Open Access
Add to BibSonomy
Abstract
Optical sectioning has been widely employed for inhibiting out-of-focus backgrounds in three-dimensional (3D) imaging of biological samples. However, point scanning imaging or multiple acquisitions for wide-field optical sectioning in epi-illumination microscopy remains time-consuming for large-scale imaging. In this paper, we propose a single-scan optical sectioning method based on the hybrid illumination (HiLo) algorithm with a line-illumination strategy. Our method combines HiLo background inhibition with confocal slit detection. It thereby offers a higher optical sectioning capability than wide-field HiLo and line-confocal imaging without extra modulation and multiple data acquisition. To demonstrate the optical-sectioning capability of our system, we imaged a thin fluorescent plane and different fluorescence-labeled mouse tissue. Our method shows an excellent background inhibition in thick tissue and thus potentially provides an alternative tool for 3D imaging of large-scale biological tissue.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Biological tissues are commonly composed of cells and intercellular substances with three-dimensional (3D) submicron-size details. Optical microscopy is suitable for the observation of fine morphological structures in biological tissues owing to its subcellular resolution. Conventional wide-field microscopy provides the simplest and most direct means of observation. Nonetheless, epi-illumination-type lighting of both in-focus and out-of-focus signals generally introduces background noise blurring of the imaging. In other words, conventional wide-field microscopy lacks an optical sectioning capability. To overcome this disadvantage, traditional histology, such as hematoxylin and eosin (HE) staining, usually mechanically cuts the biological tissue into axially thin slices [1]. This method typically leads to the loss of axial information and the failure of 3D reconstruction, which is not conducive to 3D visualization of the biological tissue [2].
In view of the current problem, different principles [3,4] have been introduced to remove out-of-focus backgrounds and achieve optical sectioning. One of the most widely employed of these methods is confocal laser scanning microscopy for obtaining high-quality optically sectioned images [5]. A pinhole mask at an optical plane conjugate is employed to reject out-of-focus fluorescence signals. In general, the point-by-point scanning method thus adopted usually limits the confocal imaging speed on large samples. Although it has been effectively mitigated by spinning disk systems [6] or line confocal systems [7] to improve imaging throughput, these strategies engender a trade-off between the imaging speed and the imaging quality of optical sectioning.
Compared to the point-scanning imaging mode, a series of wide-field optical sectioning methods have been presented to speed up the imaging process. Structured illumination microscopy (SIM) [8] employs periodic modulation in the illumination beam to inhibit the out-of-focus background. The SIM optical-sectioning images are calculated from three modulated images with different phase patterns. Owing to the existence of tissue scattering, with the increase in propagation depth, the SIM modulation contrast deteriorates gradually and directly affects the optical sectioning effect of the SIM reconstruction [9]. Although SIM has a higher throughput advantage than the point confocal system, it sacrifices the optical-sectioning performance for thick-tissue imaging. Hybrid illumination (HiLo) microscopy [10], another wide-field optical-sectioning method, requires only two raw images with uniform and structural illumination, reducing data acquisition for optical-sectioning reconstruction. Its optical sectioning has been demonstrated in previous works [11–19]. It showed a faster imaging speed than SIM and confocal [4,12]. However, the contrast of modulation patterns in a wide-field fashion could be decreased by the strong scattering of thick tissue, which weakens the optical sectioning capability [20]. Recently, line-scanning SIM [21] and HiLo [22] have demonstrated an optical sectioning capability in unclearing thick tissue. However, both these methods are time-consuming processes for acquiring multiple images. Therefore, a fast optical-sectioning technique suitable for unclearing thick tissue remains a challenge.
In this paper, we report a single-scan HiLo method to acquire optically sectioned images of uncleared thick tissue based on a line-illumination strategy. The hybrid illumination without an additional modulator is produced by the line illumination in single scanning. We imaged a thin fluorescent plane mouse liver labeled with lectin, as well as a 100-µm-thick Thy1-green fluorescent protein (GFP) transgenic mouse brain slice to demonstrate the optical sectioning capability of our method. In addition, sequential imaging in the axial direction of tdTomato transgenic mouse kidney embedded with agarose further demonstrated the optical sectioning capability of thick tissue.
