Linear scanning type X-ray plate imaging microscope system

Abstract

The invention discloses a line scanning type X-ray plate imaging microscope system, and the system comprises a sample chamber which is used for placing a sample; the illumination unit is used for providing exciting light to excite the sample to generate fluorescent light; the light sheet generating unit is used for shaping the exciting light to form a light sheet; an X-ray plate rotating unit for changing the direction of the X-ray plate; the detection unit is used for collecting fluorescence signals generated when the light plates in different directions excite the sample, so that image stacks corresponding to the light plates in different directions are obtained, and different image stacks are fused and reconstructed to obtain a three-dimensional image; wherein the fluorescence signal acquisition direction of the detection unit is orthogonal to the incidence direction of the exciting light and is orthogonal to the focal plane of the sample. According to the invention, the image stack is obtained from multiple angles by rotating the light sheet, and the image stack is reconstructed into the three-dimensional image with higher z-axis resolution, so that the requirements of high-quality imaging and rapid image acquisition can be met at the same time.

Description

Technical Field

[0001] This application relates to the field of optical technology, and in particular to a line scan light sheet imaging microscope system. Background Technology

[0002] Light sheet imaging, as an advanced microscopic imaging technique, has important applications in biomedical research. Its core principle is to illuminate the sample using a thin light sheet, and to achieve three-dimensional imaging by orthogonalizing the illumination and detection light paths.

[0003] However, existing light-sheet imaging techniques have significant limitations in image acquisition. In light-sheet microscopy, there is always a trade-off between axial resolution and field of view (FOV). Light sheets with a larger FOV are typically thicker, which is more advantageous for illuminating large sample areas. A thicker light sheet widens the point spread function along the Z-axis (which is usually the direction perpendicular to the light sheet, i.e., the depth direction), resulting in the collection of more background noise and a decrease in axial resolution. If a thinner light sheet is used, the sample is scanned by a small illumination area, increasing acquisition time. To obtain high-quality 3D images, it is often necessary to sacrifice the field of view size to ensure uniform light sheet thickness, thereby achieving high-resolution imaging. This leads to a significant increase in image acquisition time, making it difficult to meet the demand for rapid sample imaging. This problem causes many inconveniences in practical applications. For example, when imaging live samples for extended periods, the physiological state of the sample may change, affecting the accuracy and reliability of the imaging results. Furthermore, prolonged imaging processes increase the risk of photobleaching and phototoxicity, causing damage to the sample.

[0004] Therefore, there is an urgent need for a line-scan light-sheet imaging microscope system that can simultaneously meet the requirements of high-quality imaging and rapid image acquisition. Summary of the Invention

[0005] To address the above technical issues, this application provides a line-scan light-sheet imaging microscope system that obtains image stacks from multiple angles by rotating the light sheet and reconstructs them into a three-dimensional image with higher z-axis resolution, thereby simultaneously meeting the requirements for high-quality imaging and rapid image acquisition.

[0006] This application provides a line-scanning light-sheet imaging microscope system, comprising: The sample chamber is used to hold samples; An illumination unit is used to provide excitation light to excite the sample to produce fluorescence, wherein the optical axis of the excitation light is perpendicular to the sample. A light sheet generation unit is used to shape the excitation light to form a light sheet; A light sheet rotation unit is used to change the direction of the light sheet; The detection unit is used to collect fluorescence signals generated by the sample excited by the light sheet from different directions, thereby obtaining image stacks corresponding to the light sheet from different directions, and fusing and reconstructing different image stacks to obtain a three-dimensional image; wherein the direction in which the detection unit collects fluorescence signals is orthogonal to the incident direction of the excitation light and orthogonal to the focal plane of the sample.

[0007] As an improvement to the above solution, the optical sheet generation unit includes: A Powell lens is used to convert the light beam emitted by the illumination unit into a uniform linear light beam; The collimating lens group and the excitation mirror are used to adjust the direction and focusing position of the light beam, and to focus the light beam after passing through the Powell lens into a light sheet at the sample.

[0008] As an improvement to the above solution, the optical sheet rotation unit includes: A rotating stage, on which the Powell lens is mounted; An electric hollow servo motor is used to drive the rotary table to rotate, so that the Powell lens rotates synchronously with the rotary table.

