High-resolution Microscopy Images Research Articles

Ex vivo Live Cell Imaging of Nanoparticle-Cell Interactions in the Mouse Lung.

A successful clinical translation of novel nanoparticle-based cancer therapeutics requires a thorough preclinical investigation of their interaction with immune, tumor and endothelial cells as well as components of the tumor-microenvironment. Although high-resolution microscopy images of fixed tumor tissue specimens can provide valuable information in this regard, they are only static snapshots of a momentary event. Here we describe a superior alternative fluorescence microscopy approach to assess the feasibility of investigating nanoparticle-cell interactions in the mouse lung live and over time at nanometer resolution. We applied fluorescent lung tumor cells and Barium-based fluorescently labeled nanoparticles to nude mice or to CD68-EGFP transgenic mice for visualization of the monocyte-macrophage lineage. Shortly before imaging, fluorescently labeled lectin was intravenously injected for staining of the blood vessels. The lung was filled ex vivo with 1% agarose and individual lung lobes were imaged over time using a confocal microscope with Airyscan technology. Time series demonstrate that live cell imaging of lung lobes can be performed for at least 4 h post mortem. Time-lapse movies illustrate the dynamics of the nanoparticles within the pulmonary circulation and their uptake by immune cells. Moreover, the exchange of nanoparticle material between cancer cells was observed over time. Fluorescent monocytes in lungs of CD68-EGFP transgenic mice could be visualized within blood vessels in the process of interaction with tumor cells and nanoparticles. This high resolution ex vivo live cell imaging approach provides an excellent 4D tool to obtain valuable information on the behavior of tumor and immune cells at first encounter with nanoparticles and may contribute to the understanding of how nanoparticles interact with cells supporting the development of therapeutic strategies based on nanoparticulate drug delivery systems.

Multiplex Fourier ptychographic reconstruction with model-based neural network for Internet of Things

Fourier ptychographic microscopy (FPM) is a newly developed technique to capture wide field-of-view (FOV) and high-resolution images that meet the demands of Internet of Things (IoT). FPM enlarges the equivalent numerical aperture of the system and achieves phase imaging by simply employing an angle-varied illumination module. Recently, researches propose to perform the FPM reconstruction with deep learning servers which is costly and requires large datasets. In this paper, we present a new FPM image reconstruction framework termed multi-NNP for Internet of Medical Things (IoMT). Multi-NNP performs multiplex ptychographic reconstruction with a model-based neural network locally rather than on deep learning servers. Our framework simplifies the process and improves the reconstruction performance which promotes the application of wide-field, high-resolution microscopic images in IoMT. Experimental results demonstrate the performance and effectiveness of the proposed framework.

Systematic evaluation of particle loss during handling in the percutaneous transluminal angioplasty for eight different drug-coated balloons

Paclitaxel drug coated balloons (DCBs) should provide optimal drug transfer exclusively to the target tissue. The aim of this study was to evaluate the particle loss by handling during angioplasty. A robotic arm was developed for systematic and reproducible drug abrasion experiments. The contact force on eight different commercially available DCB types was gradually increased, and high-resolution microscopic images of the deflated and inflated balloons were recorded. Three types of DCBs were classified: no abrasion of the drug in both statuses (deflated and inflated), significant abrasion only in the inflated status, and significant abrasion in both statuses. Quantitative measurements via image processing confirmed the qualitative classification and showed changes of the drug area between 2.25 and 45.73% (13.28 ± 14.29%) in the deflated status, and between 1.66 and 40.41% (21.43 ± 16.48%) in the inflated status. The structures and compositions of the DCBs are different, some are significantly more susceptible to drug loss. Particle loss by handling during angioplasty leads to different paclitaxel doses in the target regions for same DCB types. Susceptibility to involuntary drug loss may cause side effects, such as varying effective paclitaxel doses, which may explain variations in studies regarding the therapeutic outcome.