2. Materials and methods
2.1 Line-scanning imaging system
A schematic diagram of the line-scanning imaging system is shown in Fig. 1. Two continuous lasers (488 and 561 nm; Cobolt, Sweden), used according to experimental requirements, excited the fluorescence signal. The laser beams were first expanded by a 4F optical imaging system comprised of two lenses (L1, f = 10 mm; L2, f = 250 mm). The light was shaped by a cylindrical lens forming a linear beam for line illumination. The linear beam was finally relayed to the focal plane of the objective via lens L3 (f = 125 mm), a dichroic mirror (DM; ZT405/488/561rpc, Chroma, USA), and a water-immersion objective lens (OBJ; NA 1.0, XLUMPLFLN 20XW, Olympus, Japan). For detection, the fluorescence signal from the specimen was collected by the OBJ and filtered by the DM and emission filter (EM; ZET405/488/561 m, Chroma, USA). The fluorescence signal was focused by the tube lens (TL; f=180 mm) to a scientific CMOS camera (sCMOS, ORCA-Flash 4.0, Hamamatsu, Japan). The sCMOS possessed 2048 × 2048 pixels, and each pixel size was 6.5 × 6.5 µm2. The central lines of the camera are operated in sub-array mode as a multi-line array detector. Volumetric data of the sample were recorded using a high-accuracy 3D translation stage (X: ABL20020, Y: ANT130, Z: AVL125, Aerotech, USA).
Fig. 1. (a) Schematic of the line-scanning imaging system. L1-3, lens; C, cylindrical lens; DM, dichroic mirror; OBJ, objective lens; EM, emission filter; TL, tube lens. (b) Schematic of uniform illumination. (c) Schematic of structured illumination.
2.2 Principle of single-scan HiLo
The excitation light was shaped into a linear beam for line illumination, and the specimen was translated along the focused dimension perpendicular to the direction of the line illumination. The distance of each movement was one-pixel size. In the detection, the center lines of the camera were operated in the sub-array mode as a virtual slit. The virtual slit at the focal plane conjugate was used to partly reject out-of-focus light. The uniform illumination image, as shown in Fig. 1(b), was generated by the time-delay integration of the N line signals at the same position in the sample [23]. In other words, the accumulation of the convolution between the illumination point spread function and the Dirac function on the time series is equal to uniform illumination. The imaging results under uniform illumination Iu can be expressed as [7]
Although the low-frequency components originating from the defocused signal were filtered out from the focal plane, the low-frequency in-focus signal was also lost. To recover the low-frequency components in focus, the structured image was employed to extract the spatial contrast for weighting the in-focus portion. Besides the speckle and fringe, any non-uniform illumination pattern with spatial variations can also be used as a structural illumination for removing off-focus information [13]. We employed the Gaussian distribution of the line illumination along the scanning direction (X) in the focus as a naturally structural illumination without an additional modulator. This is the key to achieving the single-scan HiLo imaging in this study. As depicted in Fig. 1(c), the Gaussian-modulated image of the N-line sample is directly recorded by the detector. This is the structural image of the imaged N-line sample under the Gaussian-distribution illumination modulation for HiLo reconstruction. The imaging results under structured illumination ${I_s}$ can be expressed as [7]
The modulation occurs only within the focal plane, and the structured pattern decays rapidly during the defocusing process. The spatial contrast $C({v_x},{v_y})$ is the weighting function used to weight the in-focus portion of ${I_\textrm{u}}$, which is defined as [15]
Typical values of scaling factor $\eta $ are 10 ∼ 12 in our system.