[0009] As an improvement to the above scheme, the detection unit includes: The first objective lens and the second objective lens are used to collect the fluorescence signal emitted by the sample, and the optical path collection directions of the first objective lens and the second objective lens are orthogonal. An optical lens group is used to image the fluorescence signals collected by the first objective lens and the second objective lens onto the detection surface of the imaging camera. The imaging camera is used to convert the fluorescence signals acquired by the first objective lens and the second objective lens into fluorescence images.

[0010] As an improvement to the above solution, the detection unit further includes: A dichroic mirror is disposed between the optical lens group and the imaging camera to transmit a first band spectrum and reflect a second band spectrum, thereby splitting the light beam emitted by the optical lens group into two beams with different propagation paths. The detection unit is equipped with only one imaging camera, which is used to simultaneously image the first band spectrum and the second band spectrum to obtain fluorescence images corresponding to the two bands; the imaging camera is a time-delay integral linear array camera.

[0011] As an improvement to the above scheme, the detection unit further includes a linear moving stage, on which a reflector is disposed; when the first objective lens captures a fluorescence signal, the linear moving stage is driven to a first position so that the reflector reflects the fluorescence signal captured by the first objective lens into the optical lens group; when the second objective lens captures a fluorescence signal, the linear moving stage is driven to a second position so that the fluorescence signal captured by the second objective lens enters the optical lens group.

[0012] As an improvement to the above scheme, the installation position of each optical element in the detection unit is determined by solving according to preset constraint conditions; The constraints include: the distance from the first objective lens to the optical lens group is equal to the distance from the second objective lens to the optical lens group; the size of the two beams emitted from the dichroic mirror is less than half the size of the sensor of the imaging camera; the distance between the central axes of the two beams emitted from the dichroic mirror is greater than half the size of the sensor of the imaging camera; and the beam entering the imaging camera is perpendicular to the sensor plane of the imaging camera.

[0013] As an improvement to the above solution, the line-scanning light-sheet imaging microscope system further includes: A sample displacement stage, used to adjust the position of the sample, includes an x-axis displacement stage, a y-axis displacement stage, and a z-axis displacement stage.

[0014] As an improvement to the above solution, the line-scanning light-sheet imaging microscope system further includes a controller; the controller is configured to: When the detection unit detects the first fluorescence image, the sample is moved at a constant speed to obtain the image stack corresponding to the first direction from the detection unit; wherein, the first fluorescence image is the image of the fluorescence signal emitted after the light sheet in the first direction illuminates the sample and is displayed on the imaging camera; After obtaining the image stack corresponding to the first direction, the light sheet rotation unit is controlled to make the light sheet generation unit generate a light sheet in the second direction, wherein the first direction and the second direction are orthogonal; when the detection unit detects the second fluorescence image, the sample is moved at a constant speed, and the image stack corresponding to the second direction is obtained from the detection unit; wherein, the second fluorescence image is the imaging of the fluorescence signal emitted after the light sheet in the second direction illuminates the sample on the imaging camera. The image stack corresponding to the first direction is superimposed and fused with the image stack corresponding to the second direction to obtain a three-dimensional image.

[0015] As an improvement to the above solution, the controller is further configured to: The imaging mode of the system is determined based on the preset imaging accuracy requirements and sample structural properties. When the system is in the first imaging mode, both objectives in the detection unit acquire full scan data of the sample, forming two complementary image stacks, and registering, deconvolving and weighting the pixels of the two image stacks to generate a fused three-dimensional image. When the system is in the second imaging mode, the two objective lenses in the detection unit acquire the full scan data and partial scan data of the sample respectively, forming two asymmetric image stacks, and use a deep learning algorithm to fuse and reconstruct the two image stacks to generate a three-dimensional image. When the system is in the third imaging mode, the detection unit uses two objectives with different magnifications to acquire the scanning data of the sample. The low-magnification objective is used to scan the entire sample, and the high-magnification objective is used to scan the region of interest of the sample, forming two asymmetric image stacks. The two image stacks are then fused and reconstructed using compressed sensing to generate a three-dimensional image.

