SemOpenAlex |

SemOpenAlex

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312692894> ?p ?o ?g. }

Showing items 1 to 49 of 49 with 100 items per page.

W4312692894 abstract "Article Figures and data Abstract Editor's evaluation Introduction Results Discussion Methods Data availability References Decision letter Author response Article and author information Metrics Abstract Rapid multicolor three-dimensional (3D) imaging for centimeter-scale specimens with subcellular resolution remains a challenging but captivating scientific pursuit. Here, we present a fast, cost-effective, and robust multicolor whole-organ 3D imaging method assisted with ultraviolet (UV) surface excitation and vibratomy-assisted sectioning, termed translational rapid ultraviolet-excited sectioning tomography (TRUST). With an inexpensive UV light-emitting diode (UV-LED) and a color camera, TRUST achieves widefield exogenous molecular-specific fluorescence and endogenous content-rich autofluorescence imaging simultaneously while preserving low system complexity and system cost. Formalin-fixed specimens are stained layer by layer along with serial mechanical sectioning to achieve automated 3D imaging with high staining uniformity and time efficiency. 3D models of all vital organs in wild-type C57BL/6 mice with the 3D structure of their internal components (e.g., vessel network, glomeruli, and nerve tracts) can be reconstructed after imaging with TRUST to demonstrate its fast, robust, and high-content multicolor 3D imaging capability. Moreover, its potential for developmental biology has also been validated by imaging entire mouse embryos (~2 days for the embryo at the embryonic day of 15). TRUST offers a fast and cost-effective approach for high-resolution whole-organ multicolor 3D imaging while relieving researchers from the heavy sample preparation workload. Editor's evaluation TRUST is a powerful content-rich three-dimensional imaging method. By combining iterative mechanical sectioning, automated labeling with fluorogenic dyes, UV-based surface illumination, and widefield multicolor detection, entire late gestation mouse embryos and terminally developed tissues can be imaged at cellular-scale resolution. Importantly, the strengths and weaknesses of the method are forthrightly presented. Ultimately, this work will be of great interest to developmental biologists, medical biologists, and clinicians, all of whom routinely work with specimens that are prohibitively large for labeling and imaging using classical mechanisms. https://doi.org/10.7554/eLife.81015.sa0 Decision letter Reviews on Sciety eLife's review process Introduction High-resolution three-dimensional (3D) whole-organ imaging with high fidelity has long been a scientific challenge. Mechanical sectioning followed by manual staining is a traditional approach for 3D histological imaging with almost no limitation on sample size Amunts et al., 2013. However, hundreds to thousands of slices must be mounted on glass slides, stained with dyes, and then imaged by a conventional bright-field microscope with a tile scanning to cover the whole slide. The entire procedure is highly labor-intensive and requires complicated image registration algorithms Wang et al., 2015; Ju et al., 2006 for 3D reconstruction. Moreover, sectioning before imaging could induce inevitable distortion in the reconstructed 3D image due to slice ruptures, which is impossible to be corrected by image processing approaches, hindering the broad applicability of this traditional method. Over the past decade, many advanced tissue clearing protocols Richardson and Lichtman, 2015; Richardson et al., 2021; Ueda et al., 2020 have been proposed to render opaque biological samples transparent and significantly reduce light scattering or absorption, thus facilitating volumetric imaging of thick tissue samples when combined with light microscopy approaches. Light-sheet fluorescence microscopy (LSFM) Huisken et al., 2004 is ideal for imaging large cleared specimens with its high acquisition speed, high spatial resolution, and high light efficiency. For example, single-cell resolution or subcellular-resolution imaging over the entire mouse brain has been achieved within hours Matsumoto et al., 2019 or days Murakami et al., 2018, respectively. Many issues concerning tissue clearing have been resolved or mitigated recently. For example, the toxicity of solvents, particularly benzyl alcohol and benzyl benzoate, can be addressed by using its alternatives, like ethyl cinnamate Henning et al., 2019, which also contributes to prolonged fluorescent protein emission Klingberg et al., 