3. Results and discussion
3.1 Measurement of the 3D spatial resolution
To measure the imaging performance of the single-scan HiLo system, we imaged 200-nm-diameter fluorescent beads (FluoSpheres Carboxylate-Modified Microspheres, ThermoFisher, USA) with a slit width of 6 pixels at an excitation wavelength of 488 nm. A certain diluted solution of fluorescent beads was mounted on the slide with glycerol. The lateral scanning step was 0.325 µm. We randomly chose five separate beads and calculated their full width at half maximum (FWHM) of the intensity profiles. After Gaussian curve fitting, the FWHMs in the X and Y directions were 0.44 ± 0.04 and 0.60 ± 0.07 µm, respectively [ Fig. 2(a)]. The pixel size in the focal plane was 0.325 µm, and the sampling process was undersampling. Therefore, the X resolution in the focused dimension was slightly larger than the theoretical value of 0.32 µm. The Y resolution was slightly larger than the X resolution because the line illumination was along the Y direction. We then acquired the images of the fluorescent beads with an axial step of 0.2 µm and calculated the intensity profiles using the same method. The FWHM along the Z direction was 2.09 ± 0.11 µm.
Fig. 2. 3D spatial resolutions of the single-scan HiLo system. (a) Line profile and Gaussian fitting through a bead in X and Y directions. (b) The line profile and Gaussian fitting through a bead in the Z direction.
3.2 Verification of the optical sectioning capability
To verify the feasibility of the single-scan HiLo with the line-illumination strategy, we imaged an agarose-embedded lectin-labeled mouse liver (DL-1174, Vectorlabs, USA). The tissue thickness was more than 1 cm. All animal experiments followed procedures approved by the Institutional Animal Ethics Committee of Huazhong University of Science and Technology.
Figure 3 provides a comparison of the optical sectioning capability between the uniform illumination and single-scan HiLo images. The uniform illumination image is detected by the virtual slit using a 6-pixel size, as shown in Fig. 3(a). Figure 3(b) shows a structured-illumination image modulated by Gaussian illumination with a modulation period of 6 pixels. The structural pattern with a high modulation frequency in Fig. 3(b) may not be easily visible owing to the display. Therefore, we further enlarge the yellow box in Fig. 3(b) as Fig. 3(b1) to show the distinct modulation contrast in the structural-illumination image, which guarantees the optical sectioning capability.
Fig. 3. Imaging the top surface of lectin-labeled mouse liver tissue that is thicker than 1 cm. (a) Uniform illumination image. (b) Structured-illumination image. (c) HiLo-reconstructed image. (a1–c1) Enlarged views of the areas indicated by yellow boxes in (a–c), respectively. (d) Normalized intensity profiles along yellow dash lines corresponding with Figs. (a1) and (c1). Scale bars in (a–c): 500 µm. Scale bars in (a1–c1): 30 µm.
In Fig. 3(c), the optical-sectioned image with the HiLo algorithm is calculated from Figs. 3(a) and 3(b). The areas of the yellow box in Figs. 3(a) and 3(c) are shown in the enlarged images of Figs. 3(a1) and 3(c1). The optically sectioned image in Fig. 3(c1) shows a higher imaging contrast than the line-confocal image in Fig. 3(a1). The intensity profiles along the yellow dashed lines corresponding to Figs. 3(a1) and 3(c1) are normalized and plotted in Fig. 3(d). The results illustrate that the single-scan HiLo with line-illumination has a lower background and higher contrast than the uniform illumination method.
To further quantify the effect of background removal, 20 × 20-pixel regions were obtained for each signal and background to calculate the signal-to-background ratio (SBR) [22]. The SBR of Figs. 3(a) and 3(c) are 7.52 dB and 20.18 dB, respectively. The above results demonstrate the feasibility of the optical sectioning of our method on thick tissues.
