arrangement of the tissue and hence provide only partial information. A new technique of whole-mount three-dimensional (3D) imaging termed high-resolution episcopic microscopy (HREM) can overcome this obstacle and generate a 3D model of the specimen at a near-histological resolution. Here, we describe the application of HREM technique in plants by analyzing two plant developmental processes in woody plants: oil secretory cavity development in citrus fruit and adventitious root formation in persimmon rootstock cuttings. HREM 3D models of citrus fruit peel showed that oil cavities were initiated schizogenously during the early stages of fruitlet development. Citrus secretory cavity formation, shape, volume, and distribution were analyzed, and new insights are presented. HREM 3D model comparison of persimmon rootstock clones, which differ in their rooting ability, revealed that difficult-to-root clones failed to develop adventitious roots due to their inability to initiate root primordia.

Rhesus Macaque Brain Atlas Regions Aligned to an MRI Template

To aid in the analysis of rhesus macaque brain images, we aligned digitized anatomical regions from the widely used atlas of Paxinos et al. to a published magnetic resonance imaging (MRI) template based on a large number of subjects. Digitally labelled atlas images were aligned to the template in 2D and then in 3D. The resulting grey matter regions appear qualitatively to be well registered to the template. To quantitatively validate the procedure, MR brain images of 20 rhesus macaques were aligned to the template along with regions drawn by hand in striatal and cortical areas in each subject's MRI. There was good geometric overlap between the hand drawn regions and the template regions. Positron emission tomography (PET) images of the same subjects showing uptake of a dopamine D₂ receptor ligand were aligned to the template space, and good agreement was found between tracer binding measures calculated using the hand drawn and template regions. In conclusion, an anatomically defined set of rhesus macaque brain regions has been aligned to an MRI template and has been validated for analysis of PET imaging in a subset of striatal and cortical areas. The entire set of over 200 regions is publicly available at https://www.nitrc.org/ . Graphical Abstract ᅟ.

Morphologic and morphometric study on microvasculature of developing mouse kidneys

A proper morphogenesis of the renal microvasculature is crucial not only for fulfilling the renal function but also to slow down the progression of chronic kidney disease in adulthood. However, the current description of the developing microvasculature is incomplete. The present study investigated the morphogenesis and volume densities of the renal microvasculature using computer-assisted tubular tracing, immunohistochemistry for CD34, and unbiased stereology. The earliest glomerular capillaries were observed at the lower cleft of the S-shaped nephrons, as simple loops connecting the afferent and efferent arterioles. In parallel with this, the peritubular capillaries were established. Noticeably, from early nephrogenesis on, the efferent arterioles of the early-formed glomeruli ran in close proximity to their own thick ascending limbs. In addition, the ascending vasa recta arising from the arcuate or interlobular veins also ran in close proximity to the thick descending limb. Thus, the tubules and vessels formed the typical countercurrent relation in the medulla. No loop bends were observed between descending and ascending vasa recta. The volume density of the cortical and medullary peritubular capillary increased 3.3- and 2.6-fold, respectively, from 2.34 (0.13) and 7.03 (0.09)% [means (SD)] at embryonic day 14.5 (E14.5) to 7.71 (0.44) and 18.27 (1.17)% at postnatal day 40 (P40). In contrast, the volume density of glomeruli changed only slightly during kidney development, from 4.61 (0.47)% at E14.5 to 6.07 (0.2)% at P7 to 4.19 (0.47)% at P40. These results reflect that the growth and formation of the renal microvasculature closely correspond to functional development of the tubules.

Patch-Based Nonlinear Image Registration for Gigapixel Whole Slide Images

OBJECTIVE

Image registration of whole slide histology images allows the fusion of fine-grained information-like different immunohistochemical stains-from neighboring tissue slides. Traditionally, pathologists fuse this information by looking subsequently at one slide at a time. If the slides are digitized and accurately aligned at cell level, automatic analysis can be used to ease the pathologist's work. However, the size of those images exceeds the memory capacity of regular computers.

METHODS

We address the challenge to combine a global motion model that takes the physical cutting process of the tissue into account with image data that is not simultaneously globally available. Typical approaches either reduce the amount of data to be processed or partition the data into smaller chunks to be processed separately. Our novel method first registers the complete images on a low resolution with a nonlinear deformation model and later refines this result on patches by using a second nonlinear registration on each patch. Finally, the deformations computed on all patches are combined by interpolation to form one globally smooth nonlinear deformation. The NGF distance measure is used to handle multistain images.

RESULTS

The method is applied to ten whole slide image pairs of human lung cancer data. The alignment of 85 corresponding structures is measured by comparing manual segmentations from neighboring slides. Their offset improves significantly, by at least 15%, compared to the low-resolution nonlinear registration.

