Journal of Microscopy

The recovery of critical raw materials from water electrolysers, which are used to produce green hydrogen, is essential to keep the raw materials with limited availability in the material cycle and to facilitate the expansion of production of this technology, which is supposed to be essential for the decarbonisation of our industrial society. Proton exchange membrane water electrolysers (PEMWE) use precious metals such as Ir and Pt as catalysts, which require a high recycling rate due to their natural scarcity. In order to investigate at an early‐stage mechanical recycling technologies, such as shredding for liberation and milling for decoating of these complex materials, it becomes necessary to develop small‐scale experimental methods. This is due to the low availability of End‐of‐Life samples and the high price of pristine electrolyser components. Especially decoating has shown huge potential for a highly selective separation of defined material layers; nevertheless, until now, there is no method to determine the success of decoating of the flexible polymer membrane, which is coated on both sides with particle‐based electrodes. One possible concept is presented here, using scanning electron microscope images and micro‐X‐ray fluorescence elemental maps. Image processing and segmentation is performed using the WEKA software and a simple thresholding method. This allows the efficiency of the decoating process to be determined with an accuracy of ±0.5 percentage points for decoated PEMWE cell samples. The high accuracy of the presented method framework provides the necessary tool for any further quantitative development of improved mechanical stressing for decoating.

Highly oriented pyrolytic graphite (HOPG) is one of the most used host materials for obtaining and investigating graphite intercalated compounds, because of the high degree structural order of this polycrystal. Experiments on electrochemically intercalated HOPG in sulphuric acid have a model character, as the results obtained can be usefully generalised, not only with respect to other graphite compounds but also for the intercalation of other layered host lattices. In addition, the HOPG/H2SO4 system has an attractive potential for the possibility of electrochemically producing graphite oxide, ideally, by reversible oxidation/reduction cycles, which is of interest for energy storage and graphene production on an industrial scale. However, the oxidation/reduction cycles in such electrochemical intercalation process are not reversible and topotactic, so that the HOPG structure is considerably altered. This alteration may affect, for instance, the quality of the electrochemically produced graphene. In particular, the impact the electrochemical intercalation has on the conductivity of basal planes of HOPG, and so on graphene sheets, is still debated. In this work, we investigated both the macroscopic and microscopic electron transfer (ET) kinetics of the HOPG surface, before and after the intercalation of 1 M H2SO4 to obtain graphite intercalated compound, by using cyclic voltammetry (CV) and scanning electrochemical microscopy (SECM), respectively. The heterogeneous kinetic constant (k⁰) of the HOPG was evaluated quantitatively by using the redox systems [Fe(CN)6]3–/4– and [Ru(NH3)6]3+/2+. The morphology of the samples was also investigated by atomic force microscopy (AFM), which revealed a widespread formation of blisters and precipitates during the HOPG intercalation process. The CV and SECM results indicate that, upon intercalation, the electrochemical behaviour of the HOPG changes sensibly and the ET decreases sensibly. However, this effect depends on the redox mediators employed and it results more dramatic for the [Fe(CN)6]3–/4– system, for which a decrease of k⁰ by orders of magnitude was obtained. The decrease of ET can be correlated to the blisters and precipitates, which occur during the HOPG intercalation, as observed by AFM.

Microscopy and image analysis play a vital role in parasitology research; they are critical for identifying parasitic organisms and elucidating their complex life cycles. Despite major advancements in imaging and analysis, several challenges remain. These include the integration of interdisciplinary data; information derived from various model organisms; and data acquired from clinical research. In our view, artificial intelligence—with the latest advances in machine and deep learning—holds enormous potential to address many of these challenges. This review addresses how artificial intelligence, machine learning and deep learning have been used in the field of parasitology—mainly focused on Apicomplexan, Diplomonad, and Kinetoplastid groups. We explore how gaps in our understanding could be filled by AI in future parasitology research and diagnosis in the field. Moreover, it addresses challenges and limitations currently faced in implementing and expanding the use of artificial intelligence across biomedical fields. The necessary increased collaboration between biologists and computational scientists will facilitate understanding, development, and implementation of the latest advances for both scientific discovery and clinical impact. Current and future AI tools hold the potential to revolutionise parasitology and expand One Health principles.