To quantify the optical sectioning ability of the system with different spatial frequencies, we measured the system optical sectioning capability by axially scanning a thin, uniform fluorescent plane with a step of 0.2 µm. The number of the central lines in the sub-array mode controlled the width of the virtual slit, which produced the period of the modulation patterns. A sectioning curve plots the integrated intensity as IHiLo a function of the plane defocus. The profile of the defocus function quantitatively evaluates the sectioning ability of the system. The profile of the defocus was calculated from the 20 × 20-pixel region of interest. Optical sectioning profiles of the different periods are plotted in Fig. 4. The FWHMs of the sectioning profile corresponding to 4, 6, and 8 pixels size at the imaging space are 1.8, 3.4, and 4.2 µm, respectively. The results demonstrate that the optical sectioning ability decreased with the increased width of the virtual slit. We also measured the optical sectioning of standard line confocal by single-line detection. The profile of the defocus measured by line confocal is shown in Fig. 4, and the FWHM of the sectioning profile is 5.6 µm. The results further demonstrate that optical sectioning ability of our method has an enhancement than standard line confocal.
Fig. 4. Optical sectioning performances with the different widths of the virtual slit and line confocal. The widths of the virtual slit are 4 (black), 6 (red), and 8 (blue) pixels, respectively. Line confocal (orange) is detected by one line.
We further compared the optical-sectioned capability in the thick tissue between the wide-field HiLo and single-scan HiLo systems, as shown in Fig. 5. The two systems loaded the same modulation period of 8 pixels and shared the same objective lens, tube lens, and camera. The 100-µm-thick Thy1-GFP transgenic mouse (Jackson Laboratory, Bar Harbor, ME, USA) brain slices were first imaged using a homemade wide-field HiLo microscope. The hybrid illumination was produced using a digital micro-mirror device (XD-ED01N, X-digit, China) to sequentially project the white and grid patterns. The wide-field image obtained by uniform illumination [ Fig. 5(a)] and its grid-structured modulated image reconstructed the corresponding wide-field HiLo image [Fig. 5(b)].
Fig. 5. (a) Wide-field image. (b) Wide-field HiLo image. (c) Single-scan HiLo image. (a1)–(c1) Enlarged views of the area I corresponding with (a)–(c). (a2)–(c2) Enlarged views of the area II corresponding with (a)–(c). (d)–(e) Normalized intensity profiles along yellow dash lines corresponding with Figs. (a1–c1) and (a2–c2), respectively. Scale bars in (a)–(c): 100 µm. Scale bars in (a1)–(c1) and (a2)–(c2): 20 µm.
Figure 5(c) shows a single-scan HiLo image acquired from our system. The stereoscopic sense successively decreases in Figs. 5(a), 5(b), and 5(c). The out-of-focus backgrounds in Figs. 5(b) and 5(c) are effectively removed after HiLo algorithm processing. The line-illumination strategy with the confocal slit detection maintains the spatial contrast of the structured illumination in the thick tissue [Fig. 5(c)], which further improves the optical-sectioned capability compared to the wide-field HiLo [Fig. 5(b)]. We further enlarge the two different areas indicated by the yellow and red boxes in Figs. 5(a)–5(c). The strong background signals surrounding the neuronal somas in the areas are rejected effectively by our method compared with the conventional methods. The normalized intensity profiles along the yellow dashed lines corresponding to Figs. 5(a1–a2), 5(b1–b2), and 5(c1–c2) are plotted in Figs. 5(d) and 5(e). The results illustrate that Figs. 5(c1–c2) have a lower background and higher contrast than Figs. 5(a1–a2) and 5(b1–b2). The SBR of Figs. 5(a)–5(c) are 4.62 dB, 18.34 dB, and 29.68 dB, respectively. The experimental results demonstrate that the single-scan HiLo with line illumination has a superior optical sectioning performance compared to wide-field HiLo in thick tissue.
