A neural lens for super-resolution biological imaging

Fused Filament Fabrication (FFF) is the most popular and widely used additive manufacturing process for printing polymer and composite products. Various production factors influenced the strength and stiffness of the part manufactured by 3D printing. A comprehensive experimental analysis was conducted in this study to examine the effect of FFF process parameters (infill density, pattern, and layer thickness) on mechanical properties and failure mechanism. The tensile, flexural, and impact test specimens were printed using ABS and carbon fibre reinforced ABS filaments in accordance with ASTM standards. Furthermore, dynamic properties are studied using dynamic mechanical analysis to estimate the loss factor and glass transition temperature under the impact of temperature and frequency in addition to static properties. Further, the results showed the addition of carbon fiber in ABS increases the mechanical properties. The failure modes are studied using optical microscopy and Scanning Electron Microscopy images and it has been visualized that due to improper layer deposition, poor bonding between the previous layer and low infill density creates a void in the specimen which results in poor mechanical properties. The Dynamic Mechanical Analysis showed that at higher frequency the molecular movement decreases which in turn stabilizes the composite behavior and reduces the loss factor.

To develop a multi-modal deep learning method for automatic classification of immune-mediated glomerular diseases based on images of optical microscopy (OM), immunofluorescence microscopy (IM), and transmission electron microscopy (TEM). We retrospectively collected the pathological images from 273 patients and constructed a multi-modal multi- instance model for classification of 3 immune-mediated glomerular diseases, namely immunoglobulin A nephropathy (IgAN), membranous nephropathy (MN), and lupus nephritis (LN). This model adopts an instance-level multi-instance learning (I-MIL) method to select the TEM images for multi-modal feature fusion with the OM images and IM images of the same patient. By comparing this model with unimodal and bimodal models, we explored different combinations of the 3 modalities and the optimal methods for modal feature fusion. The multi-modal multi-instance model combining OM, IM, and TEM images had a disease classification accuracy of (88.34±2.12)%, superior to that of the optimal unimodal model [(87.08±4.25)%] and that of the optimal bimodal model [(87.92±3.06)%]. This multi- modal multi- instance model based on OM, IM, and TEM images can achieve automatic classification of immune-mediated glomerular diseases with a good classification accuracy.

Microstructural information acquired from image analysis can be used in cell modeling. In order to obtain more precise Solid Oxide Fuel Cell (SOFC) microstructure parameters, an adaptive fuzzy approach is developed for three-phase identification of YSZ/Ni anode Optical Microscopic (OM) images. A new quantum-inspired clique potential Markov random field (MRF) function is proposed to considerate spatial information in fuzzy logic model, where the space distance based weight is introduced to reflect the influence of neighborhood pixels. Simulated images and real SOFC anode OM images are used to compare the effectiveness and practicability of the proposed algorithm with others. Experiment results demonstrate that the proposed methods can accurately separate there-phase of SOFC OM images, which lays the foundation of subsequent microstructural parameters extracting.

In this study, Hardox 400 steel used as substrate material was coated through solid media Thermoreactive Diffusion (TRD) method using Ferro Niobium and Ferro Boron powders from carbide forming element powders. Coating was carried out in three different temperatures ([Formula: see text]C, [Formula: see text]C and [Formula: see text]C) and three different time intervals (1, 2 and 3[Formula: see text]h). Microstructures of the coated specimens were examined by optical microscope, Scanning Electron Microscope (SEM), Energy Dispersive X-Ray Spectroscopy (EDX) and X-ray Diffraction (XRD); and hardness values were measured. The effects of coating parameters on coating thickness and hardness were analyzed by ANOVA. In addition, specimens were subjected to wear tests to determine the effect of hardness and coating parameters on wear. In the wear tests, Taguchi test design setup was used. The obtained results were compared with the Hardox 400 steel used under current conditions. It was seen from optical microscope and SEM images that Hardox 400 steel surface could be coated with TRD method depending on coating parameters. The average thickness of NbC–B coating ranged from 1.797[Formula: see text][Formula: see text]m to 5.596[Formula: see text][Formula: see text]m under different process temperature and time. Rising the coating time and temperature increased the coating thickness by 311.40%. EDX analysis showed that the coating layer was composed of B, C, Fe and Nb elements, and XRD analysis also showed that the phase in the coating layer is NbC–B. The NbC–B phase was determined to be an important factor in increasing the hardness. The coating hardness is enhanced by 320.80% depending on the coating parameters. Optimum coating thickness, hardness and wear results were obtained from high coating temperature and time. Uncoated Hardox 400 steels were worn out more compared to the coated Hardox 400 steels. The contribution of coating temperature and time to wear resistance was 1.46% and 8.02%, respectively. It was observed that the important parameter for wear volume was the applied load.