CONCLUSION/SIGNIFICANCE

The proposed method significantly improves the accuracy of multistain registration which allows us to compare different antibodies at cell level.

Non-rigid landmark-based large-scale image registration in 3-D reconstruction of mouse and rat kidney nephrons

BACKGROUND

Serial histological sections are suffering from mechanical distortions that disturb the reconstruction of 3-D objects. We have corrected such artifacts with a non-rigid landmark-based method that respects the original geometry in the tissue block. The method is exemplified on a large scale in the registration of semi-thin serial sections of the mouse and rat kidneys, and has been tested on FFPE-sections.

AIM

In this study of mouse and rat kidneys, we have measured and characterized the deformations introduced in the preparation of 2.5-μm-thick Epon sections and then eliminated them by a landmark-based non-rigid transformation (NRT).

METHODS

We obtained 2.5-μm-thick serial Epon sections from three mouse kidneys and three rat kidneys for 3-D reconstruction of the nephron tubules. First, the images from 3000 serial mouse and 13,000 serial rat sections underwent a classic rigid registration (CRR), and the distortions were measured and indexed. The section images underwent a further NRT in order to compensate for the deformations. The NRT used is a classic interactive landmark-based approach. The quality of the NRT was verified by comparing the geometry of the transformed images with corresponding block images.

RESULTS

After CRR, the 2.5-μm-thick sections had a linear deformation of up to 2%, the tubular lengths were overestimated with up to 1.5×, and it was most difficult to trace the tubules from section to section. After the additional NRT, the geometry of the images reflected the original geometry in the block, the tubular lengths were no longer overestimated, and the NRT highly facilitated the tracing of the tubular system.

CONCLUSIONS

NRT has facilitated the tracing of the tubular system in kidneys, a tracing, which would otherwise have been most difficult to perform. NRT has yielded substantial new knowledge to segmental and spatial nephron organization in the mouse and rat kidneys.

3D prostate histology reconstruction: an evaluation of image-based and fiducial-based algorithms

PURPOSE

Evaluation of in vivo prostate imaging modalities for determining the spatial distribution and aggressiveness of prostate cancer ideally requires accurate registration of images to an accepted reference standard, such as histopathological examination of radical prostatectomy specimens. Three-dimensional (3D) reconstruction of prostate histology facilitates these registration-based evaluations by reintroducing 3D spatial information lost during histology processing. Because the reconstruction accuracy may constrain the clinical questions that can be answered with these data, it is important to assess the tradeoffs between minimally disruptive methods based on intrinsic image information and potentially more robust methods based on extrinsic fiducial markers.

METHODS

Ex vivo magnetic resonance (MR) images and digitized whole-mount histology images from 12 radical prostatectomy specimens were used to evaluate four 3D histology reconstruction algorithms. 3D reconstructions were computed by registering each histology image to the corresponding ex vivo MR image using one of two similarity metrics (mutual information or fiducial registration error) and one of two search domains (affine transformations or a constrained subset thereof). The algorithms were evaluated for accuracy using the mean target registration error (TRE) computed from homologous intrinsic point landmarks (3-16 per histology section; 232 total) identified on histology and MR images, and for the sensitivity of TRE to rotational, translational, and scaling initialization errors.

RESULTS

The algorithms using fiducial registration error and mutual information had mean ± standard deviation TREs of 0.7 ± 0.4 and 1.2 ± 0.7 mm, respectively, and one algorithm using fiducial registration error and affine transforms had negligible sensitivities to initialization errors. The postoptimization values of the mutual information-based metric showed evidence of errors due to both the optimizer and the similarity metric, and variation of parameters of the mutual information-based metric did not improve its performance.

CONCLUSIONS

The extrinsic fiducial-based algorithm had lower mean TRE and lower sensitivity to initialization than the intrinsic intensity-based algorithm using mutual information. A model relating statistical power to registration error for certain imaging validation study designs estimated that a reconstruction algorithm with a mean TRE of 0.7 mm would require 27% fewer subjects than the method used to initialize the algorithms (mean TRE 1.3 ± 0.7 mm), suggesting the choice of reconstruction technique can have a substantial impact on the design of imaging validation studies, and on their overall cost.

