A deep learning approach for line-level Amharic Braille image recognition.

Braille, the most popular tactile-based writing system, uses patterns of raised dots arranged in cells to inscribe characters for visually impaired persons. Amharic is Ethiopia's official working language, spoken by more than 100 million people. To bridge the written communication gap between persons with and without eyesight, multiple Optical braille recognition systems for various language scripts have been developed utilizing both statistical and deep learning approaches. However, the need for half-character identification and character segmentation has complicated these systems, particularly in the Amharic script, where each character is represented by two braille cells. To address these challenges, this study proposed deep learning model that combines a CNN and a BiLSTM network with CTC. The model was trained with 1,800 line images with 32 × 256 and 48 × 256 dimensions, and validated with 200 line images and evaluated using Character Error Rate. The best-trained model had a CER of 7.81% on test data with a 48 × 256 image dimension. These findings demonstrate that the proposed sequence-to-sequence learning method is a viable Optical Braille Recognition (OBR) solution that does not necessitate extensive image pre and post processing. Inaddition, we have made the first Amharic braille line-image data set available for free to researchers via the link: https://github.com/Ne-UoG-git/Am-Br-line-image.github.io .

Multifocal optical projection microscopy enables label-free 3D measurement of cardiomyocyte cluster contractility.

Human induced pluripotent stem cell (hiPSC)-derived cardiomyocyte (CM) models have become an attractive tool for in vitro cardiac disease modeling and drug studies. These models are moving towards more complex three-dimensional microphysiological organ-on-chip systems. Label-free imaging-based techniques capable of quantifying contractility in 3D are needed, as traditional two-dimensional methods are ill-suited for 3D applications. Here, we developed multifocal (MF) optical projection microscopy (OPM) by integrating an electrically tunable lens to our in-house built optical projection tomography setup for extended depth of field brightfield imaging in CM clusters. We quantified cluster biomechanics by implementing our previously developed optical flow-based CM video analysis for MF-OPM. To demonstrate, we acquired and analyzed multiangle and multifocal projection videos of beating hiPSC-CM clusters in 3D hydrogel. We further quantified cluster contractility response to temperature and adrenaline and observed changes to beating rate and relaxation. Challenges emerge from light penetration and overlaying textures in larger clusters. However, our findings indicate that MF-OPM is suitable for contractility studies of 3D clusters. Thus, for the first time, MF-OPM is used in CM studies and hiPSC-CM 3D cluster contraction is quantified in multiple orientations and imaging planes.

Substrate microtopographies induce cellular alignment and affect nuclear force transduction.

Cellular physiology has been mainly studied by using two-dimensional cell culture substrates which lack in vivo-mimicking extracellular environment and interactions. Thus, there is a growing need for more complex model systems in life sciences. Micro-engineered scaffolds have been proven to be a promising tool in understanding the role of physical cues in the co-regulation of cellular functions. These tools allow, for example, probing cell morphology and migration in response to changes in chemo-physical properties of their microenvironment. In order to understand how microtopographical features, what cells encounter in vivo, affect cytoskeletal organization and nuclear mechanics, we used direct laser writing via two-photon polymerization (TPP) to fabricate substrates which contain different surface microtopographies. By combining with advanced high-resolution spectral imaging, we describe how the constructed grid and vertical line microtopographies influence cellular alignment, nuclear morphology and mechanics. Specifically, we found that growing cells on grids larger than 10 × 20 µm2 and on vertical lines increased 3D actin cytoskeleton orientation along the walls of microtopographies and abolished basal actin stress fibers. In concert, the nuclei of these cells were also more aligned, elongated, deformed and less flattened, indicating changes in nuclear force transduction. Importantly, by using fluorescence lifetime imaging microscopy for measuring Förster resonance energy transfer for a genetically encoded nesprin-2 molecular tension sensor, we show that growing cells on these microtopographic substrates induce lower mechanical tension at the nuclear envelope. To conclude, here used substrate microtopographies modulated the cellular mechanics, and affected actin organization and nuclear force transduction.

Transcriptional Landscape of 3D vs. 2D Ovarian Cancer Cell Models.