2017. Also, the active electrophoretic tissue clearing (ETC) device has been utilized in CLARITY Chung et al., 2013 to accelerate the clearing process significantly. Perfusion-assisted agent release in situ Yang et al., 2014 achieved much faster whole-body clearing without ETC, which may cause tissue degradation. In response to the potential tissue damage and molecular information loss, stabilization under harsh conditions via intramolecular epoxide linkages to prevent degradation Park et al., 2018 was proposed using a flexible polyepoxide to preserve the tissue architecture in organ-scale transparent tissues. Although both tissue clearing and LSFM are still under active development, barriers remain despite the significant accomplishment. For instance, the detective objective lens with a long working distance is generally required by LSFM when imaging large-volume samples, limiting the highest attainable resolution due to the consequent reduction in numerical aperture (NA). Besides, tissue clearing, especially for large samples, still need to balance the clearing effect and time cost to avoid tissue degradation. For example, aqueous techniques show relatively high biocompatibility, biosafety, and preservation of protein function, which, however, may take a longer time. Therefore, despite the destructive nature of the imaging process, block-face serial sectioning tomography has also gained much attention for automated and registration-free 3D imaging of large specimens. In this imaging and sectioning scheme, tissue clearing is not needed. More importantly, the imaging volume is relatively easy to be scaled up through engineering modification of the sectioning system, e.g., enlarging the sample holder. By far, spinning disk-based confocal microscopy Seiriki et al., 2017; Seiriki et al., 2019, two-photon microscopy Economo et al., 2016; Ragan et al., 2012, multiphoton microscopy Abdeladim et al., 2019, microscopy with ultraviolet surface excitation (MUSE) Fereidouni et al., 2017; Guo et al., 2019, optical coherence microscopy Min et al., 2020, and photoacoustic microscopy (PAM) Wong et al., 2017 have been utilized for block-face serial sectioning tomography. However, there are other challenges in this field, including high system cost, time-consuming tissue processing (e.g., staining or embedding), and the slow imaging speed with point-scanning-based imaging systems. To this end, wide-field large-volume tomography (WVT) Gong et al., 2016 proposed real-time labeling to reduce the time cost of sample staining significantly. However, the imaged sample must be embedded into a resin block after several days of processing to achieve fine sectioning by a microtome, which can also induce apparent tissue shrinkage due to dehydration Rodgers et al., 2021; Gong et al., 2013 and is detrimental in high-resolution whole-organ imaging. By far, numerous organ-scale imaging systems have been proposed with different focuses and strengths. With the goal of time-efficient, inexpensive, and multicolor large-volume 3D imaging, we developed translational rapid ultraviolet-excited sectioning tomography (TRUST) with the merits of both MUSE and WVT while bypassing the need for additional sample processing and keeping the entire system setup simple and cost-effective. More specifically, the short penetration depth of the UV light is utilized in TRUST for fast widefield block-face imaging. Meanwhile, formalin-fixed samples can be directly imaged by TRUST, and the tissue will be labeled layer by layer along with vibratomy-assisted serial sectioning. Compared with whole-organ pre-staining, real-time staining can tremendously reduce the overall staining time, considering that the diffusion timescale correlates quadratically with the tissue thickness Richardson et al., 2021. Furthermore, the real-time staining is well fitted with the shallow penetration depth of the UV light. While TRUST does achieve rapid 3D imaging with subcellular resolution, its unique strength lies in its high-content imaging ability with a cost-effective and straightforward setup. Oblique illumination using an inexpensive ultraviolet light-emitting diode (UV-LED) is implemented as the light source to simultaneously excite molecular-specific fluorescence and autofluorescence signals, which can then be collectively captured by a color camera. Many fluorescent dyes with different emission spectra can be excited simultaneously by the UV light Fereidouni et al., 2017, providing a broad and informative color palette. In TRUST, we mainly applied two fluorogenic dyes Gao et al., 2020 (4',6-diamidino-2-phenylindole [DAPI] and propidium