The performance of machine learning algorithms, when used for segmenting 3D biomedical images, does not reach the level expected based on results achieved with 2D photos. This may be explained by the comparative lack of high-volume, high-quality training datasets, which require state-of-the-art imaging facilities, domain experts for annotation and large computational and personal resources. The HR-Kidney dataset presented in this work bridges this gap by providing 1.7 TB of artefact-corrected synchrotron radiation-based X-ray phase-contrast microtomography images of whole mouse kidneys and validated segmentations of 33 729 glomeruli, which corresponds to a one to two orders of magnitude increase over currently available biomedical datasets. The image sets also contain the underlying raw data, threshold- and morphology-based semi-automatic segmentations of renal vasculature and uriniferous tubules, as well as true 3D manual annotations. We therewith provide a broad basis for the scientific community to build upon and expand in the fields of image processing, data augmentation and machine learning, in particular unsupervised and semi-supervised learning investigations, as well as transfer learning and generative adversarial networks.

Distraction osteogenesis is a clinically established technique for lengthening, molding and shaping bone by new bone formation. The experimental evaluation of this expensive and time-consuming treatment is of high impact for better understanding of tissue engineering but mainly relies on a limited number of histological slices. These tissue slices contain two-dimensional information comprising only about one percent of the volume of interest. In order to analyze the soft and hard tissues of the entire jaw of a single rat in a multimodal assessment, we combined micro computed tomography (µCT) with histology. The µCT data acquired before and after decalcification were registered to determine the impact of decalcification on local tissue shrinkage. Identification of the location of the H&E-stained specimen within the synchrotron radiation-based µCT data collected after decalcification was achieved via non-rigid slice-to-volume registration. The resulting bi- and tri-variate histograms were divided into clusters related to anatomical features from bone and soft tissues, which allowed for a comparison of the approaches and resulted in the hypothesis that the combination of laboratory-based µCT before decalcification, synchrotron radiation-based µCT after decalcification and histology with hematoxylin-and-eosin staining could be used to discriminate between different types of collagen, key components of new bone formation.

Atherosclerotic arteries exhibit characteristic constrictions and substantial deviations from cylindrical shape. Therefore, determining the artery’s cross-section along the centerline is challenging, although high-resolution isotropic three-dimensional data are available. Herein, we apply high-resolution computed tomography in absorption and phase to a plaque-containing human artery post-mortem, through the course of the preparation stages for histology. We identify the impact of paraffin embedding and decalcification on the artery lumen. For automatic extraction of lumen’s cross-section along centerline we present a dedicated pipeline. Comparing fixated tissue before and after paraffin embedding gives rise to shape changes with lumen reduction to 50–80%. The histological slicing induces further deformations with respect to tomography. Data acquired after decalcification show debris unintentionally distributed within the vessel preventing the reliable automatic lumen segmentation. Comparing tomography of laboratory- and synchrotron-radiation-based X rays by means of joint histogram analysis leads us to conclude that advanced desktop tomography is capable of quantifying the artery’s lumen as an essential input for blood flow simulations. The results indicate that the most reliable lumen quantification is achieved by imaging the non-decalcified specimen fixed in formalin, using phase contrast modality and a dedicated processing pipeline. This study focusses on a methodology to quantitatively evaluate diseased artery segments post-mortem and provides unique structural parameters on the treatment-induced local shrinkage, which will be the basis of future studies on the flow in vessels affected by constrictions.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

There have been great efforts on the nanoscale 3D probing of brain tissues to image subcellular morphologies. However, limitations in terms of tissue coverage, anisotropic resolution, stain dependence, and complex sample preparation all hinder achieving a better understanding of the human brain functioning in the subcellular context. Herein, X-ray nanoholotomography is introduced as an emerging synchrotron radiation-based technology for large-scale, label-free, direct imaging with isotropic voxel sizes down to 25 nm, exhibiting a spatial resolution down to 88 nm. The procedure is nondestructive as it does not require physical slicing. Hence, it allows subsequent imaging by complementary techniques, including histology. The feasibility of this 3D imaging approach is demonstrated on human cerebellum and neocortex specimens derived from paraffin-embedded tissue blocks. The obtained results are compared to hematoxylin and eosin stained histological sections and showcase the ability for rapid hierarchical neuroimaging and automatic rebuilding of the neuronal architecture at the level of a single cell nucleolus. The findings indicate that nanoholotomography can complement microscopy not only by large isotropic volumetric data but also by morphological details on the sub-100 nm level, addressing many of the present challenges in brain tissue characterization and probably becoming an important tool in nanoanatomy.

In article number 1700694, Bert Müller and co-workers launch nano-holotomography for the isotropic 3D imaging of paraffin-embedded human brain tissues. Nano-holotomography bridges the spatial resolution gap between optical and electron microscopy, while giving access to 3D data for large tissue volumes with a spatial resolution well below 100 nm. The hierarchical approach will mediate prospering nano-anatomy.