To further demonstrate the optical-sectioned capability of our method, a 100-µm Thy1-GFP transgenic mouse brain slice with neuronal fiber was imaged at a virtual slit width of 6 pixels. The beam was focused from the surface to the inside of the brain slice for imaging. Sequential imaging in the axial direction of the brain slice was recorded with a step size of 2 µm. The volumetric data in Fig. 6 was present at a 20-µm depth. Figures 6(a)–6(c) show the optical-sectioned images at different depths generated by our method. Figure 6(d) shows the maximum intensity projection (MIP) of the data stacks of ten images spanning the 20 µm-depth range. The neural soma and their dendrites and axons are clearly discernible in the individual optical-sectioning images. Axons appear to be intermittent in these images; however, the MIP proves that their information acquisition in 3D space is continuous and complete. Enlarged views in the upper right corner of Figs. 6(a)–6(d) clearly display the neural soma at different depths. The experimental results suggest that the single-scan HiLo with line-illumination enables optical-sectioning imaging in thick tissue.
Fig. 6. 3D images of Thy1-GFP transgenic mouse brain tissue. (a)–(c) Optical-sectioned images at different depths. (d) Maximum intensity projection of stacks of ten images spanning a 20-µm depth range. Scale bars: 50 µm [(a)–(d)].
Variable scattering in different biological tissues may affect the optical sectioning performance of the system. To verify this effect of our system, we imaged the renal cortex of an mT/mG mouse (007676, Jackson Laboratory, Bar Harbor, ME, USA) at a virtual slit width of 6 pixels (Fig. 7). The thick kidney tissue (> 1 cm) was embedded with agarose, and the surface of the tissue was cut flat by our homemade vibratome [24]. An axial series of images were acquired by scanning with a step size of 2 µm at a depth of 30 µm, and the imaging range was 9.17 × 4.66 mm2. The exposure time of each image was 39 µs and the synchronous moving speed of the stage was approximately 500 mm/min. The scanning time of each strip was 1.10 s in 9.17 mm. Extra time included the acceleration and deceleration time of the stage, the field of view conversion time, and the recording time of the data, equal to 2.95 s. The total acquisition time of the seven strips was 28.35 s in each layer.
Fig. 7. Imaging of a mouse renal cortex. (a) MIP of a 15-image stack spanning a 30-µm depth range. Enlarged optical-sectioned images at different depths [(b)–(d)] and MIP (e) corresponding with the yellow dotted box of Fig. 7(a). Expanded views of the areas [(b1)–(e1)] bounded by the yellow dashed lines in (b–e), respectively. Scale bars: 1 mm (a), and 100 µm [(b)–(e)], and 20 µm [(b1)–(e1)].
Figure 7(a) shows the 30-µm-thick MIP of the kidney cortex with cell membrane-localized tdTomato. Fluorescence-labeled glomeruli are spread in the cortex, increasing the difficulty of the optical observation of this thick tissue. The enlarged views in Figs. 7(b)–7(d), indicated by yellow boxes in Fig. 7(a), display the glomerular tuft and renal tubule at different depths. Figure 7(e) is their MIP. To further illustrate the details, Figs. 7(b1)–7(e1) depict an enlarged glomerular tuft corresponding to the yellow dotted box in Figs. 7(b)–7(e). These figures show that the optical-sectioning ability of our method enables the acquisition of the spatial distribution of the glomerular tuft in 3D. Owing to strong tissue scattering in the unclear specimen, the contrast of Figs. 7(b)–7(d) gradually decreases with the increasing depth. This indicates that we may need to combine our approach with mechanical cutting [24] to further extend the imaging depth of the unclear tissue.
4. Conclusion
The above experimental results demonstrate that the proposed method can achieve optical sectioning imaging of thick tissue based on single-scan HiLo with the line-illumination strategy. Generally, previous approaches have often engendered a trade-off between the imaging speed and the optical-sectioned capability when imaging thick tissue. Raw data were obtained in the line-confocal detection in single-line scanning by our method. Unlike previous conventional HiLo approaches [10–19,22], the line-illumination strategy, in a single scan provides both a wide-field image and its hybrid-illumination image, thereby simplifying the data acquisition process. Phase-sensitive calculation of multiple images inevitably introduces motion artifacts of SIM reconstruction. In contrast, our method is insensitive to the modulation pattern benefiting from the HiLo algorithm. The virtual confocal slit of the line detector suppresses the fractional out-of-focus background and further improves the contrast of our method in thick tissue.