¡Estudiando el Desarrollo Embrionario y Fetal con el Microscopio Virtual! En Tiempos de Covid-19

La Microscopía Virtual es una herramienta tecnológica que permite la visualización de imágenes digitales microscópicas de gran resolución a través de un computador imitando la funcionalidad de un microscopio óptico tradicional. El presente trabajo presenta nuestra experiencia en el uso de esta modalidad de trabajo, útil hoy en día, en medio de la pandemia por Covid-19.

RTIP: A FULLY AUTOMATED ROOT TIP TRACKER FOR MEASURING PLANT GROWTH WITH INTERMITTENT PERTURBATIONS.

RTip is a tool to quantify plant root growth velocity using high resolution microscopy image sequences at sub-pixel accuracy. The fully automated RTip tracker is designed for high-throughput analysis of plant phenotyping experiments with episodic perturbations. RTip is able to auto-skip past these manual intervention perturbation activity, i.e. when the root tip is not under the microscope, image is distorted or blurred. RTip provides the most accurate root growth velocity results with the lowest variance (i.e. localization jitter) compared to six tracking algorithms including the top performing unsupervised Discriminative Correlation Filter Tracker and the Deeper and Wider Siamese Network. RTip is the only tracker that is able to automatically detect and recover from (occlusion-like) varying duration perturbation events.

Consistency and variation of protein subcellular location annotations.

A major challenge for protein databases is reconciling information from diverse sources. This is especially difficult when some information consists of secondary, human-interpreted rather than primary data. For example, the Swiss-Prot database contains curated annotations of subcellular location that are based on predictions from protein sequence, statements in scientific articles, and published experimental evidence. The Human Protein Atlas (HPA) consists of millions of high-resolution microscopic images that show protein spatial distribution on a cellular and subcellular level. These images are manually annotated with protein subcellular locations by trained experts. The image annotations in HPA can capture the variation of subcellular location across different cell lines, tissues, or tissue states. Systematic investigation of the consistency between HPA and Swiss-Prot assignments of subcellular location, which is important for understanding and utilizing protein location data from the two databases, has not been described previously. In this paper, we quantitatively evaluate the consistency of subcellular location annotations between HPA and Swiss-Prot at multiple levels, as well as variation of protein locations across cell lines and tissues. Our results show that annotations of these two databases differ significantly in many cases, leading to proposed procedures for deriving and integrating the protein subcellular location data. We also find that proteins having highly variable locations are more likely to be biomarkers of diseases, providing support for incorporating analysis of subcellular location in protein biomarker identification and screening.

Lissajous scanning structured illumination microscopy.

High-resolution fluorescent microscopic imaging techniques are in high demand to observe detailed structures or dynamic mechanisms of biological samples. Structured illumination microscopy (SIM) has grabbed much attention in super-resolution imaging due to simple configuration, high compatibility with common fluorescent molecules, and fast image acquisition. Here, we report Lissajous scanning SIM (LS-SIM) by using a high fill-factor Lissajous scanning micromirror and laser beam modulation. The LS-SIM was realized by a Lissajous scanned structured illumination module, relay optics, and a conventional fluorescent microscope. The micromirror comprises an inner mirror and an outer frame, which are scanned at pseudo-resonance with electrostatic actuation. The biaxial scanning frequencies are selected by the frequency selection rule for high fill-factor (> 80%) Lissajous scanning. Structured illumination (SI) was then realized by modulating the intensity of a laser beam at the least common multiple (LCM) of the scanning frequencies. A compact Lissajous scanned SI module containing a fiber-optic collimator and Lissajous micromirror has been fully packaged and coupled with relay optics and a fiber-based diode pumped solid state (DPSS) laser including acousto-optic-modulator (AOM). Various structured images were obtained by shifting the phase and orientation of the illumination patterns and finally mounted with a conventional fluorescent microscope. The LS-SIM has experimentally demonstrated high-resolution fluorescent microscopic imaging of reference targets and human lung cancer cell PC-9 cells. The LS-SIM exhibits the observable region in spatial frequency space over 2x, the line-edge sharpness over 1.5x, and the peak-to-valley (P-V) ratio over 2x, compared to widefield fluorescent microscopy. This method can provide a new route for advanced high-resolution fluorescent microscopic imaging.