Lewy body disease (LBD) is the second most common cause of neurodegenerative dementia after Alzheimer's disease and presents a challenge to healthcare services in an ageing society because current therapeutic approaches do not halt the disease's progress. It has been suggested that aggregation of the protein α-synuclein plays a critical role in LBD. One pathway for aggregation involves the creation of low molecular weight α-synuclein oligomers, which have been observed to induce cell death. Precise location of Lewy bodies and their precursors could allow researchers to better relate their presence to clinical symptoms as well as infer their physiological effects. While conventional light microscopy of histological slices can visualize Lewy bodies and larger protein aggregates, it provides only two-dimensional information and diffraction-limited resolution prevents visualization of the smaller (approximately 50 nanometer) protein bundles. For this reason, we propose hard X-ray micro- and nano-tomography modalities for complementary, non-destructive, three-dimensional visualization of brain tissues taking advantage of developments at synchrotron radiation facilities. As a proof-of-concept, we imaged paraffin-embedded fusiform gyrus tissue (provided by Newcastle Brain Tissue Resource) from a DLB case at the European Synchrotron Radiation Facility (ESRF Grenoble, France provided beamtime at ID19 from proposal MD-1055) using single-distance inline phase-contrast X-ray tomography with pixel size of 1.6 micrometers. A H&E stained histology slice of the same sample was also taken. A correspondence between the micro computed tomography and histology datasets could be made by tracking the orientation of vessels in the specimen. The micro computed tomography demonstrated as shown in figure (left) sufficient contrast and resolution for identification of the lower cortical layer as well as visualization of cells in three-dimensional manner, with confirmation from the annotated histology slice.

Hard X-ray Nano-Holotomography of Formalin-Fixated and Paraffin-Embedded Human Brain Tissue Anna Khimchenko, Alexandra Pacureanu, Christos Bikis, Simone E. Hieber, Peter Thalmann, Hans Deyhle, Gabriel Schweighauser, Jürgen Hench, Stephan Frank, Magdalena Müller-Gerbl, Georg Schulz, Peter Cloetens, and Bert Müller 1. Biomaterials Science Center (BMC), DBE, University of Basel, Allschwil, Switzerland 2. European Synchrotron Radiation Facility (ESRF), Grenoble, France 3. Institute of Pathology, Basel University Hospital, Basel, Switzerland 4. Musculoskeletal Research Group, DBM, University of Basel, Basel, Switzerland # Current Addresses: Department of Dermatology, Harvard Medical School (HMS), Boston, USA and Wellman Center for Photomedicine, Massachusetts General Hospital (MGH), Boston, USA * Corresponding author bert.mueller@unibas.ch

Nearby slices (left) and H& The modalitie m-CT directing tology into the BASED MICROTOMOGRAPHY TO IDENTIFY a-SYNUCLEIN OLIGOMERS IN POSTMORTEM TISSUE Griffin Fischer Rodgers, Daniel Erskine, Ian McKieth, JohnPaul Taylor, Bert Mueller, Christos Biskis, Georg Schulz, Willy Kuo, Simone Hieber, Prabitha Urwyler, University of Basel, Allschwil, Switzerland; Newcastle University, Newcastle upon Tyne, United Kingdom; Newcastle University, Institute of Neuroscience, Newcastle upon Tyne, United Kingdom; University of Bern, Bern, Switzerland. Contact e-mail: prabitha.urwyler@artorg.unibe.ch

SummaryLocalizing a histological section in the three‐dimensional dataset of a different imaging modality is a challenging 2D‐3D registration problem. In the literature, several approaches have been proposed to solve this problem; however, they cannot be considered as fully automatic. Recently, we developed an automatic algorithm that could successfully find the position of a histological section in a micro computed tomography (μCT) volume. For the majority of the datasets, the result of localization corresponded to the manual results. However, for some datasets, the matching μCT slice was off the ground‐truth position. Furthermore, elastic distortions, due to histological preparation, could not be accounted for in this framework.In the current study, we introduce two optimization frameworks based on normalized mutual information, which enabled us to accurately register histology slides to volume data. The rigid approach allocated 81 % of histological sections with a median position error of 8.4 μm in jaw bone datasets, and the deformable approach improved registration by 33 μm with respect to the median distance error for four histological slides in the cerebellum dataset.