Three-dimensional reconstruction of the rat nephron

This study gives a three-dimensional (3D) structural analysis of rat nephrons and their connections to collecting ducts. Approximately 4,500 2.5-μm-thick serial sections from the renal surface to the papillary tip were obtained from each of 3 kidneys of Wistar rats. Digital images were recorded and aligned into three image stacks and traced from image to image. Short-loop nephrons (SLNs), long-loop nephrons (LLNs), and collecting ducts (CDs) were reconstructed in 3D. We identified a well-defined boundary between the outer stripe and the inner stripe of the outer medulla corresponding to the transition of descending thick limbs to descending thin limbs and between the inner stripe and the inner medulla, i.e., the transition of ascending thin limbs into ascending thick limbs of LLNs. In all nephrons, a mosaic pattern of proximal tubule (PT) cells and descending thin limb (DTL) cells was observed at the transition between the PT and the DTL. The course of the LLNs revealed tortuous proximal "straight" tubules and winding of the DTLs within the outer half of the inner stripe. The localization of loop bends of SLNs in the inner stripe of the outer medulla and the bends of LLNs in the inner medulla reflected the localization of their glomeruli; i.e., the deeper the glomerulus, the deeper the bend. Each CD drained approximately three to six nephrons with a different pattern than previously established in mice. This information will provide a basis for evaluation of structural changes within nephrons as a result of physiological or pharmaceutical intervention.

Spatial organization of the vascular bundle and the interbundle region: three-dimensional reconstruction at the inner stripe of the outer medulla in the mouse kidney

The vascular bundle (VB) is a complex structure that resides in the inner stripe of the outer medulla. At present, the tubulovascular spatial organization of the VB, which is crucial for the formation of the osmolarity gradient and for solute transport, is still under debate. In this study, we used computer-assisted digital tracing combined with aquaporin-1 immunohistochemistry to reconstruct all tubules and vessels in the VB of the mouse kidney. We found, first, that the descending and ascending vasa recta travelled exclusively through the VB. The ascending vasa recta received no tributaries (no branches) along their entire path in the medulla and were not connected with the capillary plexus in the interbundle region. Second, a specific group of the descending vasa recta were closely accompanied by the longest ascending vasa recta, which connected only to the capillary plexus at the tip of the papilla. Third, the descending thin limbs of all short-looped nephrons travelled exclusively through the outer part of the VB. The loops of these nephrons (both descending and ascending parts) were distributed in a regular pattern based on their length. Finally, the thick ascending limbs of all long-looped nephrons were located at the margin of the VB (except a few within the VB), which formed a layer separating the VB from the interbundle region. In conclusion, our three-dimensional analysis of the VB strongly suggest a lateral osmolarity heterogeneity across the inner stripe of the outer medulla, which might work as a driving force for water and solute transport.

Direct physical contact between intercalated cells in the distal convoluted tubule and the afferent arteriole in mouse kidneys

Recent physiological studies in the kidney proposed the existence of a secondary feedback mechanism termed ‘crosstalk’ localized after the macula densa. This newly discovered crosstalk contact between the nephron tubule and its own afferent arteriole may potentially revolutionize our understanding of renal vascular resistance and electrolyte regulation. However, the nature of such a crosstalk mechanism is still debated due to a lack of direct and comprehensive morphological evidence. Its exact location along the nephron, its prevalence among the different types of nephrons, and the type of cells involved are yet unknown. To address these issues, computer assisted 3-dimensional nephron tracing was applied in combination with direct immunohistochemistry on plastic sections and electron microscopy. ‘Random’ contacts in the cortex were identified by the tracing and excluded. We investigated a total of 168 nephrons from all cortical regions. The results demonstrated that the crosstalk contact existed, and that it was only present in certain nephrons (90% of the short-looped and 75% of the long-looped nephrons). The crosstalk contacts always occurred at a specific position – the last 10% of the distal convoluted tubule. Importantly, we demonstrated, for the first time, that the cells found in the tubule wall at the contact site were always type nonA-nonB intercalated cells. In conclusion, the present work confirmed the existence of a post macula densa physical crosstalk contact.

Interactive 3D computer model of the human corneolimbal region: crypts, projections and stem cells

PURPOSE

This study aims to clarify the existence of and to map the localization of different proposed stem cell niches in the corneal limbal region.

MATERIALS AND METHODS

One human eye was cut into 2200 consecutive sections. Every other section was stained with haematoxylin and eosin, digitized at low and high magnification, aligned, 3D reconstructed and visualized using interactive 3D visualization software. The visualization software has interactive tools that make free rotations in all directions possible and makes it possible to create virtual sections independent of the original cutting plan. In all, one low-magnification and 24 high-magnification interactive 3D models were created. Immunohistochemistry against stem cell markers p63 and ΔNp63α was performed as a supplement to the 3D models.

RESULTS

Using the interactive 3D models, we identified three types of stem cell niches in the limbal region: limbal epithelial crypts (LECs), limbal crypts (LCs) and focal stromal projections (FSPs). In all, eight LECs, 25 LCs and 105 FSPs were identified in the limbal region. The LECs, LCs and FSPs were predominantly located in the superior limbal region with seven LECs, 19 LCs and 93 FSPs in the superior limbal region and one LEC, six LCs and 12 FSPs in the inferior limbal region. Only few LECs, LCs and FSPs were localized nasally and temporally.