High-Resolution Vertical Polarization Excited Dark-Field Microscopic Imaging of Anisotropic Gold Nanorods for the Sensitive Detection and Spatial Imaging of Intracellular microRNA-21.

As an important biomarker for early diagnosis of cancers, sensitive detection and high-resolution imaging of microRNA-21 in cancer cells have become important and challengeable. In this work, highly sensitive detection and spatial imaging of intracellular microRNA-21 were realized by the reduced signal background through vertical polarization excitation with a polarizer. The lateral local surface plasmon resonance property of gold nanorods (AuNRs) displayed a pronounced green color with low scattering intensity, which was adjusted to red color with strong scattering intensity when the core-satellite gold nanoparticle (AuNP) assembly was constructed on the side of AuNRs through a catalyzed hairpin assembly (CHA) circuit in the presence of microRNA-21. This unique approach allows for effectively reducing the strong background signal to improve the sensitivity of detection. Additionally, the proposed strategy can not only realize the sensitive detection of microRNA-21 with the limit of detection as low as 2 pM (3σ) but also achieve the high spatial imaging of cancer cells, which provided a specific strategy for the construction and imaging of intracellular imaging probes. It is believed that the simple and sensitive approach on the basis of lateral local surface plasmon resonance property of anisotropic AuNRs with excellent sensitivity combined with high spatial imaging holds promising potentials to visualize intracellular microRNAs with low abundance.

Abstract 4791: Metabolic switch from glycolysis to oxidative phosphorylation (ox-phos) provides survival advantage to anti-androgen-treated prostate cancer cells and make them vulnerable to mitochondrial metabolism inhibitors IACS-010759 and CB-839

Abstract Background: Most cancer cells depend more on glycolysis for their energy need compared with their normal counterparts incubated under identical condition. In addition to glycolysis, tumor cells also engage mitochondrial ox-phos to support growth. Here, we show that PCa cells surviving under anti-androgen enzalutamide (ENZA) treatment switch from glycolysis to ox-phos for their energy need and are more vulnerable to mitochondria-targeted metabolic inhibitors. We have also standardized a high-resolution quantitative fluorescence microscopic method to detect this metabolic switch in patient circulating tumor cells (CTCs). Methods: Seahorse assay has been used to compare mitochondrial metabolism in ENZA treated anti-androgen-sensitive LNCaP and -resistant C4-2 and PCa2b cells. The Seahorse data were supported by mass spectroscopic analysis of corresponding metabolites and metabolic fluxes. An ex vivo fluorescence staining for the CTC mitochondria with high ox-phos followed by high-resolution quantitative microscopic image analysis method has been standardized to follow such changes in cultured cells and patient CTCs. Results: Seahorse, mass spectroscopy and high-resolution microscopy of mitochondrial ox-phos showed that ENZA significantly decreases glycolysis and increases ox-phos in all surviving PCa cells within 24h of treatment. These cells are more vulnerable to treatment with mitochondrial ox-phos inhibitor IACS-010759 and a glutaminase inhibitor CB-839. High-resolution microscopic analysis of CTCs has thus far been performed in 18 patient blood samples. Six out of the 18 patients developed resistance to anti-androgen therapy within 0-6 months of sample collection. CTCs from all six patients showed a relatively higher average fluorescence due to high mitochondrial ox-phos as compared with the rest of the patients. Discussion: The data presented here may lead to informed combination therapy for selected PCa patients developing resistance to anti-androgen therapy for better clinical outcome. Citation Format: Hirak S. Basu, Nathaniel Wilganowski, Samantha Robertson, Sumankalai Ramachandran, Amado Zurita-Saavedra, Mark Titus, Evan Cohen, James Reuben, George Wilding. Metabolic switch from glycolysis to oxidative phosphorylation (ox-phos) provides survival advantage to anti-androgen-treated prostate cancer cells and make them vulnerable to mitochondrial metabolism inhibitors IACS-010759 and CB-839 [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 4791.