iodide [PI]) together for double labeling to provide high color contrast and reveal rich biological information (Figure 1—figure supplement 1). Nile red and Dil have also been applied for lipid staining in TRUST (Figure 1—figure supplement 2). To characterize the performance and examine the robustness of the proposed system, all vital organs in mice were imaged. To further show the full potential of TRUST, whole mouse embryos at different stages have also been imaged. The detailed image comparison demonstrates that TRUST is highly desirable for large sample imaging, showing great promise as a tool for developmental biology studies. Results TRUST system setup and workflow In TRUST (Figure 1a–c), the short penetration depth in tissue of obliquely illuminated light from UV-LED (~285 nm) is the key to achieving widefield block-face imaging. Both fluorescence and autofluorescence signals from the tissue surface are excited by UV light, which are subsequently collected by a 10×infinity-corrected objective lens, and finally focused on a color camera by a tube lens. Two motorized stages (Motor-x and Motor-y) can drive the objective lens scanning along the x-y plane, and the manually tunable z-axis stage is used for focusing. A lab-built 2-axis angle adjustable platform (Figure 1—figure supplement 3) under the vibratome can keep the focal plane of the objective lens in parallel with the sample surface determined by the blade angle of the vibratome. Figure 1 with 9 supplements see all Download asset Open asset Overview of 3D whole-organ imaging by TRUST. (a) Schematic of the TRUST system. Light from UV-LED is obliquely projected onto the surface of an agarose-/gelatin-embedded sample (e.g., a mouse brain), which is placed on top of a sample holder inside a water tank of a vibratome filled with staining solutions. The generated fluorescence and autofluorescence signals are collected by an objective lens, refocused by an infinitely corrected tube lens, and finally detected by a color complementary metal-oxide-semiconductor camera. (b), Close-up of the region marked by orange dashed box in a, showing the components immersed in the staining solutions. (c), Viewing b from the x-z plane. A waterproof case containing two pieces of quartz keeps most of the space under the objective lens filled with air. (d), Workflow of the whole imaging process, including (1) chemical staining for ~150 s, (2) widefield imaging with raster-scanning and stitching in parallel, and (3) shaving off the imaged layer with the vibratome to expose a layer underneath. The three steps will be repeated until the entire organ has been imaged. All procedures are automated with lab-built hardware and control programs. Fixed tissue block after agarose or gelatin embedding can be directly imaged with TRUST, and the sample will be stained during imaging by submerging it under staining solutions in the water tank of the vibratome. A 3D printed plastic waterproof case (Figure 1c) can reduce the fluorescence background by minimizing the volume of staining solutions filled between the tissue surface and objective lens (Figure 1—figure supplement 4). Besides, it can prevent the objective lens from being affected by liquid evaporation or fluctuation, ensuring high-quality imaging throughout the entire whole-organ imaging process. Two pieces of quartz mounted on the waterproof case are used to transmit the UV illumination beam and the excited fluorescence signal. The workflow of the whole system (Figure 1d) can be simplified as a loop of three steps: (1) the surface layer of the sample immersed under chemical dyes will be labeled within ~150 s Figure 1—figure supplement 5; (2) the region of interest of the current layer, which consists of multiple fields of view (FOVs), will be acquired through motorized raster-scanning and subsequently stitched in parallel by a lab-built program; and (3) the vibratome will cut off the imaged layer and expose the layer underneath to the staining solutions. This loop will end when the whole organ has been imaged completely. To realize fully automated serial imaging, triggering circuits and corresponding control programs have been developed for synchronizing the entire TRUST system. Because the sample is stained during the imaging step in TRUST, the labeling protocol can be thought of as real-time staining. Fluorogenic probes Gao et al., 2020; Wang et al., 2020; Werther et al., 2021 which show an increase in fluorescence on binding to their targets, do not require a procedure to remove unbound probes and are naturally suited for the real-time staining owing to the low fluorescence