[0016] Compared with existing technologies, the advantages of the line-scanning light-sheet imaging microscope system provided in this application are as follows: By setting a rotating unit to rotate the Powell prism in the light sheet generation unit, the direction of the light sheet is changed, enabling the acquisition of fluorescence signals emitted by the sample from multiple angles. Then, the images acquired from multiple angles are stacked, fused, and reconstructed by an imaging camera, thereby improving the imaging quality of the three-dimensional image. At the same time, since this application can use a relatively thicker light sheet with a larger field of view, the acquisition speed is greatly improved, which can simultaneously meet the requirements of high-quality imaging and fast image acquisition. By combining a time-delay integral linear array camera with dual-color imaging technology, it has a fast acquisition speed and extremely small time displacement, which further improves the imaging efficiency. It is especially suitable for studying rapid cellular processes and is also beneficial for analyzing studies involving cell dynamics or interactions between biomolecules. Attached Figure Description

[0017] Figure 1 This is a side view of a line-scanning light-sheet imaging microscope system provided in an embodiment of this application; Figure 2 This is a side view of a light sheet generation unit provided in an embodiment of this application; Figure 3 This is a schematic diagram of the light spot distribution during the process of a light beam passing through a light sheet generation unit according to an embodiment of this application, wherein (a) to (f) are the light spot distributions in front of the Powell lens, the first collimating lens, the first aperture, the second collimating lens, the second aperture, and the excitation mirror, respectively; Figure 4 This is a side view of a detection unit provided in an embodiment of this application; Figure 5 This is a schematic diagram of the imaging principle of a time-delay integral linear array camera provided in an embodiment of this application; Figure 6 This is a schematic diagram of the lateral displacement of light in a dichroic mirror provided in an embodiment of this application; Figure 7 This is an optical simulation diagram of a detection unit imaging provided in an embodiment of this application; The reference numerals in the attached figures are explained as follows: 1. Illumination unit; 2. Powell lens; 3. Collimating lens group; 4. Excitation mirror; 5. Sample chamber; 6. First objective lens; 7. First reflecting mirror; 8. Optical lens group; 9. Imaging camera; 10. Second objective lens; 11. Second reflecting mirror; 12. Linear stage; 21. Dichroic mirror; 22. Third reflecting mirror; 23. Fourth reflecting mirror; 24. Fifth reflecting mirror; 31. First collimating mirror; 32. First aperture stop; 33. Second collimating lens; 34. Second aperture. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0019] Light sheet microscopy is a three-dimensional imaging technique that introduces an additional illumination beam to generate an extremely thin sheet of light. When this light illuminates the sample, it excites a thin layer of fluorescence. A probe beam then captures the fluorescence image, allowing for the acquisition of a series of planar fluorescence images by scanning the sample. These images are then stacked and reconstructed to generate a three-dimensional fluorescence image of the sample. Currently, there is always a trade-off between axial resolution and field of view (FOV) in light sheet microscopy. Light sheets with a larger field of view are typically thicker, which is more advantageous for illuminating large sample areas. However, a thicker light sheet broadens the point spread function along the z-axis (which is usually the direction perpendicular to the light sheet, i.e., the depth direction), resulting in the collection of more background noise and a decrease in axial resolution. If a thinner light sheet is used, the sample is scanned over a small illuminated area, requiring a longer acquisition time and resulting in lower efficiency.

[0020] For example, in a traditional light sheet microscope, the objective lens images a sample with dimensions of 0.72 mm × 1 mm × 1 mm. Assuming the objective lens parameters are as follows: magnification M = 25X, numerical aperture NA = 1.0, and field number OFN = 18 mm; then, according to the Rayleigh criterion, when the wavelength λ = 0.5 μm, the lateral resolution is: Its axial resolution dz is close to its lateral resolution dxy, i.e., dz = 0.305 μm. Assuming the sheet thickness is 2W0 = 0.5 μm, the effective sheet length is: in It is half the Rayleigh distance. This indicates half the thickness of the sheet.

[0021] The volume of the sample The image size is 720μm × 1000μm × 1000μm, and the field of view S of a single image is 720μm × 0.8μm. Assume a single-image exposure time... 50ms, overlap ratio 5%, axial step size If the aperture is 0.5 μm, then the imaging time t for the entire sample is: The above process illustrates the trade-off between axial resolution and field of view. To maintain a fast acquisition speed, axial resolution is often sacrificed, which is more beneficial for imaging a large sample area.

[0022] In existing technologies, to improve axial resolution, the sample is typically rotated to obtain an image stack from multiple angles, which is then reconstructed into a 3D image with higher z-axis resolution (perpendicular to the light sheet direction). However, to avoid positional shifts between images, this method requires the sample to be placed at the center of the rotation axis. If the sample is not properly mounted or supported, rotating it may cause image blurring, resulting in decreased accuracy.