High-Resolution Ion Microscope Imaging over Wide Mass Ranges Using Electrodynamic Postextraction Differential Acceleration.

A time-dependent postextraction differential acceleration (PEDA) potential was used to temporally focus increasingly heavy ions in a stigmatic imaging mass spectrometer, allowing them to be imaged with high mass and spatial resolutions over a broad mass-to-charge (m/z) range. By applying a linearly rising potential to the ion extraction electrode, sequential m/z ratios were subjected to a changing electric field, allowing their foci to coincide at the detector. Using this approach, at least 75% of the maximum mass resolution was obtained over a 300–600 Da range when the ion microscope was focused around 450 Da, representing more than a 10-fold increase over the conventional single-field PEDA method.

Molecular and cellular correlates in Kv channel clustering: entropy-based regulation of cluster ion channel density

Scaffold protein-mediated ion channel clustering at unique membrane sites is important for electrical signaling. Yet, the mechanism(s) by which scaffold protein-ion channel interactions lead to channel clustering or how cluster ion channel density is regulated is mostly not known. The voltage-activated potassium channel (Kv) represents an excellent model to address these questions as the mechanism underlying its interaction with the post-synaptic density 95 (PSD-95) scaffold protein is known to be controlled by the length of the extended ‘ball and chain’ sequence comprising the C-terminal channel region. Here, using sub-diffraction high-resolution imaging microscopy, we show that Kv channel ‘chain’ length regulates Kv channel density with a ‘bell’-shaped dependence, reflecting a balance between thermodynamic considerations controlling ‘chain’ recruitment by PSD-95 and steric hindrance due to the spatial proximity of multiple channel molecules. Our results thus reveal an entropy-based mode of channel cluster density regulation that mirrors the entropy-based regulation of the Kv channel-PSD-95 interaction. The implications of these findings for electrical signaling are discussed.

Electron-Beam Excitation Assisted (EXA) Optical Microscopy for High Resolution Imaging

Electron-Beam Excitation Assisted (EXA) Optical Microscopy for High Resolution Imaging

A comparison of nano-scale pore attributes of Barakar Formation gas shales from Raniganj and Wardha Basin, India using low pressure sorption and FEG-SEM analysis

This study investigates the micropore-mesopore spectrum in gas shales by augmenting two different pore size distribution calculation methods using two probe gases in low-pressure sorption analysis. The Barakar Formation shale samples collected from Wardha and Asansol belonging to two key potential basins in India, namely Wardha and Raniganj, were studied to identify possible variations in pore attributes in shales, and a comparison was drawn. A combination of CO2 and N2 low-pressure gas adsorption studies was implemented to resolve micro- and meso-pores, using the respective probe adsorbate. The visual representation of the pore structure and its disposition was facilitated by high-resolution microscopic imaging, and aided by image analysis. The results show a positive correlation between clay mineral composition and the mesopore attributes as well as organic matter composition and micropore attributes. This confirms that both inorganic and organic content of shales contains large pore volumes, albeit in different pore dimension ranges with some overlap. High N2 adsorption potential of Asansol shale, as well as high CO2 adsorption potential of Wardha valley shale, is conformable with their respective pore size distributions. Fractal characteristics of mesopores suggest that the larger mesopores have a rougher surface, whereas the smaller mesopores have a smoother surface (with average fractal dimensions of 2.6 and 2.2, respectively). Although all examined shales belong to the same geological formation, they exhibit distinctly variable pore attributes and gas storage potential.