Recently, our group has just published another work of digital structured modulation (DSM) [24]. Compared with it, the optical sectioning capability of our work here is more flexibly due to being adjusted by both the width of the virtual slit [15,22] and the tuned parameter of the bandpass filter in HiLo algorithm [12]. While, the optical sectioning capability of DSM depends only on the former, similar with traditional wide-field SIM. The single-scan HiLo method compensates for the gap between the imaging speed and the capability of optical sectioning for large-scale imaging. This enables the potential to achieve the desired optical sectioning effect with the imaging throughput of the detector limit without the cost of extra data acquisition. This makes it possible for this method to be applied to live imaging and other scenes requiring a high imaging speed. Our method, combined with mechanical cutting [25,26], potentially provides a new technological means for 3D large-scale and high-resolution imaging.
Funding
National Key Research and Development Program of China (2017YFA0700402); National Natural Science Foundation of China (81827901, 91749209).
Disclosures
The authors declare no conflicts of interest.
References
1. E. S. Lein, M. J. Hawrylycz, N. Ao, M. Ayres, A. Bensinger, A. Bernard, A. F. Boe, M. S. Boguski, K. S. Brockway, and E. J. Byrnes, “Genome-wide atlas of gene expression in the adult mouse brain,” Nature 445(7124), 168–176 (2007). [CrossRef]
2. A. Li, H. Gong, B. Zhang, Q. Wang, C. Yan, J. Wu, Q. Liu, S. Zeng, and Q. Luo, “Micro-optical sectioning tomography to obtain a high-resolution atlas of the mouse brain,” Science 330(6009), 1404–1408 (2010). [CrossRef]
3. J.-A. Conchello and J. W. Lichtman, “Optical sectioning microscopy,” Nat. Methods 2(12), 920–931 (2005). [CrossRef]
4. J. Mertz, “Optical sectioning microscopy with planar or structured illumination,” Nat. Methods 8(10), 811–819 (2011). [CrossRef]
5. J. Pawley, Handbook of Biological Confocal Microscopy (Springer Science & Business Media, 2006), Vol. 236.
6. A. Egner, V. Andresen, and S. W. Hell, “Comparison of the axial resolution of practical Nipkow-disk confocal fluorescence microscopy with that of multifocal multiphoton microscopy: Theory and experiment,” J. Microsc. (Oxford, U. K.) 206(1), 24–32 (2002). [CrossRef]
7. T. Wilson and S. Hewlett, “Imaging in scanning microscopes with slit-shaped detectors,” J. Microsc. 160(2), 115–139 (1990). [CrossRef]
8. M. A. Neil, R. Juškaitis, and T. Wilson, “Method of obtaining optical sectioning by using structured light in a conventional microscope,” Opt. Lett. 22(24), 1905–1907 (1997). [CrossRef]
9. N. Hagen, L. Gao, and T. S. Tkaczyk, “Quantitative sectioning and noise analysis for structured illumination microscopy,” Opt. Express 20(1), 403–413 (2012). [CrossRef]
10. D. Lim, K. K. Chu, and J. Mertz, “Wide-field fluorescence sectioning with hybrid speckle and uniform-illumination microscopy,” Opt. Lett. 33(16), 1819–1821 (2008). [CrossRef]
11. T. N. Ford, D. Lim, and J. Mertz, “Fast optically sectioned fluorescence HiLo endomicroscopy,” J. Biomed. Opt. 17(2), 021105 (2012). [CrossRef]