CONCLUSION

Interactive 3D models are a powerful tool that may help to shed more light on the existence and spatial localization of the different stem cell niches (LECs, LCs and FSPs) in the corneal limbal region.

Imaging heart development using high-resolution episcopic microscopy

Development of the heart in vertebrate embryos is a complex process in which the organ is continually remodelled as chambers are formed, valves sculpted and connections established to the developing vascular system. Investigating the genetic programmes driving these changes and the environmental factors that may influence them is critical for our understanding of congenital heart disease. A recurrent challenge in this work is how to integrate studies as diverse as those of cardiac gene function and regulation with an appreciation of the localised interactions between cardiac tissues not to mention the manner in which both may be affected by cardiac function itself. Meeting this challenge requires an accurate way to analyse the changes in 3D morphology of the developing heart, which can be swift or protracted and both dramatic or subtle in consequence. Here we review the use of high-resolution episcopic microscopy as a simple and effective means to examine organ structure and one that allows modern computing methods pioneered by clinical imaging to be applied to the embryonic heart.

JULIDE: a software tool for 3D reconstruction and statistical analysis of autoradiographic mouse brain sections

In this article we introduce JULIDE, a software toolkit developed to perform the 3D reconstruction, intensity normalization, volume standardization by 3D image registration and voxel-wise statistical analysis of autoradiographs of mouse brain sections. This software tool has been developed in the open-source ITK software framework and is freely available under a GPL license. The article presents the complete image processing chain from raw data acquisition to 3D statistical group analysis. Results of the group comparison in the context of a study on spatial learning are shown as an illustration of the data that can be obtained with this tool.

Large Deformation Diffeomorphic Metric Mapping Registration of Reconstructed 3D Histological Section Images and in vivo MR Images

Our current understanding of neuroanatomical abnormalities in neuropsychiatric diseases is based largely on magnetic resonance imaging (MRI) and post mortem histological analyses of the brain. Further advances in elucidating altered brain structure in these human conditions might emerge from combining MRI and histological methods. We propose a multistage method for registering 3D volumes reconstructed from histological sections to corresponding in vivo MRI volumes from the same subjects: (1) manual segmentation of white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF) compartments in histological sections, (2) alignment of consecutive histological sections using 2D rigid transformation to construct a 3D histological image volume from the aligned sections, (3) registration of reconstructed 3D histological volumes to the corresponding 3D MRI volumes using 3D affine transformation, (4) intensity normalization of images via histogram matching, and (5) registration of the volumes via intensity based large deformation diffeomorphic metric (LDDMM) image matching algorithm. Here we demonstrate the utility of our method in the transfer of cytoarchitectonic information from histological sections to identify regions of interest in MRI scans of nine adult macaque brains for morphometric analyses. LDDMM improved the accuracy of the registration via decreased distances between GM/CSF surfaces after LDDMM (0.39 ± 0.13 mm) compared to distances after affine registration (0.76 ± 0.41 mm). Similarly, WM/GM distances decreased to 0.28 ± 0.16 mm after LDDMM compared to 0.54 ± 0.39 mm after affine registration. The multistage registration method may find broad application for mapping histologically based information, for example, receptor distributions, gene expression, onto MRI volumes.

Towards ultra-high resolution fibre tract mapping of the human brain - registration of polarised light images and reorientation of fibre vectors

Polarised light imaging (PLI) utilises the birefringence of the myelin sheaths in order to visualise the orientation of nerve fibres in microtome sections of adult human post-mortem brains at ultra-high spatial resolution. The preparation of post-mortem brains for PLI involves fixation, freezing and cutting into 100-μm-thick sections. Hence, geometrical distortions of histological sections are inevitable and have to be removed for 3D reconstruction and subsequent fibre tracking. We here present a processing pipeline for 3D reconstruction of these sections using PLI derived multimodal images of post-mortem brains. Blockface images of the brains were obtained during cutting; they serve as reference data for alignment and elimination of distortion artefacts. In addition to the spatial image transformation, fibre orientation vectors were reoriented using the transformation fields, which consider both affine and subsequent non-linear registration. The application of this registration and reorientation approach results in a smooth fibre vector field, which reflects brain morphology. PLI combined with 3D reconstruction and fibre tracking is a powerful tool for human brain mapping. It can also serve as an independent method for evaluating in vivo fibre tractography.