To enhance the indexing rate of conventional electron backscatter diffraction (EBSD), this study employed EBSD to collect and analyse the mapping data of cubic phase materials. Kikuchi bands were identified using Hough transform, and the pattern centre was optimised through a genetic algorithm. Four objective functions were designed to investigate the influence of varying population sizes on the convergence of the algorithm. The results revealed that the calculation stabilised when the population size reached 400, with the HMAE (H‐mean angular error) objective function exhibiting superior performance in screening by integrating the number of matched Kikuchi bands and mean angular error (MAE). Furthermore, to address indexing errors resulting from overlapping Kikuchi patterns at grain boundaries, an indexing optimisation method based on pattern similarity matching was proposed, significantly improving the indexing rate of EBSD mapping data. Finally, neighbourhood search strategy was implemented to further refine the indexing process, ensuring high indexing accuracy while substantially reducing computational time. This study offers novel methodologies and insights for improving the efficiency and precision of EBSD mapping data acquisition and analysis.

Influenza B virus and SARS‐CoV‐2 virus are the two most representative respiratory infectious diseases. These two viruses not only show similarities in clinical symptoms but also have numerous similarities in microstructure, which is difficult to distinguish and poses great challenges for diagnosis. In this work, the three‐dimensional structures and surface features of influenza B virus and SARS‐CoV‐2 virus were investigated using atomic force microscopy. The results indicated that there were substantial differences in surface morphology and structure between the two viruses. Specifically, the average diameter of SARS‐CoV‐2 virus particles was around 222.8 nm while that of influenza B virus particles is smaller at about 191.2 nm. The height of SARS‐CoV‐2 particles was also larger, averaging about 30–60 nm, while that of influenza B virus particles averaged around 10–30 nm. Additionally, the crown‐like structure on the surface of the SARS‐CoV‐2 virus was sparser and more prominent than that of the influenza virus. These findings offer significant insights into the distinction between the two viruses, aiding in the accurate characterisation of SARS‐CoV‐2 and influenza viruses and facilitating timely and effective treatment strategies.

Energy storage technologies that are efficient are in constant demand. Supercapacitors have attracted much interest among these gadgets because of their superior cycle stability and high‐power density. This work used a simple and cost‐effective sonication‐assisted electrodeposition approach to develop tin oxide nanoparticles on functionalised carbon foam substrate with different concentration ratios (1 mM, 3 mM, and 5 mM). FTIR, XRD, and SEM validated the chemical, structural, and morphological characteristics of all nanostructured electrodes. The tetragonal structure with spherical shape was the result of the fine crystallisation of the tin oxide nanoparticles. The electrochemical characteristics are evaluated by CV, EIS, and GCD testing. Among all electrodes, Sn 1 @CF has a larger electrochemically active surface area, low internal resistance, and high specific capacitance. These findings underscore that the binder‐free Sn 1 @CF electrode is a promising candidate for high‐efficiency supercapacitor applications.

X‐rays, secondary electrons, and other emitted electrons need to be extracted at a large solid angle to enhance electron collection efficiency in transmission electron microscopy. The finite element method is employed to investigate the effects of different polepiece angles on the spherical aberration coefficient of polepiece. The research findings reveal that the azimuthal angle β of the upper polepiece has a substantial effect on the spherical aberration coefficient. When β = 30°, the minimum spherical aberration coefficient is achieved. When β ≥ 50°, the spherical aberration coefficient increases significantly, which adversely affects imaging. The aperture size of the upper polepiece has a relatively minor effect on the spherical aberration. The design of the large‐angle polepiece offers novel design concepts for future emission X‐ray/electron collection devices, while also offering a new reference for the design of objective lenses in transmission electron microscopy.

A successful scanning tunnelling microscopy (STM) experiment relies on both delicate sample preparation and measurement, and careful image filtering and analysis to provide clear and solid results. Processing and analysis of STM images may result in a tricky task, due to the complexity and specificity of the probed systems. In this paper, we introduce our recently developed, simple Python-based methods for filtering and analysing STM images, with the aim of providing a semi-quantitative treatment of the input data. Case studies will be presented using images obtained through electrochemical STM. Additionally, we propose a straightforward yet effective universal drift-correction tool for SPM image sequences.