UV irradiation of HSV-1 virions induces the differential recruitment of host intrinsic and innate immune regulators to infecting viral genomes

Recognition of virus-derived nucleic acids is a key process in the activation of immune defences to viral infection. However, the spatiotemporal recruitment of host immune regulators to infecting viral DNA (vDNA) genomes remains poorly defined. Here we utilize 5-Ethynyl-2’-deoxycytidine (EdC) nucleotide labelling of WT or UV-irradiated HSV-1 genomes in combination with click chemistry and high-resolution confocal microscopy imaging to investigate the recruitment and proximity of key intracellular immune regulators to infecting vDNA during HSV-1 infection. We report that UV irradiation of HSV-1 virions restricts vDNA decompaction upon delivery to infected cells. UV treatment reduced the frequency in PML-NB recruitment to nuclear genomes relative to non-irradiated vDNA. These data demonstrate an important role of genome decompaction in the recruitment of intrinsic host factors to nuclear infecting vDNA. Additionally, UV-irradiated HSV-1 genomes favoured a premature cytosolic deposition phenotype, due to ineffective association with microtubules as shown by nocodazole treatment. The cytosolic UV irradiated vDNA showed enhanced association with both cGAS and STING. Interestingly, high-resolution Airyscan imaging identified a consistent unreported 3D interaction between both cGAS and STING, the strength of association of which was enhanced in the context of UV infection. This enhanced recruitment correlated with elevated levels of IFN-β transcription relative to non-irradiated vDNA, despite the presence of non-irradiated cytosolic genomes. Our data demonstrates that UV irradiation of herpesviruses, a well-established procedure for virus inactivation, has multiple effects on genome structure and sub-cellular localisation that differentially influence the spatiotemporal recruitment of intrinsic and innate host immune regulators to infecting genomes.

Fine structure analysis of perineuronal nets in the ketamine model of schizophrenia.

Perineuronal nets (PNNs) represent a highly condensed specialized form of brain extracellular matrix (ECM) enwrapping mostly parvalbumin-positive interneurons in the brain in a mesh-like fashion. PNNs not only regulate the onset and completion of the critical period during postnatal brain development, control cell excitability, and synaptic transmission but are also implicated in several brain disorders including schizophrenia. Holes in the perineuronal nets, harboring the synaptic contacts, along with hole-surrounding ECM barrier can be viewed as PNN compartmentalization units that might determine the properties of synapses and heterosynaptic communication. In this study, we developed a novel open-source script for Fiji (ImageJ) to semi-automatically quantify structural alterations of PNNs such as the number of PNN units, area, mean intensity of PNN marker expression in 2D and 3D, shape parameters of PNN units in the ketamine-treated Sprague-Dawley rat model of schizophrenia using high-resolution confocal microscopic images. We discovered that the mean intensity of ECM within PNN units is inversely correlated with the area and the perimeter of the PNN holes. The intensity, size, and shape of PNN units proved to be three major principal factors to describe their variability. Ketamine-treated rats had more numerous but smaller and less circular PNN units than control rats. These parameters allowed to correctly classify individual PNNs as derived from control or ketamine-treated groups with ≈85% reliability. Thus, the proposed multidimensional analysis of PNN units provided a robust and comprehensive morphometric fingerprinting of fine ECM structure abnormalities in the experimental model of schizophrenia.

Colorectal cysts as a validating tool for CAR therapy

BackgroundTreatment of cancers has largely benefited from the development of immunotherapy. In particular, Chimeric Antigen Receptor (CAR) redirected T cells have demonstrated impressive efficacy against B-cell malignancies and continuous efforts are made to adapt this new therapy to solid tumors, where the immunosuppressive tumor microenvironment is a barrier for delivery. CAR T-cell validation relies on in vitro functional assays using monolayer or suspension cells and in vivo xenograft models in immunodeficient animals. However, the efficacy of CAR therapies remains difficult to predict with these systems, in particular when challenged against 3D organized solid tumors with highly intricate microenvironment. An increasing number of reports have now included an additional step in the development process in which redirected T cells are tested against tumor spheres.ResultsHere, we report a method to produce 3D structures, or cysts, out of a colorectal cancer cell line, Caco-2, which has the ability to form polarized spheroids as a validation tool for adoptive cell therapy in general. We used CD19CAR T cells to explore this method and we show that it can be adapted to various platforms including high resolution microscopy, bioluminescence assays and high-throughput live cell imaging systems.ConclusionWe developed an affordable, reliable and practical method to produce cysts to validate therapeutic CAR T cells. The integration of this additional layer between in vitro and in vivo studies could be an important tool in the pre-clinical workflow of cell-based immunotherapy.