background. Although lots of different fluorogenic probes have been synthesized for labeling various components, like proteins Gao et al., 2020; Werther et al., 2021; Grimm et al., 2017, lipid droplets Fam et al., 2018, cytoskeleton Lukinavičius et al., 2014, and mitochondrial Fang et al., 2020, two of the most commonly used and commercially available fluorogenic probes, DAPI and PI, were applied together in TRUST for nucleic acids staining. The fluorescence intensity of DAPI or PI will be amplified over 20 folds when bonded with the nucleic acids Barcellona et al., 1990; Unal Cevik and Dalkara, 2003, so that the fluorescence background from the staining solutions will be negligible. Finally, to maintain the concentration of chemical dyes in the water tank over a long period for whole-organ imaging, a high-precision water pump (LabN1-YZ1515x, Baoding Shenchen Precision Pump Co., Ltd) can be utilized to supply additional dyes into the water tank at a certain rate (Supplementary file 1a). Another peristaltic pump (KCP PRO2-N16, Kamoer Fluid Tech Co., Ltd.) with its suction pipe placed close to the sample holder of the vibratome can be utilized to collect the sectioned layers automatically for follow-up studies, like immunostaining, if necessary (Figure 1—video 1). Whole mouse brain imaging with TRUST Our TRUST system first imaged a mouse brain to demonstrate its high imaging speed, molecular-specific real-time staining, and multicolor/multicontrast imaging capability. The mouse brain was first harvested and fixed in formalin for 24 hr. Subsequently, without staining or clearing, the mouse brain was directly embedded into 2% w/v agarose. The sectioning thickness of the vibratome was set as 50 µm to provide a good balance between the sectioning quality and the z-axis sampling interval. DAPI and PI solutions with a concentration of 5 µg/ml were used to label cell nuclei. The total imaging volume (12.1 mm × 8.6 mm×17.4 mm [xyz]) consists of approximately 7.8×1011 voxels with 24 bits RGB channels, of which the uncompressed dataset is ~2.1 terabytes (TB). Without image registration, images of 347 coronal sections acquired by TRUST can be directly stacked to reconstruct the 3D model (Figure 2a). Including staining, two-dimensional (2D) raster-scanning, and mechanical sectioning, the total acquisition time is ~64 hr, which is highly manageable and practical. Figure 2 with 7 supplements see all Download asset Open asset Whole mouse brain imaging with TRUST. (a, b), Side and top views of the reconstructed 3D model of a fixed mouse brain, respectively. (c–f) Four coronal sections with positions indicated in b by the white dashed lines. (g), (h), Close-up images of the blue solid and green dashed regions marked in c. (i–k), Close-up images of the red dotted, yellow dashed, and purple solid regions marked in d. (l–n), Close-up images of the orange dotted, green solid, and blue dashed regions marked in e.( o), (p), Close-up images of the red solid and black dashed regions marked in f. (q), The cell nuclei in the mouse brain extracted from the red channel in a. (r), ( s), Close-up images of the blue solid and orange dashed regions marked in q. (t), 3D structure of the nerve tracts or fibers extracted from the green channel of the autofluorescence signal in a. (u), (v), Close-up images of the red dashed and black solid regions marked in t. (w), Vessel network extracted from another mouse brain and rendered with different colors corresponding to different coronal layers. (x), Close-up image of the orange dashed region marked in w. MCL: mitral cell layer; IPL: inner plexiform layer; EPL: external plexiform layer. Scale bars: 1 mm (a–f, q,t,w), 200 µm (x), 100 µm (r–v), and 50 µm (g–p). The real-time staining is well fitted with TRUST. Apart from the high time cost, traditional staining protocols Seiriki et al., 2019 may also lead to uneven staining (Figure 2—figure supplement 1) due to the nature of passive diffusion. More importantly, agents like Triton X-100, which is used for increasing the permeability of tissue and accelerating the staining speed, can increase the transparency of the sample by washing away the lipids inside the tissue. Then, the imaging contrast of TRUST will be deteriorated due to the increased penetration depth of UV light (Figure 2—figure supplement 1b–g). In comparison, four coronal sections (Figure 2c–f) with positions indicated by the white dashed lines in Figure 2b show the stable sectioning performance of vibratome and uniform staining throughout the whole-brain imaging process. When compared with other advanced 3D imaging systems, the advantages of