Automatic Segmentation of Pathological Glomerular Basement Membrane in Transmission Electron Microscopy Images with Random Forest Stacks.

Pathological classification through transmission electron microscopy (TEM) is essential for the diagnosis of certain nephropathy, and the changes of thickness in glomerular basement membrane (GBM) and presence of immune complex deposits in GBM are often used as diagnostic criteria. The automatic segmentation of the GBM on TEM images by computerized technology can provide clinicians with clear information about glomerular ultrastructural lesions. The GBM region on the TEM image is not only complicated and changeable in shape but also has a low contrast and wide distribution of grayscale. Consequently, extracting image features and obtaining excellent segmentation results are difficult. To address this problem, we introduce a random forest- (RF-) based machine learning method, namely, RF stacks (RFS), to realize automatic segmentation. Specifically, this work proposes a two-level integrated RFS that is more complicated than a one-level integrated RF to improve accuracy and generalization performance. The integrated strategies include training integration and testing integration. Training integration can derive a full-view RFS1 by simultaneously sampling several images of different grayscale ranges in the train phase. Testing integration can derive a zoom-view RFS2 by separately sampling the images of different grayscale ranges and integrating the results in the test phase. Experimental results illustrate that the proposed RFS can be used to automatically segment different morphologies and gray-level basement membranes. Future study on GBM thickness measurement and deposit identification will be based on this work.

A Rare Lysozyme Crystal Form Solved Using Highly Redundant Multiple Electron Diffraction Datasets from Micron-Sized Crystals.

Recent developments of novel electron diffraction techniques have shown to be powerful for determination of atomic resolution structures from micron- and nano-sized crystals, too small to be studied by single-crystal X-ray diffraction. In this work, the structure of a rare lysozyme polymorph is solved and refined using continuous rotation MicroED data and standard X-ray crystallographic software. Data collection was performed on a standard 200 kV transmission electron microscope (TEM) using a highly sensitive detector with a short readout time. The data collection is fast (∼3 min per crystal), allowing multiple datasets to be rapidly collected from a large number of crystals. We show that merging data from 33 crystals significantly improves not only the data completeness, overall I/σ and the data redundancy, but also the quality of the final atomic model. This is extremely useful for electron beam-sensitive crystals of low symmetry or with a preferred orientation on the TEM grid.

Annular dark field transmission electron microscopy for protein structure determination.

Recently annular dark field (ADF) transmission electron microscopy (TEM) has been advocated as a means of recording images of biological specimens with better signal to noise ratio (SNR) than regular bright field images. I investigate whether and how such images could be used to determine the three-dimensional structure of proteins given that an ADF aperture with a suitable pass-band can be manufactured and used in practice. I develop an approximate theory of ADF-TEM image formation for weak amplitude and phase objects and test this theory using computer simulations. I also test whether these simulated images can be used to calculate a three-dimensional model of the protein using standard software and discuss problems and possible ways to overcome these.

Localization of fluorescently tagged protein to plasmodesmata by correlative light and electron microscopy.

Plasmodesmata (PD) are intercellular communication channels that form long, membrane-lined cylinders across cellular junctions. A fluorescent-tagging approach is most commonly used for an initial assessment to address whether a protein of interest may localize or associate with PD domain. However, owing to the dimension of PD being at nanoscale, PD-associated fluorescent signals are detected only as small spots scattered at the cell periphery, hence requiring additional confirmatory evidence. Immunogold labeling provides such information, but suitable antibodies are not always available and morphological preservation is often compromised with this approach. Here we describe an alternative approach using a correlative light and electron microscopy (CLEM) technique, which combines fluorescent imaging and transmission electron microscopy. By employing this method, a clear correlation between fluorescent speckles and the presence of individual or clusters of PD is achieved.

Macromolecular 3D SEM reconstruction strategies: signal to noise ratio and resolution.

Three-dimensional scanning electron microscopy generates quantitative volumetric structural data from SEM images of macromolecules. This technique provides a quick and easy way to define the quaternary structure and handedness of protein complexes. Here, we apply a variety of preparation and imaging methods to filamentous actin in order to explore the relationship between resolution, signal-to-noise ratio, structural preservation and dataset size. This information can be used to define successful imaging strategies for different applications.

Choice of operating voltage for a transmission electron microscope.