Self-Healing Dyes-Keeping the Promise?

Self-healing dyes have emerged as a new promising class of fluorescent labels. They consist of two units, a fluorescent dye and a photostabilizer. The latter heals whenever the fluorescent dye is in danger of taking a reaction pathway toward photobleaching. We describe the underlying concepts and summarize the developmental history and state-of-the-art, including latest applications in high-resolution microscopy, live-cell, and single-molecule imaging. We further discuss remaining limitations, which are (i) lower photostabilization of most self-healing dyes when compared to solution additives, (ii) limited mechanistic understanding on the influence of the biochemical environment and molecular oxygen on self-healing, and (iii) the lack of cheap and facile bioconjugation strategies. Finally, we provide ideas on how to further advance self-healing dyes, show new data on redox blinking caused by double-stranded DNA, and highlight forthcoming work on intramolecular photostabilization of fluorescent proteins.

Optimizing the 1525 Nm Emission of Er3+-Sensitized Nanoprobe for In Vivo NIR-IIb Luminescence Bioimaging

Introduction Fluorophores emitting in the second near-infrared window (NIR-II, 900-1700 nm)[1] have exhibited significant improvements in detection depth and resolution owing to low tissue autofluorescence and up to a 1000-fold reduction in scattering losses compared to those for visible (400-780 nm)[2] and NIR-I regions (780-900 nm)[3]. Especially, the range of 1500-1700 nm is denoted as the “NIR-IIb” region, which provides the lowest photon scattering in the NIR-II region, thus allowing for the most desirable imaging clarity and deep penetration using existing InGaAs detectors [4]. Unfortunately, NIR-IIb fluorescent probes with sufficient brightness are scarce. Herein, we present an efficient NIR-IIb imaging agent NaErF4:0.5%Tm3+@NaLuF4 with a shell thickness of 4.30 nm. The influence of the types of epitaxial heterogeneous shells, the doping effect and optimal doping concentration of Tm3+ ions, as well as the critical shell thickness for obtaining the surface quenching-assisted downshifting were systematically investigated to acquire efficient 1525 nm emission of Er3+ ions under 800 nm excitation. The quantum yield in 1500-1700 nm region reached to 13.92%, enabling high-resolution through-skull microscopic cerebrovascular imaging and large-depth in vivo gastrointestinal tract imaging with extremely low excitation powder density of 35mW/cm2. Method The 793 nm laser beam emitted by the semiconductor laser (Suzhou Rugkuta Optoelectronics Co., Ltd., China) was collimated and expanded through the lens so that the sample can be uniformly illuminated. The emitting fluorescent signal was collected onto the InGaAs camera (Tekwin, China) through a prime lens (focal length: 50 mm, Edmund Optics). A 1450 nm long-pass filter (Thorlabs) was placed in front of the detector to extract the pure NIR-IIb signals. After anesthesia, the mice were placed on the imaging platform. The PEGylated β-NaErF4:0.5%Tm@4.30nm NaLuF4 nanoparticles (200 μL, 5 mg/mL) was injected into the tail vein, then the laser was turned on immediately and the image was recorded at different time points.The β-NaErF4:0.5%Tm@4.30nm NaLuF4 nanoparticles (200 μL, 5 mg/mL) were perfused into the stomach of nude mice. The images of the mice were recorded using the above-mentioned NIR-IIb fluorescence macroscopic imaging system. Results and Conclusions The brightest β-NaErF4:0.5%Tm@4.30nm NaLuF4 nanoparticels (NPs) were used for the in vivo bio-imaging. Firstly, 200 μL NPs (PEGylated β-NaErF4:0.5%Tm@4.30nm NaLuF4, 5mg/mL) were intravenously injected into the mouse for the whole body imaging (Fig.1), which verified the feasibility of the probe for NIR-IIb in vivo imaging. Under the excitation of a 793 nm laser (~35 mW/cm2), a bright, clear vascular network in the nearly “transparent” body was presented. Within 5 min post-injection, it could be seen that the liver has been gradually brighter, for which the NPs were captured by the liver (Fig.1B). Thanks to the low autofluorescence and suppressed tissue scattering in NIR-IIb region, a clear outline of liver was distinguished. It is worth noticing that, owing to the excellent luminescence properties of the probe, the excitation power density at the imaging plane could be as low as ~35 mW/cm2, which is much smaller than that in previously reported work.[5] The deep-tissue penetration in the NIR-IIb region was well validated in Fig. 2. After anesthesia, a mouse with a cranial window and a mouse with intact skull were injected with PEGylated β-NaErF4:0.5%Tm@4.30nm NaLuF4 NPs (200 μL, 100 mg/mL) respectively and fixed in the imaging system for microscopic cerebrovascular imaging. The 793 nm laser beam was expanded, reflected, and uniformly illuminated in the region of interest. The NIR-IIb fluorescence signals emitted by the NPs in the blood vessel were collected by the objective lens and focused by the tube lens onto the InGaAs detector (Tekwin, China). In order to visually show the structure of the blood vessels in the mouse brain, we stitched several images recorded by a scan lens (LSM03, Thorlabs) into an image with a large field of view. Compared to open-skull brain imaging (Fig.2A), NIR-IIb through-skull imaging (Fig.2B) also provided fairly rich details, and a small blood vessel with the diameter of 18.609 μm was easily identified.