TRUST are more related to its high-content multicolor imaging capability with an outstanding balance in terms of time-efficiency (sample preparation time and imaging speed), imaging resolution, and cost-effectiveness (Supplementary file 1b), by enjoying the benefits of the UV surface excitation and double labeling. Unlike the staining protocol used in MUSE or WVT, we developed a real-time double-labeling protocol that perfectly fits our TRUST system. The double labeling with DAPI and PI helps to reveal more biological information and achieve better image quality than that of staining with only PI (Figure 1—figure supplement 1). First, PI staining can clearly reveal cytomorphological details of neurons while the cell bodies of glial cells are only slightly labeled Hezel et al., 2012, hence differentiating neurons from glial cells in the brain through morphological differences (Figure 1—figure supplement 1a). In contrast, DAPI almost only stains cell nuclei. As a result, with the combination of DAPI and PI, TRUST can provide similar imaging contrast as Nissl stains (Figure 2—figure supplement 2), where the Purkinje cell layer can be easily differentiated from the granular layer or molecular layer. Furthermore, the color contrast of images can be improved by double labeling, especially for regions with a high density of cells (Figure 1—figure supplement 1g and h) because the cytoplasm stained by PI (red) can act as a background for DAPI-labeled cell nuclei to stand out (green and blue). Although intrinsic autofluorescence is unwanted information in many cases, which can be regarded as background signals, much effort has been spent to reveal meaningful biological information in 2D label-free imaging Ojaghi et al., 2020; Bhartia et al., 2010; Kaza et al., 2021; Yung et al., 2016; Costantini et al., 2021, which can also be utilized in TRUST. For example, the axon (Figure 2g), nerve tracts (e.g., lateral olfactory tract Figure 2h or anterior commissure Figure 2i), and myelinated fiber bundles in the caudate putamen (Figure 2j) have been identified with TRUST even without any labeling. Also, blood vessels (Figure 2k) can be identified based on negative contrast because their autofluorescence signal intensity is lower than that of the surrounding tissues Mehrvar et al., 2021; Staniszewski et al., 2013. To further enhance the contrast of the vessel network in the brain (Figure 2w and x), the mouse was perfused transcardially with a mixture of black ink and 3% w/v gelatin (Figure 2—figure supplement 3). In TRUST, the mixed dyes with a broad emission spectrum (blue to red) make the fluorescence signals less affected by the autofluorescence background by extracting signals from different color channels for different types of tissue components. For example, although hepatocytes in liver tissue are hard to be differentiated from the autofluorescence background in the red channel, the image contrast is high in the green and blue channels (Figure 2—figure supplement 4a–d). Also, although the autofluorescence background is strong in the green and blue channels for some regions in the mouse brain, the cells remain evident in the red channel (Figure 2—figure supplement 4e–h). With background subtraction and dynamic range adjustment in different color channels of TRUST images, the 3D distribution of cell nuclei (Figure 2—figure supplement 1q-s) and nerve fiber bundles (Figure 2t–v) in the brain can be digitally extracted. The 3D animation of the whole mouse brain, including serial coronal or sagittal sections, has been rendered as shown in Figure 2—video 1. The vessel network has also been rendered, as shown in Figure 2—video 2. 3D imaging of other organs with TRUST To demonstrate the generalizability and robustness of the TRUST system, other mouse organs with various sizes are imaged, including a heart (Figure 3a–e), liver (Figure 3f–j), kidney (Figure 3k–o), lung (Figure 3p–t), and spleen (Figure 3u–y). A fixed mouse heart was first imaged by TRUST with a sectioning thickness of 50 µm. The whole imaging and staining automated procedure for the entire volume (10.4 mm × 8.2 mm × 6.1 mm, 1.8×109 voxels with 24 bits RGB channels, 122 sections) took ~21 hr, and the reconstructed 3D model is shown in Figure 3a. One representative section is shown in Figure 3b with its position marked by a white dashed line at the top right-hand corner. Three close-up images (Figure 3c–e) indicate that not only the cell nucleus, but other components, like adipose tissue (Figure 3c) or cardiac muscle (Figure 3d), can also be clearly imaged. More details of the whole dataset, including multiple zoomed-in