Laser scanning fluorescence microscopy (LSFM) is a widely used imaging method, but image quality is often degraded by noise. Averaging techniques can enhance the signal‐to‐noise ratio (SNR), but while this can improve image quality, excessive frame accumulation can introduce photobleaching and may lead to unnecessarily long acquisition times. A classical software method called PerfectlyAverage is presented to determine the optimal number of frames for averaging in LSFM using SNR, photobleaching, and power spectral density (PSD) measurements. By assessing temporal intensity variations across frames in a time series, PerfectlyAverage identifies the point where additional averaging ceases to provide significant noise reduction. Experiments with fluorescently stained tissue paper and fibroblast cells validated the approach, demonstrating that up to a fourfold reduction in averaging time may be possible. PerfectlyAverage is open source, compatible with any LSFM data, and it is aimed at improving imaging workflows while reducing the reliance on subjective criteria for choosing the number of averages.

Propagation‐based imaging (one method of X‐ray phase contrast imaging) with microcomputed tomography (PBI‐µCT) offers the potential to visualise low‐density materials, such as soft tissues and hydrogel constructs, which are difficult to be identified by conventional absorption‐based contrast µCT. Conventional µCT reconstruction produces edge‐enhanced contrast (EEC) images which preserve sharp boundaries but are susceptible to noise and do not provide consistent grey value representation for the same material. Meanwhile, phase retrieval (PR) algorithms can convert edge enhanced contrast to area contrast to improve signal‐to‐noise ratio (SNR), and contrast‐to‐noise ratio (CNR) but usually results to over‐smoothing, thus creating inaccuracies in quantitative analysis. To alleviate these problems, this study developed a deep learning‐based method called edge view enhanced phase retrieval (EVEPR), by strategically integrating the complementary spatial features of denoised EEC and PR images, and further applied this method to segment the hydrogel constructs in vivo and ex vivo. EVEPR used paired denoised EEC and PR images to train a deep convolutional neural network (CNN) on a dataset‐to‐dataset basis. The CNN had been trained on important high‐frequency details, for example, edges and boundaries from the EEC image and area contrast from PR images. The CNN predicted result showed enhanced area contrast beyond conventional PR algorithms while improving SNR and CNR. The enhanced CNR especially allowed for the image to be segmented with greater efficiency. EVEPR was applied to in vitro and ex vivo PBI‐µCT images of low‐density hydrogel constructs. The enhanced visibility and consistency of hydrogel constructs was essential for segmenting such material which usually exhibit extremely poor contrast. The EVEPR images allowed for more accurate segmentation with reduced manual adjustments. The efficiency in segmentation allowed for the generation of a sizeable database of segmented hydrogel scaffolds which were used in conventional data‐driven segmentation applications. EVEPR was demonstrated to be a robust post‐image processing method capable of significantly enhancing image quality by training a CNN on paired denoised EEC and PR images. This method not only addressed the common issues of over‐smoothing and noise susceptibility in conventional PBI‐µCT image processing but also allowed for efficient and accurate in vitro and ex vivo image processing applications of low‐density materials.

Cryogenic scanning electron microscopy (cryo‐SEM) is a powerful imaging technique used in cellular biology, providing high‐resolution micrographs that show the complexity and dynamics of biological systems. The use of high‐pressure freezing (HPF) for specimen fixation preserves cellular structures in their native, hydrated state, avoiding the artefacts introduced by conventional chemical fixation, while modern microscopes provide high‐resolution imaging at low electron acceleration voltage, giving fine structural details. That makes cryo‐SEM a unique tool for understanding cellular complexity. However, operating the SEM at cryogenic conditions requires careful optimisation of working parameters to avoid artefacts. In our work, we explore the potential of cryo‐SEM for haematology and general cell studies. We discuss the impact of a combination of different signals and work distance on specimen appearance and present examples of studies on healthy human blood cells under physiological conditions. Our findings illustrate the breadth of information that can be obtained from these data, highlighting the technique's capacity to enhance our understanding of cellular biology.