Brain maps on the go: functional imaging during motor challenge in animals

Brain mapping in the freely moving animal is useful for studying motor circuits, not only because it avoids the potential confound of sedation or restraints, but because activated brain states may serve to accentuate differences that only manifest partially while a subject is in the resting state. Perfusion or metabolic mapping using autoradiography allows one to examine changes in brain function at the circuit level across the entire brain with a spatial resolution (approximately 100 micro) appropriate for the rat or mouse brain, and a temporal resolution (seconds-minutes) sufficient for capturing acute brain changes. Here we summarize the application of these methods to the functional brain mapping of behaviors involving locomotion of small animals, methods for the three-dimensional reconstruction of the brain from autoradiographic sections, voxel based analysis of the whole brain, and generation of maps of the flattened rat cortex. Application of these methods in animal models promises utility in improving our understanding of motor function in the normal brain, and of the effects of neuropathology and treatment interventions such as exercise have on the reorganization of motor circuits.

Three-dimensional reconstruction of stained histological slices and 3D non-linear registration with in-vivo MRI for whole baboon brain

The correlation between post-mortem data and in-vivo brain images is of high interest for studying neurodegenerative diseases. This paper describes a protocol that matches a series of stained histological slices of a baboon brain with an anatomical MRI scan of the same subject using an intermediate 3D-consistent volume of "blockface" photographs taken during the sectioning process. Each stained histological section of the baboon brain was first registered to its corresponding blockface photograph using a novel "hemi-rigid" transformation. This piecewise rigid 2D transformation was specifically adapted to the registration of slices which contained both hemispheres. Subsenquently, to correct the global 3D deformations of the brain caused by histological preparation and fixation, a 3D elastic transformation was estimated between the blockface volume and the MRI data. This 3D elastic transformation was then applied to the histological volume previously aligned using the hemi-rigid method to complete the registration of the series of stained histological slices with the MRI data. We assessed the efficacy of our method by evaluating the quality of matching of anatomical features as well as the difference of volume measurements between the MRI and the histological images. Two complete baboon brains (with the exception of cerebellum) were successfully processed using our protocol.

Three-dimensional reconstruction of the mouse nephron

Renal function is crucially dependent on renal microstructure which provides the basis for the regulatory mechanisms that control the transport of water and solutes between filtrate and plasma and the urinary concentration. This study provides new, detailed information on mouse renal architecture, including the spatial course of the tubules, lengths of different segments of nephrons, histotopography of tubules and vascular bundles, and epithelial ultrastructure at well-defined positions along Henle's loop and the distal convolution of nephrons. Three-dimensional reconstruction of 200 nephrons and collecting ducts was performed on aligned digital images, obtained from 2.5-mum-thick serial sections of mouse kidneys. Important new findings were highlighted: (1) A tortuous course of the descending thin limbs of long-looped nephrons and a winding course of the thick ascending limbs of short-looped nephrons contributed to a 27% average increase in the lengths of the corresponding segments, (2) the thick-walled tubules incorporated in the central part of the vascular bundles in the inner stripe of the outer medulla were identified as thick ascending limbs of long-looped nephrons, and (3) three types of short-looped nephron bends were identified to relate to the length and the position of the nephron and its corresponding glomerulus. The ultrastructure of the tubule segments was identified and suggests important implications for renal transport mechanisms that should be considered when evaluating the segmental distribution of water and solute transporters within the normal and diseased kidney.

Statistical parametric mapping applied to an autoradiographic study of cerebral activation during treadmill walking in rats

Autoradiographs are conventionally analyzed by a region-of-interest (ROI) analysis. However, definition of ROIs on an image set is labor intensive, is subject to potential inter-rater bias, and is not well suited for anatomically variable structures that may not consistently correspond to specific ROIs. Most importantly, the ROI method is poorly suited for whole-brain analysis, where one wishes to detect all activations resulting from an experimental paradigm. A system developed for analysis of imaging data in humans, Statistical Parametric Mapping (SPM), avoids some of these limitations but has not previously been adapted as a tool for the analysis of autoradiographs. Here, we describe the application of SPM to an autoradiographic data set mapping cerebral activation in rats during treadmill walking. We studied freely moving, non-tethered rats that received injections of the cerebral blood flow tracer [14C]-iodoantipyrine, while they were performing a treadmill task (n = 7) or during a quiescent control condition (n = 6). Results obtained with SPM were compared to those previously reported using a standard ROI-based method of analysis [J. Cereb. Blood Flow Metab. 23(2003) 925]. The SPM method confirmed most areas detected as significant using the ROI approach. However, in the subcortex, SPM detected additional significant regions that, because of their irregular structures, fell short of statistical significance when analyzed by ROI. The SPM approach offers the ability to perform a semi-automated whole-brain analysis, and coupled with autoradiography, provides an effective means to globally localize functional activity in small animals.