Background: We wanted to achieve a three-dimensional (3D), non-destructive imaging and automatic post-analysis and evaluation of reconstructed peripheral nerves without involving cutting and staining processes. New method: We used a laboratory-based micro computed tomography system for imaging, as well as a custom analysis protocol. The sample preparation was also adapted in order to achieve 3D images with true micrometer resolution and suitable contrast. Results: Analysis of the acquired tomograms enabled the quantitative assessment of 3D tissue structures, i.e., surface morphology, nerve fascicles, nerve tissue volume, geometry, and vascular regrowth. The resulting data showed significant differences between operated animals and non-operated controls. Comparison with existing methods: Our approach avoids the sampling error associated with conventional 2D visualization approaches and holds promise for automation of the analysis of large series of datasets. Conclusions: We have presented a potential way for 3D imaging and analysis of entire regenerated nerves non-destructively, paving the way for high-throughput analysis of therapeutic conditions of treating adult nerve injuries. (C) 2017 The Author(s). Published by Elsevier B.V.

Our body is hierarchically organized down to individual cells. Cutting-edge clinical imaging facilities reach a spatial resolution of a fraction of a millimeter, living cells invisible. A decade ago, post-mortem X-ray imaging by means of synchrotron radiation enabled the identification of Os-stained ganglion and unstained Purkinje cells. Very recently, even sub-cellular structures, such as nucleolus and the dendritic tree of Purkinje cells, were extracted by means of phase-contrast single-distance synchrotron radiation-based hard X-ray tomography. At the same time, conventional absorption-contrast, laboratory-based micro computed tomography was successfully applied to visualize brain components including individual Purkinje cells within a cerebellum specimen. Thus, the goal of isotropic-cellular-resolution visualization of soft tissues within a laboratory environment without application of any dedicated contrast agent was achieved. In this communication, we are discussing (1) to which extend the quality gain of the laboratory-based absorption-contrast tomography can be driven with respect to optical microscopy of stained tissue sections and (2) what value such a technique would add. As a proof of principle, four histological sections were affine-registered to corresponding three-dimensional (3D) tomography dataset. We are discussing a semi-automatic landmark-based 2D-3D registration framework and compare registration results based on mean square difference (MSD) metrics.

Background: Histological assessment of peripheral nerve regeneration in animals is tedious, time-consuming and challenging for three-dimensional analysis. New method: The present study reports on how and to what extent micro computed tomography of paraffin-embedded samples can provide a reliable three-dimensional approach for quantitative analysis of peripheral nerves. Results: Rat sciatic nerves were harvested, formalin-fixated, positioned into nerve conduits (NC), paraffin-embedded, and imaged using a laboratory-based X-ray microtomography system with an isotropic voxel length of 4 p.m. Suitable quantitative measures were identified and automatically evaluated, i.e. nerve length, cross-sectional area and volume, as well as vascular structures, to be used as an assessment and comparison indicator of regeneration quality. Comparison with existing methods: Compared to imaging using contrast agents, the investigated specimens can subsequently undergo the conventional histological analysis without requiring additional preparation steps. Contrast and spatial resolution are also increased significantly. Conclusions: We demonstrate the potential of the micro computed tomography for non-destructive monitoring of peripheral nerves inside the conduits. (C) 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license.

Hard X-ray phase contrast imaging techniques have become most suitable for the non-destructive three-dimensional visualization of soft tissues at the microscopic level. Among the hard X-ray grating interferometry methods, a single-grating approach (XSGI) has been implemented by simplifying the established double-grating interferometer (XDGI). We quantitatively compare the XSGI and XDGI tomograms of a human nerve and demonstrate that both techniques provide sufficient contrast to allow for the distinction of tissue types. The two-fold binned data show spatial resolution of (5.2 ± 0.6) μm and (10.7 ± 0.6) μm, respectively, underlying the performance of XSGI in soft tissue imaging.

Brain tissues have been an attractive subject for investigations in neuropathology, neuroscience, and neurobiol- ogy. Nevertheless, existing imaging methodologies have intrinsic limitations in three-dimensional (3D) label-free visualisation of extended tissue samples down to (sub)cellular level. For a long time, these morphological features were visualised by electron or light microscopies. In addition to being time-consuming, microscopic investigation includes specimen fixation, embedding, sectioning, staining, and imaging with the associated artefacts. More- over, optical microscopy remains hampered by a fundamental limit in the spatial resolution that is imposed by the diffraction of visible light wavefront. In contrast, various tomography approaches do not require a complex specimen preparation and can now reach a true (sub)cellular resolution. Even laboratory-based micro computed tomography in the absorption-contrast mode of formalin-fixed paraffin-embedded (FFPE) human cerebellum yields an image contrast comparable to conventional histological sections. Data of a superior image quality was obtained by means of synchrotron radiation-based single-distance X-ray phase-contrast tomography enabling the visualisation of non-stained Purkinje cells down to the subcellular level and automated cell counting. The question arises, whether the data quality of the hard X-ray tomography can be superior to optical microscopy. Herein, we discuss the label-free investigation of the human brain ultramorphology be means of synchrotron radiation-based hard X-ray magnified phase-contrast in-line tomography at the nano-imaging beamline ID16A (ESRF, Grenoble, France). As an example, we present images of FFPE human cerebellum block. Hard X-ray tomography can provide detailed information on human tissues in health and disease with a spatial resolution below the optical limit, improving understanding of the neuro-degenerative diseases.