We present here a modulating effect on lysozyme derived Amyloid β fibrils by aqueous magnetic fluid. This non-conventional approach of treatment of lysozyme derived Amyloid β fibrils showed lysing of Amyloid fibrils to its secondary structures which can be seen using optical microscope and scanning electron microscopic image. The size of lysozyme derived amyloid fibrils before and after treatment was measured using dynamic light scattering technique. The mechanism of defragmentation of lysozyme derived Amyloid β fibrils by magnetic fluid is explained. This is a first report to identify the secondary structure of protein using Fourier Transform Infrared (FTIR) and Circular Dichroism (CD) spectra after lysing. The cyto-toxicity study of this magnetic fluid on neuronal (SH-SY5Y) and non-neuronal (NRK) cell lines shows non-toxicity up to a concentration of 250 μg/mL. The study indicates a novel and unique complementary approach to treat the amyloidogenic brain diseases.

In their Progress Report, Sytnyk et al. claimed the epitaxial deposition of metal halide perovskites by inkjet printing on various substrates. The claims of epitaxy are doubtful due to a lack of convincing experimental evidence. Both optical microscopy and scanning electron microscopy images alone are insufficient to prove epitaxy. Even if the images suggest that epitaxial growth might have been achieved, direct evidences from X‐ray diffraction (XRD) or electron diffraction are still needed. Among the six perovskite‐substrate systems, only two CH3NH3PbI3 deposits on (111)PbTe/BaF2 are characterized with XRD. According to their XRD patterns, actually, it is estimated that each of their MAPbI3 deposits contains only <20% epitaxial components and the component of nearly random orientation is predominant besides considerable impurity PbI2. None of the MAPbI3 deposits can be considered epitaxial. It is critical to point out the errors in the methodology for verifying epitaxy, so that researchers would not be misled by Sytnyk's paper. Importantly, a new methodology to quantify the volume fraction of each epitaxial component in a thin film from XRD measurements is proposed. It is reliable to estimate the degree of texture according to the proposed equations, quite different from using Lotgering degree of orientation.

Inverse dynamical photon scattering (IDPS), an artificial neural network based algorithm for three-dimensional quantitative imaging in optical microscopy, is introduced. Because the inverse problem entails numerical minimization of an explicit error metric, it becomes possible to freely choose a more robust metric, to introduce regularization of the solution, and to retrieve unknown experimental settings or microscope values, while the starting guess is simply set to zero. The regularization is accomplished through an alternate directions augmented Lagrangian approach, implemented on a graphics processing unit. These improvements are demonstrated on open source experimental data, retrieving three-dimensional amplitude and phase for a thick specimen.

We determine the fracture toughness of a duplex stainless steel using the COD-concept (crack tip opening displacement). The CODi values are obtained by stereophotogrammetric reconstruction of fracture surfaces. This is done by analyzing stereoscopic scanning electron microscope (SEM) images of the surfaces with an automatic image processing system. This system allows the automatic generation of a digital elevation model (DEM) with approximately 50000 points for each fracture surface image. Height-profiles at the point of crack initiation can be obtained from this DEM. The examination of corresponding profiles from both specimen-halves leads to CODi.