Reconstruction of serially acquired slices using physics-based modeling

This paper presents an accurate, computationally efficient, fast, and fully automated algorithm for the alignment of two-dimensional (2-D) serially acquired sections forming a 3-D volume. The approach relies on the determination of interslice correspondences. The features used for correspondence are extracted by a 2-D physics-based deformable model parameterizing the object shape. Correspondence affinities and global constrains render the method efficient and reliable. The method accounts for one of the major shortcomings of 2-D slices alignment of a 3-D volume, namely variable and nonuniform thickness of the slices. Moreover, no particular alignment direction is privileged, avoiding global offsets, biases, and error propagation. The method was evaluated on real images and the experimental results demonstrated its accuracy, as reconstruction errors were smaller than 1 degree in rotation and smaller than 1 pixel in translation.

A global energy function for the alignment of serially acquired slices

An accurate, computationally efficient, and fully automated algorithm for the alignment of two-dimensional (2-D) serially acquired sections forming a three-dimensional (3-D) volume is presented. The approach relies on the optimization of a global energy function, based on the object shape, measuring the similarity between a slice and its neighborhood in the 3-D volume. Slice similarity is computed using the distance transform measure in both directions. No particular direction is privileged in the method avoiding global offsets, biases in the estimation and error propagation. The method was evaluated on real images [medical, biological, and other computerized tomography (CT) scanned 3-D data] and the experimental results demonstrated its accuracy as reconstuction errors are less than one degree in rotation and less than one pixel in translation.

A robust statistics-based global energy function for the alignment of serially acquired autoradiographic sections

Autoradiographic analysis of the functional changes occurring in the rat brain are most often performed on coronal sections that allow a good insight into the events occurring at the structural level but lacks the 3D context which is necessary to fully understand the involvement of the brain structures in specific situations like focal seizures with or without generalization. Therefore a robust, fully-automated algorithm for the registration of serially acquired autoradiographic sections is presented. The method accounts for the main difficulties of autoradiographic alignment: corrupted data (cuts and tears), dissimilarities or discontinuities between slices, non parallel or missing slices. The approach relies on the minimization of a global energy function based on robust statistics. The energy function measures the similarity between a slice and its neighborhood in the 3D volume. No particular direction is privileged in the method, so that global offsets, biases in the estimation or error propagations are avoided. The method is evaluated qualitatively and quantitatively on real autoradiographic data. Rat brain autoradiographic volumes are reconstructed with registration errors less than 1 degree in rotation and less than 1 pixel in translation.

Digital three-dimensional reconstruction and ultrastructure of the mouse proximal tubule

Mice are prime targets of experimental gene modification and have become object of an increasing number of biologic studies in renal physiology, development, and molecular biology. Phenotypic changes in response to gene modification require detailed information on normal structure. However, detailed analyses of normal mouse kidney structure and organization are lacking. This study describes the 3D organization and ultrastructural, segmental variation of the mouse kidney proximal tubule. A total of 160 proximal tubules in three C57/BL/6J mouse kidneys were analyzed on 800 serial sections from each kidney from the surface to the inner stripe of the outer zone of medulla. All tubules were reconstructed in 3D and visualized by interactive computer graphics. A quantitative ultrastructural analysis of the mouse proximal tubule at every 300 to 400 micro m was performed. The 3D representation revealed a distinct organization of the mouse proximal tubule, each occupying a separate domain within the cortex. Superficial proximal tubules have long straight parts converging into clusters within the medullary rays. Tubules originating deeper within the cortex become longer and increasingly tortuous. In the medullary rays, these are arranged in layers outside the clusters of more superficial tubules. In contrast to rat and human kidney, no major segmental variation in the ultrastructure of the proximal tubule was identified, and no parameters enabled definition of distinct segments in this strain of mice. In conclusion, significant new information on the 3D organization of the murine proximal tubule has been obtained. Quantitative, ultrastructural analyses of mouse proximal tubules reveal substantial differences compared with other species.

Extending the resolution of light microscopy and electron microscopy digitized images with reference to cellular changes after in vivo low oxygen exposure

When processing frame-grabbed images from light microscopy (LM) and electron microscopy (EM), even a state-of-the-art digital camera is the weakest link between the microscope and the image processor. Details, which can be seen directly in the ocular at LM and in a negative recorded at EM, will not necessarily be represented in the frame-grabbed images. Because of this, there is a tendency to prefer a higher magnification at the expense of overview, i.e. only smaller areas are described. We find that the inadequacy of the camera can be overcome by taking multiple images of the same object, and align, expand, and add them into a more highly resolved image. At the LM level, the method has proved useful for describing the relation of the zinc pattern versus local in vivo oxygen measurements. At the EM level, we show that it is possible to achieve information about the spatial conditions in a given area, and the method may have applications on, e.g., visualization of ultra-small antibody-bound gold particles. The method can be performed on color and black and white images at any magnification and it has been tested in Adobe Photoshop (4.0 and higher) in WINDOWS 95, 98, and 2000.