An accelerating voltage of 100-300kV remains a good choice for the majority of TEM or STEM specimens, avoiding the expense of high-voltage microscopy but providing the possibility of atomic resolution even in the absence of lens-aberration correction. For specimens thicker than a few tens of nm, the image intensity and scattering contrast are likely to be higher than at lower voltage, as is the visibility of ionization edges below 1000eV (as required for EELS elemental analysis). In thick (>100nm) specimens, higher voltage ensures less beam broadening and better spatial resolution for STEM imaging and EDX spectroscopy. Low-voltage (e.g. 30kV) TEM or STEM is attractive for a very thin (e.g. 10nm) specimen, as it provides higher scattering contrast and fewer problems for valence-excitation EELS. Specimens that are immune to radiolysis suffer knock-on damage at high current densities, and this form of radiation damage can be reduced or avoided by choosing a low accelerating voltage. Low-voltage STEM with an aberration-corrected objective lens (together with a high-angle dark-field detector and/or EELS) offers atomic resolution and elemental identification from very thin specimens. Conventional TEM can provide atomic resolution in low-voltage phase-contrast images but requires correction of chromatic aberration and preferably an electron-beam monochromator. Many non-conducting (e.g. organic) specimens damage easily by radiolysis and radiation damage then determines the TEM image resolution. For bright-field scattering contrast, low kV can provide slightly better dose-limited resolution if the specimen is very thin (a few nm) but considerably better resolution is possible from a thicker specimen, for which higher kV is required. Use of a phase plate in a conventional TEM offers the most dose-efficient way of achieving atomic resolution from beam-sensitive specimens.

Atomic structure from large-area, low-dose exposures of materials: a new route to circumvent radiation damage.

Beam-induced structural modifications are a major nuisance in the study of materials by high-resolution electron microscopy. Here, we introduce a new approach to circumvent the radiation damage problem by a statistical treatment of large, noisy, low-dose data sets of non-periodic configurations (e.g. defects) in the material. We distribute the dose over a mixture of different defect structures at random positions and with random orientations, and recover representative model images via a maximum likelihood search. We demonstrate reconstructions from simulated images at such low doses that the location of individual entities is not possible. The approach may open a route to study currently inaccessible beam-sensitive configurations.

Changes of synaptic ultrastructure in the guinea pig interpositus nuclei associate with response magnitude and timing after trace eyeblink conditioning.

Learning-induced changes of synaptic ultrastructure have long been proposed as a mechanism that may contribute to support memory formation. Although recent studies have demonstrated that the interpositus nuclei (IN) play critical role in acquisition and retention of trace conditioned eyeblink responses (CRs), there is now limited evidence associating trace eyeblink conditioning with changes of synaptic ultrastructure in the IN. Here, we investigated this issue using a transmission electron microscope. Adult guinea pigs were randomly allocated to either a trace-paired, delay-paired, unpaired or exposure-only condition. The IN tissue was taken for morphological analysis 1h after the completion of the tenth training session. Serial section analysis of synaptic ultrastructure revealed that trace eyeblink conditioning induced increases in the thickness of excitatory PSD. Classification of the synapses into shape subtypes indicated that the increased thickness of excitatory PSD was mainly attributable to increase in the concave- and convex-shaped synapses. On the contrary, trace eyeblink conditioning resulted in decreases in the thickness of inhibitory PSD. Specifically, these significant changes of PSD thickness were limited to occur in the animals with good behavioral performance. Further analysis of correlations between the trace CR performance and synaptic ultrastructural modifications showed that the thickness of excitatory PSD within the IN correlated with the peak amplitude of trace CRs, whereas the thickness of inhibitory PSD correlated with the onset latency. The present findings suggest that trace eyeblink conditioning induces structural plasticity in the IN, which may play a crucial role in acquiring and executing adaptive eyeblink movements.

Automated grain mapping using wide angle convergent beam electron diffraction in transmission electron microscope for nanomaterials.

The grain size statistics, commonly derived from the grain map of a material sample, are important microstructure characteristics that greatly influence its properties. The grain map for nanomaterials is usually obtained manually by visual inspection of the transmission electron microscope (TEM) micrographs because automated methods do not perform satisfactorily. While the visual inspection method provides reliable results, it is a labor intensive process and is often prone to human errors. In this article, an automated grain mapping method is developed using TEM diffraction patterns. The presented method uses wide angle convergent beam diffraction in the TEM. The automated technique was applied on a platinum thin film sample to obtain the grain map and subsequently derive grain size statistics from it. The grain size statistics obtained with the automated method were found in good agreement with the visual inspection method.

High precision measurements of atom column positions using model-based exit wave reconstruction.

In this paper, it has been investigated how to measure atom column positions as accurately and precisely as possible using a focal series of images. In theory, it is expected that the precision would considerably improve using a maximum likelihood estimator based on the full series of focal images. As such, the theoretical lower bound on the variances of the unknown atom column positions can be attained. However, this approach is numerically demanding. Therefore, maximum likelihood estimation has been compared with the results obtained by fitting a model to a reconstructed exit wave rather than to the full series of focal images. Hence, a real space model-based exit wave reconstruction technique based on the channelling theory is introduced. Simulations show that the reconstructed complex exit wave contains the same amount of information concerning the atom column positions as the full series of focal images. Only for thin samples, which act as weak phase objects, this information can be retrieved from the phase of the reconstructed complex exit wave.

Interactive histology of large-scale biomedical image stacks.