Gene expression is dynamic and heterogeneous across all cell types. It serves several fundamental functions such as adaptation for changing environment and developing tissue specific phenotype and functionality during development and in adult organism. Control of gene expression mostly occurs at the transcription stage. Two main factors that regulate transcription in high eukaryotes are chromatin organization and transcription factor (TF) protein dynamics. Until recently they were mostly studied by methods that are performed in vitro or/and require ensemble averaging. We have been contributing to a development of new single molecule imaging methods that allow quantifying chromatin organization and TF dynamics in vivo and at single cell/single molecule level. Here we present our methodological developments and apply them to study how chromatin organization and TF dynamics regulate transcription. On the technical side of the study we, first, developed a method of imaging DNA with enhanced resolution in living cells. Super-resolution (SR) point localization based microscopy allows visualizing sub-diffraction organization of cellular structures and their dynamics in vivo. However, until recently almost exclusively proteins were imaged in SR microscopy. We found photoswitching conditions to image with stochastic optical reconstruction microscopy (STORM) DNA directly using the site-specific DNA –binding dye Picogreen (Chapter 2). We achieved a resolution of 50-70 nm which is 5 fold better than conventional microscopy. Due to the excellent dye preservation we were able to do time lapse SR imaging. This study was a first demonstration of using a site-specific dye for STORM SR imaging in living cells. Photoswitching principle used in SR microscopy can be applied to single molecule (SM) tracking. It provides up to 1000x increase in density of trajectories compared to classic SM tracking. Initially, photoactivatable (PA) fusion proteins were used for high density SR based tracking. However, PA proteins are relatively dim and there is a limited choice of them. We demonstrated a new approach to SR based high density tracking by using organic dyes’ photoswitching to track proteins (Chapter 4). We showed that different dyes can be photoswitched in similar live-cell compatible media on membrane and in different organelles. Together with orthogonal protein labelling schemes this allows us to perform multicolour high density tracking across different compartments. This approach gives flexibility of labelling and increased track length compared to PA-FP SR based tracking. In a separate study, using tracking we developed algorithm which can account for long-living molecules in live cell SR imaging and allows us to correct clustering artefacts (Chapter 5). [...]

Accurate identification and classification of atmospheric particulates can provide the basis for their source apportionment. Most current research studies mainly focus on the classification of atmospheric particles based on the energy spectrum of particles, which has the problems of low accuracy and being time-consuming. It is necessary to study the classification method of atmospheric particles with higher accuracy. In this paper, a convolutional neural network (CNN) model with attention mechanism is proposed to identify and classify the scanning electron microscopy (SEM) images of atmospheric particles. First, this work established a database, Qingdao 2016–2018, for atmospheric particles classification research. This database consists of 3469 SEM images of single particulates. Secondly, by analyzing the morphological characteristics of single particle SEM images, it can be divided into four categories: fibrous particles, flocculent particles, spherical particles, and mineral particles. Thirdly, by introducing attention mechanism into convolutional neural network, an Attention-CNN model for the identification and classification of the four types of atmospheric particles based on the SEM images is established. Finally, the Attention-CNN model is trained and tested based on the SEM images database, and the results of identification and classification for four types of particles are obtained. Under the same SEM images database, the classification results from Attention-CNN are compared with those of CNN and SVM. It is found that Attention-CNN has higher classification accuracy and reduces significantly the misclassification number of particles, which shows the focusing effect of attention mechanism.

Polycrystalline silicon thin films on glass substrate

We present here our results on surface micro-structuring via nanosecond pulsed laser irradiation of Titanium and Stainless Steel cathode samples. Laser based surface micro-structuring leading to formation of self-assembled micro-tips can potentially enhance field emission efficiency of such surface treated cathodes. Microstructure of the laser treated surfaces has been observed under a Scanning Electron Microscope (SEM) and SEM images were further analyzed using software attached with Optical Microscope. To study the effect of laser fluence on developed surface microstructure, the target surface was irradiated in different regions with laser beams at varying laser fluence levels ranging from 2-10 J/cm 2 for a period of 1 to 15 minutes corresponding to 600 to 9000 laser pulses. Mean height of the generated micro-cones was observed to increase from 17 to 30µm on increasing number of irradiating laser pulses from 3000 to 9000 in case of Stainless Steel samples. In case of laser treated Titanium average periodicity of generated self-assembled micro-cones decreased from 10.8μm to 6.5μm when laser fluence was increased from 5 to 10J/cm 2 with a total of 600 laser pulses used for irradiating the sample.