Use of "reference series" to realign histological serial sections for three-dimensional reconstructions of the positions of cellular events in the developing brain

The present study demonstrates how, predominantly by external fiducials, histological serial sections used to reconstruct patterns of individually marked cellular events in large organs or whole embryos can be realigned with the help of "reference series". Resin-embedded embryos were cut at 1 microm and consecutive sections were alternately placed on two sets of slides. For cytological diagnosis and acquisition of embryonic contours, stained sections of the first series, termed "working series", were scanned with the x 100 objective using "Huge Image", a recently established image acquisition system. For acquisition of the contours of the resin block, adjacent unstained sections of the second series, termed "reference series", were scanned with the x 5 objective. Thereafter, "hybrid sections" were created which combined vectorized embryonic contours and cellular events taken from the working series with vectorized block contours taken from the reference series. For realignment, consecutive "hybrid sections" were matched by best-fit of the block contours. Stacks of realigned "hybrid sections" were shaped like truncated pyramids and, thus, reflected repeated "trimming" of the resin block during the sectioning procedure. Among 266 "hybrid sections" at intervals of 8 microm, needed to reconstruct the brain of a 15-day-old embryo of Tupaia belangeri (Scandentia), internal fiducials were required five times for realigning a total of six adjacent truncated pyramids. Application of this method provided realistic reconstructions of the positions of apoptotic cells in the entire developing brain without the need of secondary introduction of external fiducials.

A 3D fiber model of the human brainstem

A new neuroanatomic approach to evaluate the fiber orientation in gross histological sections of the human brain was developed. Serial sections of a human brainstem were used to derive fiber orientation maps by analysis of polarized light sequences of these sections. Fiber inclination maps visualize angles of inclination, and fiber direction maps show angles of direction. These angles define vectors which can be visualized as RGB-colors. The serial sections were aligned to each other using the minimized Euclidian distance as fit criterion. In the 3D data set of the human brainstem the major fiber tracts were segmented, and three-dimensional models of these fiber tracts were generated. The presented results demonstrate that two kinds of fiber atlases are feasible: a fiber orientation atlas representing a vector in each voxel, which shows the nerve fiber orientation, and a volume-based atlas representing the major fiber tracts. These models can be used for the evaluation of diffusion tensor data as well as for neurosurgical planning.

A new method for reliable and efficient reconstruction of 3-dimensional images from autoradiographs of brain sections

Valuable information on metabolism and function of distinct brain regions can be extracted from autoradiographs of 2D brain sections, e.g. after labelling with the non-metabolisable sugar derivate [14C]-2-fluoro-deoxyglucose (2-FDG). For a more complete comprehension of the data and for a comparison with information obtained by modern functional imaging techniques it is essential to have a reliable and efficient method for the 3D reconstruction of autoradiographs of serial sections. This paper describes a new method for the alignment of 2D 2-FDG images, that combines two established algorithms, i.e. principal axes alignment followed by consistent matrix transformation. The power and efficiency of this new 2-step method is compared to those of various previously described procedures.

Computer-assisted visualization of the rat epididymis: a methodological study based on paraffin sections autometallographically stained for zinc ions

A concept for the computer-assisted visualization of tubular organs is presented. Unmarked histological zinc-stained serial sections from the epididymis of the Wistar rat were aligned to demonstrate the concept. Virtual images were made through the aligned sections and served as controls for the alignment process. Animation of the serial sections and the virtual images revealed new information about the structure of the organ under investigation. The analysis was used to upgrade the anatomical knowledge of rat epididymis by describing how the epididymal duct runs through the structure. The proximal parts of the epididymis contain large communicating septa of connective tissue dividing the caput and the upper part of the corpus epididymidis into segments. The tortuousness was high in the caput with many turns within a small area of the epididymis, whereas longer loops were found in the lower part of the corpus and cauda epididymidis. The tube of the vas deferens was found to become an integrated part of the ductal system in the cauda epididymidis, although it was histologically easy to distinguish from the epididymal duct. The total number of cross-sections of the ductus epididymidis in the 2254, 15-microm-thick, tissue sections analysed was 104700, giving a minimum length of the ductal system of 1.5 m.

PC-assisted three-dimensional description of organs containing tubular structures, applied on the epididymis of the rat

A concept for three-dimensional computer-assisted reconstruction of tubular organs, e.g. the epididymis, is described. Histologic serial sections without artificial landmarks from the epididymis of the Wistar rat were aligned. Virtual images through the aligned sections served as a control of the alignment process and can reveal new information about the structure of the organ under investigation. The method can be used for improving the anatomical description of the epididymis, i.e. how the ductus epididymidis is coiled along this organ. Other tubular tissues and organs can be investigated and analysed with this PC-assisted method, e.g. testis and kidney.