Histology is the study of the structure of biological tissue using microscopy techniques. As digital imaging technology advances, high resolution microscopy of large tissue volumes is becoming feasible; however, new interactive tools are needed to explore and analyze the enormous datasets. In this paper we present a visualization framework that specifically targets interactive examination of arbitrarily large image stacks. Our framework is built upon two core techniques: display-aware processing and GPU-accelerated texture compression. With display-aware processing, only the currently visible image tiles are fetched and aligned on-the-fly, reducing memory bandwidth and minimizing the need for time-consuming global pre-processing. Our novel texture compression scheme for GPUs is tailored for quick browsing of image stacks. We evaluate the usability of our viewer for two histology applications: digital pathology and visualization of neural structure at nanoscale-resolution in serial electron micrographs.

The utility of electron microscopy in cytopathology.

BACKGROUND: The use of ultrastructural analysis in the diagnostic work-up of histologic specimens has been well studied but less is known about the utility of electron microscopy (EM) in cytopathology. DESIGN: 149,006 non-gynecologic cytology cases at the Massachusetts General Hospital between the years 1993 and 2006 were searched to identify those in which material had been submitted for EM. Cytologic and EM diagnoses were correlated with available histologic diagnoses. The results were put into one of three categories: confirmatory, diagnostic, or insufficient material for diagnosis (IMFD). RESULTS: Material for EM was obtained from 178 cytology cases that included 131 fine-needle aspirates (FNA) and 47 exfoliative specimens. EM provided additional diagnostic information beyond that offered by cytologic examination alone in 32% of cases, and in 48% of cases EM confirmed the cytologic findings. Insufficient material and discrepant results were noted for EM evaluation in 19% of cases and in 1% cases respectively. EM was most useful when applied to FNAs for subclassifying tumors as epithelial or mesenchymal (45.6%), for the diagnosis of non-neoplastic processes (15.7%) such as alveolar proteinosis and amyloidosis, and for the identification of microorganisms (12.2%). In our study, although EM was infrequently applied to exfoliative specimens to distinguish mesothelioma from adenocarcinoma, it proved to be very useful in this setting. CONCLUSION: When adequate material is obtained, EM can contribute significantly to the evaluation of both FNA and exfoliative cytology cases, including the diagnosis and subclassification of epithelial and mesenchymal tumors, non-neoplastic processes, and the identification of microorganisms.

Characterization of a direct detection device imaging camera for transmission electron microscopy.

The complete characterization of a novel direct detection device (DDD) camera for transmission electron microscopy is reported, for the first time at primary electron energies of 120 and 200 keV. Unlike a standard charge coupled device (CCD) camera, this device does not require a scintillator. The DDD transfers signal up to 65 lines/mm providing the basis for a high-performance platform for a new generation of wide field-of-view high-resolution cameras. An image of a thin section of virus particles is presented to illustrate the substantially improved performance of this sensor over current indirectly coupled CCD cameras.

Role of immunofluorescence and electron microscopy in the evaluation of renal biopsies in nephrotic syndrome in a developing country.

To determine the role of immunofluorescence (IF) and electron microscopy (EM) in the evaluation of renal biopsies in a developing country, the authors carried out a study in 200 patients with nephrotic syndrome. Renal biopsies were studied by light microscopy, IF, and EM. IF study was useful in all, being essential in 23.5% and helpful in remaining cases. EM was useful in 94.5% cases, being essential in 43% and helpful in 51.5% cases. The results demonstrate that IF and EM are essential in the evaluation of renal biopsies in nephrotic syndrome and these should be employed in the pathologic evaluation of renal biopsies.

A statistically harmonized alignment-classification in image space enables accurate and robust alignment of noisy images in single particle analysis.

In determining the three-dimensional (3D) structure of macromolecular assemblies in single particle analysis, a large representative dataset of two-dimensional (2D) average images from huge number of raw images is a key for high resolution. Because alignments prior to averaging are computationally intensive, currently available multireference alignment (MRA) software does not survey every possible alignment. This leads to misaligned images, creating blurred averages and reducing the quality of the final 3D reconstruction. We present a new method, in which multireference alignment is harmonized with classification (multireference multiple alignment: MRMA). This method enables a statistical comparison of multiple alignment peaks, reflecting the similarities between each raw image and a set of reference images. Among the selected alignment candidates for each raw image, misaligned images are statistically excluded, based on the principle that aligned raw images of similar projections have a dense distribution around the correctly aligned coordinates in image space. This newly developed method was examined for accuracy and speed using model image sets with various signal-to-noise ratios, and with electron microscope images of the Transient Receptor Potential C3 and the sodium channel. In every data set, the newly developed method outperformed conventional methods in robustness against noise and in speed, creating 2D average images of higher quality. This statistically harmonized alignment-classification combination should greatly improve the quality of single particle analysis.

A refined circular template matching method for classification of human cytomegalovirus capsids in TEM images.

An automatic image analysis method for describing, segmenting, and classifying human cytomegalovirus capsids in transmission electron micrograph (TEM) images of host cell nuclei has been developed. Three stages of the capsid assembly process in the host cell nucleus have been investigated. Each class is described by a radial density profile, which is the average grey-level at each radial distance from the center. A template, constructed from the profile, is used to find possible capsid locations by correlation based matching. The matching results are further refined by size and distortion analysis of each possible capsid, resulting in a final segmentation and classification.