Three-dimensional (3D) cancer models are revolutionising research, allowing for the recapitulation of an in vivo-like response through the use of an in vitro system, which is more complex and physiologically relevant than traditional monolayer cultures. Cancers such as ovarian (OvCa) are prone to developing resistance, are often lethal, and stand to benefit greatly from the enhanced modelling emulated by 3D cultures. However, the current models often fall short of the predicted response, where reproducibility is limited owing to the lack of standardised methodology and established protocols. This meta-analysis aims to assess the current scope of 3D OvCa models and the differences in the genetic profiles presented by a vast array of 3D cultures. An analysis of the literature (Pubmed.gov) spanning 2012-2022 was used to identify studies with paired data of 3D and 2D monolayer counterparts in addition to RNA sequencing and microarray data. From the data, 19 cell lines were found to show differential regulation in their gene expression profiles depending on the bio-scaffold (i.e., agarose, collagen, or Matrigel) compared to 2D cell cultures. The top genes differentially expressed in 2D vs. 3D included C3, CXCL1, 2, and 8, IL1B, SLP1, FN1, IL6, DDIT4, PI3, LAMC2, CCL20, MMP1, IFI27, CFB, and ANGPTL4. The top enriched gene sets for 2D vs. 3D included IFN-α and IFN-γ response, TNF-α signalling, IL-6-JAK-STAT3 signalling, angiogenesis, hedgehog signalling, apoptosis, epithelial-mesenchymal transition, hypoxia, and inflammatory response. Our transversal comparison of numerous scaffolds allowed us to highlight the variability that can be induced by these scaffolds in the transcriptional landscape and identify key genes and biological processes that are hallmarks of cancer cells grown in 3D cultures. Future studies are needed to identify which is the most appropriate in vitro/preclinical model to study tumour microenvironments.

Advances in optical biopsy--correlation of malignancy and cell density of primary brain tumors using Raman microspectroscopic imaging.

Raman spectroscopy is a promising tool towards biopsy under vision as it provides label-free image contrast based on intrinsic vibrational spectroscopic fingerprints of the specimen. The current study applied the spectral unmixing algorithm vertex component analysis (VCA) to probe cell density and cell nuclei in Raman images of primary brain tumor tissue sections. Six Raman images were collected at 785 nm excitation that consisted of 61 by 61 spectra at a step size of 2 micrometers. After data acquisition the samples were stained with hematoxylin and eosin for comparison. VCA abundance plots coincided well with histopathological findings. Raman spectra of high grade tumor cells were found to contain more intense spectral contributions of nucleic acids than those of low grade tumor cells. Similarly, VCA endmember signatures of Raman images from high grade gliomas showed increased nucleic acid bands. Further abundance plots and endmember spectra were assigned to tissue containing proteins and lipids, and cholesterol microcrystals. Since no sample preparation is required, an important advantage of the Raman imaging methodology is that all tissue components can be observed - even those that may be lost in sample staining steps. The results demonstrate how morphology and chemical composition obtained by Raman imaging correlate with histopathology and provide complementary, diagnostically relevant information at the cellular level.

Optical projection tomography as a quantitative tool for analysis of cell morphology and density in 3D hydrogels.

Assessing cell morphology and function, as well as biomaterial performance in cell cultures, is one of the key challenges in cell biology and tissue engineering (TE) research. In TE, there is an urgent need for methods to image actual three-dimensional (3D) cell cultures and access the living cells. This is difficult using established optical microscopy techniques such as wide-field or confocal microscopy. To address the problem, we have developed a new protocol using Optical Projection Tomography (OPT) to extract quantitative and qualitative measurements from hydrogel cell cultures. Using our tools, we demonstrated the method by analyzing cell response in three different hydrogel formulations in 3D with 1.5 mm diameter samples of: gellan gum (GG), gelatin functionalized gellan gum (gelatin-GG), and Geltrex. We investigated cell morphology, density, distribution, and viability in 3D living cells. Our results showed the usability of the method to quantify the cellular responses to biomaterial environment. We observed that an elongated morphology of cells, thus good material response, in gelatin-GG and Geltrex hydrogels compared with basic GG. Our results show that OPT has a sensitivity to assess in real 3D cultures the differences of cellular responses to the properties of biomaterials supporting the cells.

Gaussian Light Model in Brightfield Optical Projection Tomography.

This study focuses on improving the reconstruction process of the brightfield optical projection tomography (OPT). OPT is often described as the optical equivalent of X-ray computed tomography, but based on visible light. The detection optics used to collect light in OPT focus on a certain distance and induce blurring in those features out of focus. However, the conventionally used inverse Radon transform assumes an absolute focus throughout the propagation axis. In this study, we model the focusing properties of the detection by coupling Gaussian beam model (GBM) with the Radon transform. The GBM enables the construction of a projection operator that includes modeling of the blurring caused by the light beam. We also introduce the concept of a stretched GBM (SGBM) in which the Gaussian beam is scaled in order to avoid the modeling errors related to the determination of the focal plane. Furthermore, a thresholding approach is used to compress memory usage. We tested the GBM and SGBM approaches using simulated and experimental data in mono- and multifocal modes. When compared with the traditionally used filtered backprojection algorithm, the iteratively computed reconstructions, including the Gaussian models GBM and SGBM, provided smoother images with higher contrast.