Optical slicing and 3-D characterization of hippocampal capillaries in the rat visualized by autometallographic silver enhancement of colloidal gold particles

In order to visualize the vascular system of the rat brain, 10 Wistar rats were perfused transcardially with glutaraldehyde and a 40 degrees C gold-gelatine solution. The brains were post-fixed with glutaraldehyde and vibratomized into 100-micron-thick slices, and the gold particles were developed by autometallography. In this way, the colloidal gold particles in the vessels became encased in silver and thereby made visible. The developed gold staining is stable and does not interfere with further dehydration and counterstaining. Images were frame grabbed during optical slicing, and classic stereograms and 'shadow' 3-D images were produced. We found a high variation of capillary density in the hippocampal region reflecting known subregional structures. The silver-enhanced vessels acted as natural markers and made it possible to study and measure aspects of the complexity of dehydration and staining artifacts. We found a non-linear shrinking of 13-17% in the x- and y-directions and a spatial shrinking up to 50% in some regions after the dehydration and staining process. This observation may be of interest not only in relation to tissue subjected to this fixation protocol but also to other fixation procedures. The gold-gelatine autometallographic technique and the present stereograms can release data for stereological use as well.

Distinct differences in partial oxygen pressure at micrometer ranges in the rat hippocampal region

A mapping at micrometer ranges of the partial oxygen pressure in the rat hippocampus was performed. The oxygen tension in the rat hippocampal region was measured using a glass oxygen microsensor in 30-microm steps along straight lines at a set of stereotactic coordinates. In the hippocampus the pattern of the oxygen tensions reflected the autometallographic zinc sulphide (AMG(ZnS)) pattern, i.e. the pattern of zinc enriched (ZEN) terminals. The highest levels of oxygen tension were recorded in the areas that are most heavily stained with the autometallographic zinc sulphide (AMG(ZnS)) method, like hilus fasciae dentatae. The zinc ions located in synaptic vesicles of the ZEN terminals can also be demonstrated by AMG silver amplification in brains from animals in vivo treated with sodium selenite. This method depends on the presence of a substantial reduction capacity of the tissues as selenite ions (SeO(2)(3)-) must to be reduced to selenide ions (Se2-) before the catalytic zinc selenide crystals can be formed. At some point, either during the transport from the infusion site to the actual target tissue or in the target tissue itself, selenium is reduced from Se(+ IV) to Se(- II). The importance of the reduction capacity of the target tissue in this process is demonstrated by the fact that areas found to have the highest concentration of zinc ions, e.g. hilus fasciae dentatae and the mossy fibres of CA3, are almost unstained after 1 h of i.p. Na2SeO3 exposure. An explanation of this phenomenon could be that the reduction process Se(+ IV) <==> Se(- II) leading to the formation of Se2- is moved to the left by the presence of oxygen, thus inhibiting the precipitation of ZnSe crystals. It is suggested that the subtle oxygen pressure pattern found in the rat hippocampus might also reflect essential biological zinc-related mechanisms vital to brain function.

A comparison of synapses onto the somata of intrinsically bursting and regular spiking neurons in layer V of rat SmI cortex

Regular spiking (RS) and intrinsically bursting (IB) neurons show distinct differences in their inhibitory responses. Under various conditions, the synaptic responses of RS cells display marked inhibitory postsynaptic potentials (IPSPs), whereas the responses of most IB cells do not (Silva et al: Soc Neurosci Abstr 14:883, 1988; Chagnac-Amitai and Connors: J Neurophysiol 61:747, 62:1149, 1989; Connors and Gutnick: TINS 13:99, 1990). This investigation is designed to determine if differences in the inhibitory responses of RS versus IB cells are reflected in differences in the concentration of inhibitory synapses onto their somata. RS and IB neurons in rat somatosensory cortex were identified by using intracellular recording and labeling, examined with the light microscope, and then serial thin-sectioned prior to examination with the electron microscope. Axonal terminals presynaptic to their somata and proximal dendrites were identified and classified according to criteria described by Peters and coworkers (Peters et al: J Neurocytol 19:584, 1990; Peters and Harriman: J Neurocytol 19:154, 1990; 21:679, 1992). The locations of these boutons were displayed on the surfaces of 3-D reconstructions of the somata and proximal dendrites. The reconstructions were produced directly from the serial thin sections by using a novel, electron microscopic, image-processing computer resource. Our analysis showed no significant difference in the types and concentration of boutons presynaptic to the cell bodies and proximal dendrites of intrinsically bursting versus regular spiking neurons. We conclude that the differences observed in the inhibitory responses of intrinsically bursting versus regular spiking neurons cannot be explained by differences in the concentrations of synapses onto their somata.