regions, can be found in the rendered 3D animation (Figure 3—video 1). Figure 3 with 10 supplements see all Download asset Open asset 2D/3D image gallery of other organs in mice with TRUST. (a), Reconstructed 3D model of the whole mouse heart. (b), One section of the heart with the position indicated by the white dashed line at the top right-hand corner. (c–e), Three close-up images of the orange dashed, white solid, and red dotted regions marked in b. (f), One block of a fixed mouse liver. (g), One section of the liver with the position indicated by the curved white dashed line marked in f. (h), (i), Close-up images of the white solid and yellow dashed regions marked in g. (j), Vessel network extracted based on negative contrast and rendered with pseudo color. (k), Reconstructed 3D model of a whole mouse kidney. (l), One section in the middle of the kidney with the position indicated by the white dashed line at the top right-hand corner. (m), (n), Close-up images of the red solid and white dashed regions marked in l. (o), Vessel network and glomeruli extracted from the whole kidney and rendered with pseudo color. (p), Part of a mouse lung imaged with TRUST. The curved white dashed line indicates the position of the section in q.( r), (s), Close-up images of the orange solid and white dashed regions marked in q. (t), A small zoomed-in region in p.( u), Rendered 3D model of a mouse spleen, and the curved white dashed line indicates the position of the section in w. (v), White pulp regions extracted from u through the blue channel and rendered with pseudo color. (x), (y), Close-up images of the white dashed and white solid regions marked in w. Scale bars: 1 mm (b,g,l,q,w) and 50 µm (c–e, h), (i), (m), (n), (r), (s), (x), (y). Part of a fixed mouse liver (Figure 3f) has also been imaged by TRUST with a sectioning thickness of 50 µm. The entire imaging and staining process for the total volume (8.8 mm × 12.5 mm × 3 mm, 1.4×1011 voxels with 24 bits RGB channels, 60 sections) took ~11 hr. The curved white dashed line marked in Figure 3f indicates the position of the section in Figure 3g. Two close-up images (Figure 3h and i) of the white solid and yellow dashed regions in Figure 3g show that hepatocytes and typical anatomical structures, like portal vein, sinusoids, and focal inflammation, can be well differentiated with TRUST. Based on the negative contrast of blood vessels, as shown in Figure 3i, the vessel network (Figure 3j) of the whole dataset can also be extracted with several image processing steps (Figure 3—figure supplement 1). Besides, by extracting features through different color channels and subsequent image binarization, cytoplasm and the cell nucleus can be separated to estimate the nuclear/cytoplasmic ratio (Figure 3—figure supplement 2). 3D animation of the whole dataset has been rendered as shown in Figure 3—video 2. Moreover, another block of the liver with a volume of 500 µm × 500 µm × 250 µm (Figure 3—figure supplement 3) has also been imaged by TRUST with a sectioning thickness of 10 µm, which is also matched with the penetration depth of UV light (Figure 3—figure supplement 4). A finer sectioning thickness results in a decrease in the z-sampling interval, which is helpful for resolving microstructures, such as the sinusoids in 3D space. Imaging for the whole mouse kidney (Figure 3k) took~19hr and the sectioning thickness of the volume (13.2 mm × 8.6 mm×4.7 mm, 2.3×1011 voxels with 24 bits RGB channels, 93 sections) was also set as 50µm. One typical section is shown in Figure 3l with its position marked by the white dashed line at the top right-hand corner. Vein and glomerulus can be clearly recognized as shown in Figure 3m and n, respectively. All glomeruli in the kidney were extracted with the assistance of an open-source state-of-the-art detection and segmentation machine learning library, Detectron2 Wu et al., 2019. The reconstructed vessel network together with the glomeruli of the whole kidney is shown in Figure 3—video 1 of the whole dataset has been rendered as shown in Figure 3—video 3. Sectioning lung tissue directly with a vibratome can be problematic due to porous structures, e.g., alveoli or bronchioles (Figure 3—figure supplement 5a and b). To achieve better sectioning performance, we first filled the lung with 10%w/v melted gelatin by syringe injection through its trachea, and the processed lung sample should be cooled down quickly and moved into formalin solution at 4° C overnight for post-fixation (Figure 3—figure supplement 5c). The entire imaging and staining process for the total volume (15.4 mm × 4.7 mm × 4.6 mm, 1.4×1011 