For sodium (Na)‐ion batteries (SIBs), the next generation of sustainable batteries, hard carbon (HC) composite electrodes are the most used anodes. Here, we demonstrate the potential of modulated electrochemical force microscopy (mec‐AFM) to investigate electrochemical strain due to ion insertion at the electrolyte/electrode interface. HC composite anodes have a complex, multiphase structure, which include the HC particles, conductive carbon nanoparticles (carbon black) and the binder. To address the effect of the composite material on the sodium‐ion transport, we employ mec‐AFM. A HC composite anode was embedded in an epoxy‐polymer matrix and was polished to expose a micro‐sized area that enabled high‐frequency modulation of the ion transport. We analyse the influence of the modulation on interfacial forces and its role in generating electrochemical strain in the composite anode. Multichannel mec‐AFM imaging at varying electrode potentials revealed that the observed electrochemical strain predominantly occurred in the softer binder matrix rather than in the HC microparticles. Our findings underscore the significance of ionic transport pathways through the binder matrix and establish mec‐AFM as a novel AFM‐derived technique for visualising ion dynamics at battery interfaces.

Supramolecular self‐assembly on surfaces enables tailored interfaces with applications in nanotechnology. While factors like temperature and solute concentration influence self‐assembled molecular networks (SAMNs), the role of spatial confinement remains less explored. Here, we investigate the self‐assembly of an alkylated quinonoid zwitterion (QZ‐C16) at the liquid–solid interface using scanning tunnelling microscopy (STM), both in in situ as well as ex situ nanocorrals. Engineered nanocorrals not only provide a confined environment for molecular assembly, but also serve as platforms for probing the impact of geometric constraints on self‐assembly behaviour. Understanding the intricate dynamics of self‐assembly at the nanoscale, particularly the mechanisms by which confinement influences structural organisation, can inform strategies for achieving desired molecular architectures.

Single molecule localisation microscopy (SMLM) can provide two‐dimensional super‐resolved image data from conventional fluorescence microscopes, while three dimensional (3D) SMLM usually involves a modification of the microscope, for example, to engineer a predictable axial variation in the point spread function. Here we demonstrate a 3D SMLM approach (we call ‘easyZloc') utilising a lightweight Convolutional Neural Network that is generally applicable, including with ‘standard’ (unmodified) fluorescence microscopes, and which we consider may be practically useful in a high throughput SMLM workflow. We demonstrate the reconstruction of nuclear pore complexes with comparable performance to previously reported methods but with a significant reduction in computational power and execution time. 3D reconstructions of the nuclear envelope and an actin sample over a larger axial range are also shown.

Atomic force microscopy (AFM) plays a pivotal role in cell biology research. It enables scientists to observe the morphology of cell surfaces at the nanoscale, providing essential data for understanding cellular functions, including cell‐cell interactions and responses to the microenvironment. Nevertheless, AFM‐captured cell images frequently suffer from artefacts, which significantly hinder detailed analyses of cell structures. In this study, we developed a cross‐module resolution enhancement method for post‐processing AFM cell images. The method leverages the AFM topological deep learning neural network. We propose an enhanced spatial fusion structure and an optimised back‐projection mechanism within an adversarial‐based super‐resolution network to detect weak signals and complex textures unique to AFM cell images. Furthermore, we designed a crossover‐based frequency division module, capitalising on the distinct frequency characteristics of AFM images. This module effectively separates and enhances features pertinent to cell structure. In this paper, experiments were conducted using AFM images of various cells, and the results demonstrated the model's superiority. It substantially enhances image quality compared to existing methods. Specifically, the peak signal‐to‐noise ratio (PSNR) of the reconstructed image increased by 1.65 decibels, from 28.121 to 29.771, the structural similarity (SSIM) increased by 0.041, from 0.746 to 0.787, the Learned Perceptual Image Patch Similarity (LPIPS) decreased by 0.205, from 0.437 to 0.232, the Fréchet Inception Distance (FID) decreased by 6.996, from 55.442 to 48.446 and the Natural Image Quality Evaluator (NIQE) decreased by 0.847, from 4.296 to 3.449. Lay abstract : This study proposes a deep learning‐based cross‐module method for super‐resolving AFM cell images, integrating frequency division and adaptive fusion modules. It boosts PSNR by 1.65 dB and SSIM by 0.041, accurately recovering cellular microstructures, thus significantly aiding cell biology research and biomedicine applications.