.We present a multimodal visible light optical coherence microscopy (OCM) and fluorescence imaging (FI) setup. Specification and phantom measurements were performed to characterize the system. Two applications in neuroimaging were investigated. First, curcumin-stained brain slices of a mouse model of Alzheimer’s disease were examined. Amyloid-beta plaques were identified based on the fluorescence of curcumin, and coregistered morphological images of the brain tissue were provided by the OCM channel. Second, human brain tumor biopsies retrieved intraoperatively were imaged prior to conventional neuropathologic work-up. OCM revealed the three-dimensional structure of the brain parenchyma, and FI added the tumor tissue-specific contrast. Attenuation coefficients computed from the OCM data and the florescence intensity values were analyzed and showed a statistically significant difference for 5-aminolevulinic acid (5-ALA)-positive and -negative brain tissues. OCM findings correlated well with malignant hot spots within brain tumor biopsies upon histopathology. The combination of OCM and FI seems to be a promising optical imaging modality providing complementary contrast for applications in the field of neuroimaging.

To investigate optical coherence microscopy (OCM) imaging features and the application value of these high-resolution images for identifying endocervical canal lesions (ECLs), which is a clinical dilemma in cervical cancer screening programs. In total, 520 OCM images were obtained by scanning the cervical canal lesions with an ultra-high-resolution OCM system (204 specimens from 73 patients). The OCM morphologic characteristics of ECLs were observed and summarized, and then 3 researchers performed a diagnostic test of OCM images of cervical canal lesions. The accuracy, sensitivity, specificity, positive predictive value, negative predictive value, 95% confidence interval of each parameter, and interinvestigator agreement (κ) were calculated. Normal endocervix, cysts, squamous metaplasia, high-grade squamous intraepithelial lesions involving glands, and invasive carcinoma had distinct OCM characteristics, which correlated well with corresponding H&E histologic sections. The accuracy, sensitivity, and specificity of the 3 researchers were 90.6%, 89.3% (95% CI, 86.5%-91.7%) and 91.6% (95% CI, 89.2%-93.5%), respectively. The positive predictive value was 90.1% (95% CI, 87.3%-92.4%), and the negative predictive value was 90.9% (95% CI, 88.5%-92.9%), with almost perfect agreement (κ = 0.874). The application of the OCM system in cervical canal lesions is feasible and could help improve detection of occult ECLs in cervical cancer screening programs. This study lays the foundation for further research on OCM in cervical canal lesions in vivo, which also has a potential impact on projecting pathologic evaluation beyond what is currently possible, perhaps globally.

Research on the comparative morphology of pollen grains depends crucially on the application of appropriate microscopy techniques. Information on the performance of microscopy techniques can be used to inform that choice. We compared the ability of several microscopy techniques to provide information on the shape and surface texture of three pollen types with differing morphologies. These techniques are: widefield, apotome, confocal and two-photon microscopy (reflected light techniques), and brightfield and differential interference contrast microscopy (DIC) (transmitted light techniques). We also provide a first view of pollen using super-resolution microscopy. The three pollen types used to contrast the performance of each technique are: Croton hirtus (Euphorbiaceae), Mabea occidentalis (Euphorbiaceae) and Agropyron repens (Poaceae). No single microscopy technique provided an adequate picture of both the shape and surface texture of any of the three pollen types investigated here. The wavelength of incident light, photon-collection ability of the optical technique, signal-to-noise ratio, and the thickness and light absorption characteristics of the exine profoundly affect the recovery of morphological information by a given optical microscopy technique. Reflected light techniques, particularly confocal and two-photon microscopy, best capture pollen shape but provide limited information on very fine surface texture. In contrast, transmitted light techniques, particularly differential interference contrast microscopy, can resolve very fine surface texture but provide limited information on shape. Texture comprising sculptural elements that are spaced near the diffraction limit of light (∼250 nm; NDL) presents an acute challenge to optical microscopy. Super-resolution structured illumination microscopy provides data on the NDL texture of A. repens that is more comparable to textural data from scanning electron microscopy than any other optical microscopy technique investigated here. Maximizing the recovery of morphological information from pollen grains should lead to more robust classifications, and an increase in the taxonomic precision with which ancient vegetation can be reconstructed.