voxels with 24 bits RGB channels, 91 sections) took~11hr with a sectioning thickness of 50µm, and its reconstructed 3D model is shown in Figure 3p. To exhibit the improved sectioning performance, all sections of the whole dataset including multiple zoomed-in regions have been rendered in the 3D animation video (Figure 3—video 4). Two close-up images (Figure 3r and s) of the yellow solid and white dashed regions marked in Figure 3q show that the common anatomical structures, including artery, bronchiole, and alveolus, can be well imaged. The final whole organ that we imaged was a mouse spleen. The entire imaging and staining process for the total volume (12.7 mm × 8.2 mm×3.8 mm, 1.7×1011 voxels with 24 bits RGB channels, 75 sections) of the mouse spleen embedded in gelatin block took ~14 hr. The reconstructed 3D model is shown in Figure 3u, and the curved white dashed line indicates the position of the section shown in Figure 3w. Two close-up images (Figure 3x and y) of the white dashed and white solid regions marked in Figure 3w correspond to two major components in the spleen: white pulp and red pulp, respectively. Typical structures, like germinal center or marginal zone, can be clearly differentiated. As the overall appearance of the white pulp regions (Figure 3x) look blue, as shown in Figure 3v, they can be extracted from the blue channel with simple thresholding. Finally, 3D animation of the whole dataset also has been rendered, as shown in Figure 3—video 5. 2.4. 3D imaging of whole mouse embryos with TRUST Under embryonic development, there are dramatic changes in cell/structural morphology and arrangement. Therefore, 3D imaging of a whole mouse embryo is vitally essential to assist biologists in understanding any anatomical and functional changes. However, imaging the whole embryo is still a scientific challenge and suffers from issues like low imaging resolution (micro-CT Wong et al., 2012 and optical projection tomography Ban et al., 2019) or a limited number of sections (histological imaging Crawford et al., 2010; Chen et al., 2017). In the above, we have already demonstrated the advantages" @default.
W4312692894 created "2023-01-05" @default.
W4312692894 date "2022-09-04" @default.
W4312692894 modified "2023-09-27" @default.
W4312692894 title "Decision letter: Translational rapid ultraviolet-excited sectioning tomography for whole-organ multicolor imaging with real-time molecular staining" @default.
W4312692894 doi "https://doi.org/10.7554/elife.81015.sa1" @default.
W4312692894 hasPublicationYear "2022" @default.
W4312692894 type Work @default.
W4312692894 citedByCount "0" @default.
W4312692894 crossrefType "peer-review" @default.
W4312692894 hasBestOaLocation W43126928941 @default.
W4312692894 hasConcept C120665830 @default.
W4312692894 hasConcept C121332964 @default.
W4312692894 hasConcept C136339569 @default.
W4312692894 hasConcept C142724271 @default.
W4312692894 hasConcept C150903083 @default.
W4312692894 hasConcept C163716698 @default.
W4312692894 hasConcept C185592680 @default.
W4312692894 hasConcept C207001950 @default.
W4312692894 hasConcept C71924100 @default.
W4312692894 hasConcept C74864618 @default.
W4312692894 hasConcept C86803240 @default.
W4312692894 hasConceptScore W4312692894C120665830 @default.
W4312692894 hasConceptScore W4312692894C121332964 @default.
W4312692894 hasConceptScore W4312692894C136339569 @default.
W4312692894 hasConceptScore W4312692894C142724271 @default.
W4312692894 hasConceptScore W4312692894C150903083 @default.
W4312692894 hasConceptScore W4312692894C163716698 @default.
W4312692894 hasConceptScore W4312692894C185592680 @default.
W4312692894 hasConceptScore W4312692894C207001950 @default.
W4312692894 hasConceptScore W4312692894C71924100 @default.
W4312692894 hasConceptScore W4312692894C74864618 @default.
W4312692894 hasConceptScore W4312692894C86803240 @default.
W4312692894 hasLocation W43126928941 @default.
W4312692894 hasOpenAccess W4312692894 @default.
W4312692894 hasPrimaryLocation W43126928941 @default.
W4312692894 hasRelatedWork W159416972 @default.
W4312692894 hasRelatedWork W1971324172 @default.
W4312692894 hasRelatedWork W2016025723 @default.
W4312692894 hasRelatedWork W2042848525 @default.
W4312692894 hasRelatedWork W2045354512 @default.
W4312692894 hasRelatedWork W2054128329 @default.
W4312692894 hasRelatedWork W2084656721 @default.
W4312692894 hasRelatedWork W2372363491 @default.
W4312692894 hasRelatedWork W2794902547 @default.
W4312692894 hasRelatedWork W2987044308 @default.
W4312692894 isParatext "false" @default.
W4312692894 isRetracted "false" @default.
W4312692894 workType "peer-review" @default.