LAY DESCRIPTION Atomic force microscope (AFM) is an incredibly powerful tool used by scientists to explore the tiny world of atoms and molecules. It works like an ultra‐precise magnifying glasses, allowing researchers to ‘see’ the surfaces of materials at the nanoscale—far smaller than anything visible to the naked eye. However, even the most advanced tool faces challenges. One of the common problems with AFM is that its mechanical parts, especially the piezoelectric actuators that move the scanning tip, do not always behave perfectly. Over time, these parts may drift, lag, or deform slightly, especially during long scans. As a result, the final image may appear stretched, warped, or blurry—similar to what happens when trying to take a photo while the camera is moving. To tackle this issue, a new scanning method has been introduced. Instead of scanning line by line like a lawnmower, the microscope now scans in a spiral pattern—starting from the centre and gradually moving outward like the shell of a snail. This new method also breaks the image into smaller, overlapping sections. These overlaps allow the system to continuously compare parts of the image and correct drift in real time. It's similar to how a GPS navigation app recalculates the route when the vehicle drifts off course, ensuring the scan remains aligned and accurate. This spiral scanning method was tested and compared with traditional scanning techniques. In tests using 600‐pixel‐wide images, the amount of distortion was reduced by about 95%. This improvement could help scientists capture clearer, more accurate images during long experiments, which is especially useful in fields like materials science or biology where precise measurements are critical. Essentially, the spiral method helps the microscope ‘stay on track’ better, producing sharper pictures of the nanoworld.

Despite the development of 3D imaging technology, the reconstruction of three‐dimensional (3D) microstructure from a single two‐dimensional (2D) image is still a prominent problem. In this paper, we propose a hierarchical reconstruction method based on simulated annealing, which is named hierarchical simulated annealing method (HSA), with the multiscale entropy statistics as the morphological information descriptor to reconstruct its corresponding three‐dimensional (3D) microstructure from a single two‐dimensional (2D) image. Both hierarchical simulated annealing (HSA) method and simulated annealing (SA) method are used to perform on the 2D and 3D microstructure reconstruction from a single 2D image, where the two‐point cluster function and the standard two‐point correlation function are used as the measurement metrics for the reconstructed 2D and 3D structures. From the 2D reconstructions, it can be seen that all the reconstructions of HSA method and SA method not only captures the similar morphological information with the original images, but also have a good agreement with the target microstructures in two‐point cluster function. For the reconstructed 3D microstructures, the comparison of two‐point correlation function shows that both HSA method and SA method can effectively reconstruct its 3D microstructure and the comparison of the reconstruction time between HSA method and SA method shows that the reconstruction speed of HSA method is an order of magnitude faster than that of SA method.

We previously developed the CellPhe toolkit, an open‐source R package for automated cell phenotyping from ptychography time‐lapse videos. To align with the growing adoption of python‐based image analysis tools and to enhance interoperability with widely used software for cell segmentation and tracking, we developed a python implementation of CellPhe, named CellPhePy. CellPhePy preserves all of the core functionality of the original toolkit, including single‐cell phenotypic feature extraction, time‐series analysis, feature selection and cell type classification. In addition, CellPhePy introduces significant enhancements, such as an improved method for identifying features that differentiate cell populations and extended support for multiclass classification, broadening its analytical capabilities. Notably, the CellPhePy package supports CellPose segmentation and TrackMate tracking, meaning that a set of microscopy images are the only required input with segmentation, tracking and feature extraction fully automated for downstream analysis, without reliance on external applications. The workflow's increased flexibility and modularity make it adaptable to different imaging modalities and fully customisable to address specific research questions. CellPhePy can be installed via PyPi or GitHub, and we also provide a CellPhePy GUI to aid user accessibility.