There is a lack of the necessary methodology for three-dimensional (3D) investigation of soft tissues with cellular resolution without staining or tissue transformation. Synchrotron radiation based hard X-ray in-line phase contrast tomography using single-distance phase reconstruction (SDPR) provides high spatial resolution and density contrast for the visualization of individual cells using a standard specimen preparation and data reconstruction. In this study, we demonstrate the 3D characterization of a formalin-fixed paraffin-embedded (FFPE) human cerebellum specimen by SDPR at the Diamond-Manchester Imaging Branchline I13-2 (Diamond Light Source, UK) at pixel sizes down to 0.45 μm. The approach enables visualization of cerebellar layers (Stratum moleculare and Stratum granulosum), the 3D characterization of individual cells (Purkinje, stellate and granule cells) and can even resolve some subcellular structures (nucleus and nucleolus of Purkinje cells). The tomographic results are qualitatively compared to hematoxylin and eosin (H&E) stained histological sections. We demonstrate the potential benefits of hard X-ray microtomography for the investigations of biological tissues in comparison to conventional histology.

Brain tissue evaluation is essential for gaining in-depth insight into its diseases and disorders. Imaging the human brain in three dimensions has always been a challenge on the cell level. In vivo methods lack spatial resolution, and optical microscopy has a limited penetration depth. Herein, we show that hard X-ray phase tomography can visualise a volume of up to 43 mm(3) of human post mortem or biopsy brain samples, by demonstrating the method on the cerebellum. We automatically identified 5,000 Purkinje cells with an error of less than 5% at their layer and determined the local surface density to 165 cells per mm(2) on average. Moreover, we highlight that three-dimensional data allows for the segmentation of sub-cellular structures, including dendritic tree and Purkinje cell nucleoli, without dedicated staining. The method suggests that automatic cell feature quantification of human tissues is feasible in phase tomograms obtained with isotropic resolution in a label-free manner.

Micro computed tomography (mCT) is well established in virtually all fields of biomedical research, allowing for the non-destructive volumetric visualization of tissue morphology. A variety of specimens can be investigated, ranging from soft to hard tissue to engineered structures like scaffolds. Similarly, the size of the objects of interest ranges from a fraction of a millimeter to several tens of centimeters. While synchrotron radiation-based μCT still offers unrivaled data quality, the ever-improving technology of cathodic tube-based machines offers a valuable and more accessible alternative. The Biomaterials Science Center of the University of Basel operates a nanotomOR m (phoenix|x-ray, GE Sensing and Inspection Technologies GmbH, Wunstorf, Germany), with a 180 kV source and a minimal spot size of about 0.9 μm. Through the adjustable focus-specimen and focus-detector distances, the effective pixel size can be adjusted from below 500 nm to about 80 μm. On the high-resolution side, it is for example possible to visualize the tubular network in sub-millimeter thin dentin specimens. It is then possible to locally extract parameters such as tubule diameter, density, or alignment, giving information on cell movements during tooth formation. On the other side, with a horizontal shift of the 3,072 pixels x 2,400 pixels detector, specimens up to 35 cm in diameter can be scanned. It is possible, for example, to scan an entire human knee, albeit with inferior resolution. Lab source μCT machines are thus a powerful and flexible tool for the advancement of biomedical research, and a valuable and more accessible alternative to synchrotron radiation facilities.

Cell visualization and counting plays a crucial role in biological and medical research including the study of neurodegenerative diseases. The neuronal cell loss is typically determined to measure the extent of the disease. Its characterization is challenging because the cell density and size already differs by more than three orders of magnitude in a healthy cerebellum. Cell visualization is commonly performed by histology and fluorescence microscopy. These techniques are limited to resolve complex microstructures in the third dimension. Phase- contrast tomography has been proven to provide sufficient contrast in the three-dimensional imaging of soft tissue down to the cell level and, therefore, offers the basis for the three-dimensional segmentation. Within this context, a human cerebellum sample was embedded in paraffin and measured in local phase-contrast mode at the beamline ID19 (ESRF, Grenoble, France) and the Diamond Manchester Imaging Branchline I13-2 (Diamond Light Source, Didcot, UK). After the application of Frangi-based filtering the data showed sufficient contrast to automatically identify the Purkinje cells and to quantify their density to 177 cells per mm3 within the volume of interest. Moreover, brain layers were segmented in a region of interest based on edge detection. Subsequently performed histological analysis validated the presence of the cells, which required a mapping from the two- dimensional histological slices to the three-dimensional tomogram. The methodology can also be applied to further tissue types and shows potential for the computational tissue analysis in health and disease.