12. D. Lim, T. N. Ford, K. K. Chu, and J. Mertz, “Optically sectioned in vivo imaging with speckle illumination HiLo microscopy,” J. Biomed. Opt. 16(1), 016014 (2011). [CrossRef]
13. J. Mertz and J. Kim, “Scanning light-sheet microscopy in the whole mouse brain with HiLo background rejection,” J. Biomed. Opt. 15(1), 016027 (2010). [CrossRef]
14. S. Santos, K. K. Chu, D. Lim, N. Bozinovic, T. N. Ford, C. Hourtoule, A. C. Bartoo, S. K. Singh, and J. Mertza, “Optically sectioned fluorescence endomicroscopy with hybrid-illumination imaging through a flexible fiber bundle,” J. Biomed. Opt. 14(3), 030502 (2009). [CrossRef]
15. C. Y. Lin, W. H. Lin, J. H. Chien, J. C. Tsai, and Y. Luo, “In vivo volumetric fluorescence sectioning microscopy with mechanical-scan-free hybrid illumination imaging,” Biomed. Opt. Express 7(10), 3968–3978 (2016). [CrossRef]
16. Y. Xue, K. P. Berry, J. R. Boivin, D. Wadduwage, E. Nedivi, and P. T. C. So, “Scattering reduction by structured light illumination in line-scanning temporal focusing microscopy,” Biomed. Opt. Express 9(11), 5654–5666 (2018). [CrossRef]
17. K. Philipp, A. Smolarski, N. Koukourakis, A. Fischer, M. Sturmer, U. Wallrabe, and J. W. Czarske, “Volumetric HiLo microscopy employing an electrically tunable lens,” Opt. Express 24(13), 15029–15041 (2016). [CrossRef]
18. W. Lin, D. Wang, Y. Meng, and S. C. Chen, “Multi-focus microscope with HiLo algorithm for fast 3-D fluorescent imaging,” PLoS One 14(9), e0222729 (2019). [CrossRef]
19. Q. Zhang, D. Pan, and N. Ji, “High-resolution in vivo optical-sectioning widefield microendoscopy,” Optica 7(10), 1287–1290 (2020). [CrossRef]
20. J. Mazzaferri, D. Kunik, J. M. Belisle, K. Singh, S. Lefrançois, and S. Costantino, “Analyzing speckle contrast for HiLo microscopy optimization,” Opt. Express 19(15), 14508–14517 (2011). [CrossRef]
21. O. Mandula, M. Kielhorn, K. Wicker, G. Krampert, I. Kleppe, and R. Heintzmann, “Line scan-structured illumination microscopy super-resolution imaging in thick fluorescent samples,” Opt. Express 20(22), 24167–24174 (2012). [CrossRef]
22. H. Zhang, K. Vyas, and G. Z. Yang, “Line scanning, fiber bundle fluorescence HiLo endomicroscopy with confocal slit detection,” J. Biomed. Opt. 24(11), 1 (2019). [CrossRef]
23. T. Yang, T. Zheng, Z. Shang, X. Wang, X. Lv, J. Yuan, and S. Zeng, “Rapid imaging of large tissues using high-resolution stage-scanning microscopy,” Biomed. Opt. Express 6(5), 1867–1875 (2015). [CrossRef]
24. Q. Zhong, C. Jiang, D. Zhang, S. Chen, R. Jin, H. Gong, and J. Yuan, “High-throughput optical sectioning via line-scanning imaging with digital illumination,” Opt. Lett. 46(3), 504–507 (2021). [CrossRef]
25. T. Jiang, B. Long, H. Gong, T. Xu, X. Li, Z. Duan, A. Li, L. Deng, Q. Zhong, X. Peng, and J. Yuan, “A platform for efficient identification of molecular phenotypes of brain-wide neural circuits,” Sci. Rep. 7(1), 1–12 (2017). [CrossRef]
26. H. Gong, D. Xu, J. Yuan, X. Li, C. Guo, J. Peng, Y. Li, L. A. Schwarz, A. Li, B. Hu, B. Xiong, Q. Sun, Y. Zhang, J. Liu, Q. Zhong, T. Xu, S. Zeng, and Q. Luo, “High-throughput dual-colour precision imaging for brain-wide connectome with cytoarchitectonic landmarks at the cellular level,” Nat. Commun. 7(1), 12142 (2016). [CrossRef]