Determining the concentration of elements in subcellular structures poses a significant challenge. By locating an elemental species at high spatial resolution and with subcellular context, and subsequently quantifying it on an absolute scale, new information about cellular function can be revealed. Such measurements have not as yet been realised with existing techniques due to limitations on spatial resolution and inherent difficulties in detecting elements present in low concentrations. In this paper, we use scanning transmission electron microscopy (STEM) to establish a methodology for localising and quantifying high‐Z elements in a biological setting by measuring elastic electron scattering. We demonstrate platinum (Pt) deposition within neuronal cell bodies following in vivo administration of the Pt‐based chemotherapeutic oxaliplatin to validate this novel methodology. For the first time, individual Pt atoms and nanoscale Pt clusters are shown within subcellular structures. Quantitative measurements of elastic electron scattering are used to determine absolute numbers of Pt atoms in each cluster. Cluster density is calculated on an atoms‐per‐cubic‐nanometre scale, and used to show clusters form with densities below that of metallic Pt. By considering STEM partial scattering cross‐sections, we determine that this new approach to subcellular elemental detection may be applicable to elements as light as sodium. LAY DESCRIPTION: Heterogeneous elemental distributions drive fundamental biological processes within cells. While carbon, hydrogen, oxygen and nitrogen comprise by far the majority of living matter, concentrations and locations of more than a dozen other species must also be tightly controlled to ensure normal cell function. Oxaliplatin is a first‐line and adjuvant treatment for colorectal cancer. However, pain in the body's extremities (fingers and toes) significantly impairs clinical usage as this serious and persistent side effect impacts on both patient cancer care and quality of life. Annular dark‐field (ADF) imaging in the scanning transmission electron microscope (STEM) provides an image with strong atom‐number contrast and is sufficient to distinguish between different cell types and different organelles within the cells of the DRG. We also show that Pt may be imaged at the single atom level and be localised at very high resolution while still preserving a degree of ultrastructural context. The intrinsic image contrast generated is sufficient to identify these features without the need for heavy metal stains and other extensive processing steps which risk disturbing native platinum distributions within the tissue. We subsequently demonstrate that by considering the total elastic scattering intensity generated by nanometre‐sized Pt aggregations within the cell, the ADF STEM may be used to make a measurement of local concentration of Pt in units of atoms per cubic nanometre. We further estimate the minimum atomic number required to visualise single atoms in this setting, concluding that in similar samples it may be possible to detect species as light as sodium with atomic sensitivity.

The morphological characterisation is crucial for analysing cell states, especially for red blood cells (RBCs), which are used in transfusions. This study compared the applicability of atomic force microscopy (AFM) and confocal optical profilometry in the accurate characterisation of the RBC morphological parameters. The imaging of RBCs thawed after cryopreservation with immediate and delayed washing steps (deglycerolisation) was performed, and the morphological data obtained with AFM and optical profilometry were compared with the clinical laboratory studies. Both techniques provided close data on the morphological parameters, but optical profilometry allowed a faster and more convenient data acquisition. However, the membrane roughness analysis on discocytes and the submembrane cytoskeleton analysis on RBC ghosts was only possible with AFM due to its higher spatial resolution. Both techniques confirmed that delayed washing did not have negative effects on cells compared to immediate washing. Additional 3‐day storage of both types of RBCs resulted in increased haemolysis. A decrease in the fraction of area occupied by pores in the submembrane cytoskeleton with the storage time was observed, possibly associated with the cytoskeleton deterioration. The studied conditions model the transportation of thawed RBCs in a cryoprotectant solution to medical facilities that have technical conditions to wash thawed RBCs and confirm its feasibility.