3-D-Printed Registration Phantom for Combined Ultrasound and Optical Imaging of Biological Tissues

Efforts to develop quantitative ultrasound biomarkers would benefit from comparisons between ultrasound data and higher-resolution images of the tissue microstructure, such as from optical microscopy. However, only a few studies have used these methods for multiscale imaging because it is difficult to register low-resolution (>100 μm) ultrasound images to high-resolution microscopy images. To address this need, we have designed a 3-D-printed registration phantom that is made of a hard fluorescent resin, fits into a glass-bottom dish and can be used to calculate a coordinate system transform between ultrasound and optical microscopy. We report the phantom design, a registration protocol and an example registration using 18.5-MHz ultrasound and second harmonic generation microscopy. We evaluate the registration precision, achieving standard deviations smaller than the ultrasound resolution across all axes, and illustrate on a mouse mammary gland that this method yields results superior to those of manual landmark registration.

Application of CNN networks for an automatic determination of critical loads in scratch tests on a-C:H:W coatings

To enable an automatic evaluation of the scratch test according to DIN EN ISO 20502:2016-11, a new approach using convolutional neural networks (CNN) is developed in this study. For this purpose, a dataset of 2860 image sections is created using high resolution microscopy images of macro- and micro-scratches in a-C:H(:W) coatings, deposited by physical vapour deposition (PVD). Through transfer learning, four different pre-trained CNN networks are qualified to automatically detect the damage patterns of the critical loads Lc1 to Lc3. The classification accuracy is checked with a dataset of separated test images. Verification is done by the evaluation of additional scratch tracks, for which the values of Lc1 to Lc3 in Newton (N) are automatically determined by addition of a suitable image processing. The CNN network models AlexNet, Inception-v3, ResNet-101 and VGG-19 are being investigated for their applicability. All models show an exact detection range of the damage of Lc2 and Lc3, but only the network VGG-19 also classifies Lc1 in good agreement to the human evaluation. The automatic evaluation is robust against image or coating defects. In addition, the method exhibits a high potential for a future implementation on other coating types and surface conditions.