A mathematical model and iterative inversion for fluorescent optical projection tomography.

Solving the fluorophore distribution in a tomographic setting has been difficult because of the lack of physically meaningful and computationally applicable propagation models. This study concentrates on the direct modelling of fluorescence signals in optical projection tomography (OPT), and on the corresponding inverse problem. The reconstruction problem is solved using emission projections corresponding to a series of rotational imaging positions of the sample. Similarly to the bright field OPT bearing resemblance with the transmission x-ray computed tomography, the fluorescent mode OPT is analogous to x-ray fluorescence tomography (XFCT). As an improved direct model for the fluorescent OPT, we derive a weighted Radon transform based on the XFCT literature. Moreover, we propose a simple and fast iteration scheme for the slice-wise reconstruction of the sample. The developed methods are applied in both numerical experiments and inversion of fluorescent OPT data from a zebrafish embryo. The results demonstrate the importance of propagation modelling and our analysis provides a flexible modelling framework for fluorescent OPT that can easily be modified to adapt to different imaging setups.

Bioactive glass ions induce efficient osteogenic differentiation of human adipose stem cells encapsulated in gellan gum and collagen type I hydrogels.

BACKGROUND: Due to unmet need for bone augmentation, our aim was to promote osteogenic differentiation of human adipose stem cells (hASCs) encapsulated in gellan gum (GG) or collagen type I (COL) hydrogels with bioactive glass (experimental glass 2-06 of composition [wt-%]: Na2O 12.1, K2O 14.0, CaO 19.8, P2O5 2.5, B2O3 1.6, SiO2 50.0) extract based osteogenic medium (BaG OM) for bone construct development. GG hydrogels were crosslinked with spermidine (GG-SPD) or BaG extract (GG-BaG). METHODS: Mechanical properties of cell-free GG-SPD, GG-BaG, and COL hydrogels were tested in osteogenic medium (OM) or BaG OM at 0, 14, and 21 d. Hydrogel embedded hASCs were cultured in OM or BaG OM for 3, 14, and 21 d, and analyzed for viability, cell number, osteogenic gene expression, osteocalcin production, and mineralization. Hydroxyapatite-stained GG-SPD samples were imaged with Optical Projection Tomography (OPT) and Selective Plane Illumination Microscopy (SPIM) in OM and BaG OM at 21 d. Furthermore, Raman spectroscopy was used to study the calcium phosphate (CaP) content of hASC-secreted ECM in GG-SPD, GG-BaG, and COL at 21 d in BaG OM. RESULTS: The results showed viable rounded cells in GG whereas hASCs were elongated in COL. Importantly, BaG OM induced significantly higher cell number and higher osteogenic gene expression in COL. In both hydrogels, BaG OM induced strong mineralization confirmed as CaP by Raman spectroscopy and significantly improved mechanical properties. GG-BaG hydrogels rescued hASC mineralization in OM. OPT and SPIM showed homogeneous 3D cell distribution with strong mineralization in BaG OM. Also, strong osteocalcin production was visible in COL. CONCLUSIONS: Overall, we showed efficacious osteogenesis of hASCs in 3D hydrogels with BaG OM with potential for bone-like grafts.

Mechanically Biomimetic Gelatin-Gellan Gum Hydrogels for 3D Culture of Beating Human Cardiomyocytes.

To promote the transition of cell cultures from 2D to 3D, hydrogels are needed to biomimic the extracellular matrix (ECM). One potential material for this purpose is gellan gum (GG), a biocompatible and mechanically tunable hydrogel. However, GG alone does not provide attachment sites for cells to thrive in 3D. One option for biofunctionalization is the introduction of gelatin, a derivative of the abundant ECM protein collagen. Unfortunately, gelatin lacks cross-linking moieties, making the production of self-standing hydrogels difficult under physiological conditions. Here, we explore the functionalization of GG with gelatin at biologically relevant concentrations using semiorthogonal, cytocompatible, and facile chemistry based on hydrazone reaction. These hydrogels exhibit mechanical behavior, especially elasticity, which resembles the cardiac tissue. The use of optical projection tomography for 3D cell microscopy demonstrates good cytocompatibility and elongation of human fibroblasts (WI-38). In addition, human-induced pluripotent stem cell-derived cardiomyocytes attach to the hydrogels and recover their spontaneous beating in 24 h culture. Beating is studied using in-house-built phase contrast video analysis software, and it is comparable with the beating of control cardiomyocytes under regular culture conditions. These hydrogels provide a promising platform to transition cardiac tissue engineering and disease modeling from 2D to 3D.