In the domain of battery research, the processing of high‐resolution microscopy images is a challenging task, as it involves dealing with complex images and requires a prior understanding of the components involved. The utilisation of deep learning methodologies for image analysis has attracted considerable interest in recent years, with multiple investigations employing such techniques for image segmentation and analysis within the realm of battery research. However, the automated analysis of high‐resolution microscopy images for detecting phases and components in composite materials is still an underexplored area. This work proposes a novel workflow for FFT‐based segmentation, periodic component detection and phase segmentation from raw high‐resolution Transmission Electron Microscopy (TEM) images using a trained U‐Net segmentation model. The developed model can expedite the detection of components and their phase segmentation, diminishing the temporal and cognitive demands associated with scrutinising an extensive array of TEM images, thereby mitigating the potential for human errors. This approach presents a novel and efficient image analysis approach with broad applicability beyond the battery field and holds potential for application in other related domains characterised by phase and composition distribution, such as alloy production.

Electron ptychography is a powerful and versatile tool for high‐resolution and dose‐efficient imaging. Iterative reconstruction algorithms are powerful but also computationally expensive due to their relative complexity and the many hyperparameters that must be optimised. Gradient descent‐based iterative ptychography is a popular method, but it may converge slowly when reconstructing low spatial frequencies. In this work, we present a method for accelerating a gradient descent‐based iterative reconstruction algorithm by training a neural network (NN) that is applied in the reconstruction loop. The NN works in Fourier space and selectively boosts low spatial frequencies, thus enabling faster convergence in a manner similar to accelerated gradient descent algorithms. We discuss the difficulties that arise when incorporating a NN into an iterative reconstruction algorithm and show how they can be overcome with iterative training. We apply our method to simulated and experimental data of gold nanoparticles on amorphous carbon and show that we can significantly speed up ptychographic reconstruction of the nanoparticles.

High‐pressure freezing (HPF) is an electron microscopy (EM) preparation technique with superb ultrastructural preservation. Combined with EM tomography it provides virtual EM serial sections with extraordinary spatial resolution. For HPF, cells are usually cultured on a rigid sapphire disc that provides a tight fit in the holding bracket of the HPF apparatus. Since we are using extensible elastic silicone membranes as a growth support to perform cell stretch experiments, we developed a method to clamp the stretched silicone membrane and place it instead of the sapphire disc into the HPF holding bracket. Compared to chemical fixation the HPF immobilised cells showed improved structural preservation, partly even on a molecular level. However, the outstanding quality of HPF immobilised cells on sapphire discs was not achieved. Moreover, regions with obvious freezing artefacts seemed to be more abundant in the HPF silicone membranes, probably caused by lower heat transfer rates of the silicone membrane during the HPF process. Taken together, we have shown that HPF immobilisation can be performed on growth supports different than sapphire discs. Since even stretched membranes can be used with the new method, also other unconventional growth supports should not pose a problem.

Background: The major advantage of scanning electron microscope (SEM) in biological research is that one can examine the morphology and surface features of specimens at high resolution. Specimens may differ from individual cells grown in culture to solid tissues or entire organisms measuring several centimetres in size. It literally permits an ‘in‐depth’ study of such specimens with great topography due to the incredible depth of field obtainable to the operator. Current study covers practical approaches of various biological samples' preparation and visualisation via scanning electron microscope. Methods: Alternate method of drying was employed over standard drying method; Critical Point Drying (CPD). Natural state of the microstructures of delicate specimens could be preserved by applying recommended reagents/ fixatives. Samples were treated with 2.5% w/w glutaraldehyde and reduced 1% Osmium tetroxide as primary and secondary fixatives. Samples were then serially dehydrated by graded ethanol (EtOH) and finally treated with chemical dehydrant Hexamethyldisilazane (HMDS). Results: Biological specimens, bacteria (Salmonella typhi and Staphylococcus aureus), bacterial crystal proteins, viruses (SARS‐CoV‐2), fungi (Aspergillus flavus), immune cells (monocytes) and invertebrates (Aedes aegypti), were studied and high‐resolution images were captured. Detailed structural features were studied using high voltage electron beams (10–20 KV). Secondary electrons and backscattered electrons were detected to reveal detailed surface features of the specimens. Conclusion: Chemical critical drying was found to be an economic and yet effective method with less apparent deterioration of the surface features. The advantages of using a chemical dehydrant like Hexamethyldisilazane (HMDS) include ease of use, relative quickness, and less expense than a CPD. Same technique can be applied for different specimens with same results.