[0023] In order to address the problem that the requirements for high-quality imaging and rapid image acquisition cannot be met simultaneously in the prior art, this application proposes a line-scan light-sheet imaging microscope system. By rotating the light sheet, an image stack can be obtained from multiple angles and reconstructed into a three-dimensional image with higher z-axis resolution.

[0024] Please see Figure 1 , Figure 1 This is a side view of a line-scanning light-sheet imaging microscope system provided in an embodiment of this application. The line-scanning light-sheet imaging microscope system includes: Sample chamber 5 is used to place samples; Illumination unit 1 is used to provide excitation light to excite the sample to produce fluorescence, wherein the optical axis of the excitation light is perpendicular to the sample; A light sheet generation unit is used to shape the excitation light to form a light sheet; A light sheet rotation unit is used to change the direction of the light sheet; The detection unit is used to collect fluorescence signals generated by the sample excited by the light sheet from different directions, thereby obtaining image stacks corresponding to the light sheet from different directions, and fusing and reconstructing different image stacks to obtain a three-dimensional image; wherein the direction in which the detection unit collects fluorescence signals is orthogonal to the incident direction of the excitation light and orthogonal to the focal plane of the sample.

[0025] Specifically, the light sheet generation unit includes a Powell lens 2, a collimating lens group 3, and an excitation mirror 4. The excitation light emitted from the illumination unit 1 passes through the Powell lens 2, the collimating lens group 3, and the excitation mirror 4 before illuminating the sample in the sample chamber 5, forming an illumination light path. The light sheet rotation unit acts on the Powell lens 2 to rotate it by a preset angle, thereby changing the direction of the light sheet.

[0026] Furthermore, the illumination unit 1 can provide excitation light of multiple wavelengths. Excitation light refers to light of a specific wavelength that can excite a substance to emit light, used to excite electrons in the substance (such as fluorescent dyes or fluorescent materials), causing them to transition from the ground state to the excited state. Specifically, the excitation light wavelengths provided by the illumination unit include 405nm, 488nm, 561nm, and 637nm, which are suitable for exciting fluorescent markers DAPI (4',6-diamino-2-phenylpyridine), GFP (green fluorescent protein), mCherry (monomer cherry red fluorescent protein), and Cy5 (anthocyanin 5), respectively.

[0027] Furthermore, the detection unit includes: a first objective lens 6, a first reflecting mirror 7, a second objective lens 10, a second reflecting mirror 11, a linear stage 12, an optical lens group 8, and an imaging camera 9. The first objective lens 6 and the second objective lens 10 respectively collect fluorescence signals from two orthogonal directions, forming two detection optical paths. The fluorescence signal from the first detection optical path sequentially passes through the first objective lens 6, the first reflecting mirror 7, the reflecting mirror on the linear stage 12, and the optical lens group 8 before entering the imaging camera 9, generating a fluorescence image. The fluorescence signal from the second detection optical path sequentially passes through the second objective lens 10, the second reflecting mirror 11, and the optical lens group 8 before entering the imaging camera 9, generating a fluorescence image. Finally, the different fluorescence images acquired by the imaging camera are fused and reconstructed to generate a three-dimensional image of the sample.

[0028] As one optional embodiment, the line-scan light-sheet imaging microscope system further includes: A sample displacement stage, used to adjust the position of the sample, includes an x-axis displacement stage, a y-axis displacement stage, and a z-axis displacement stage.

[0029] Specifically, in this embodiment, the sample chamber 5 is made of a transparent material to allow the light sheet to pass through and excite the sample to generate fluorescence, while facilitating the detection unit to capture the fluorescence signal. The sample chamber 5 is equipped with temperature, humidity, and gas environment control functions to maintain the physiological state of the sample, especially for long-term imaging of live cells or live tissues.

[0030] The microscope system of this application embodiment is provided with a sample displacement stage, which includes an x-axis displacement stage, a y-axis displacement stage and a z-axis displacement stage. The sample displacement stage can adjust the position of the sample in the sample chamber, thereby ensuring that a specific area of ​​the sample is accurately aligned with the light plate and achieving high-quality image acquisition.

[0031] As one optional embodiment, the light sheet generation unit includes: Powell lens 2 is used to convert the light beam emitted by the illumination unit 1 into a uniform linear light beam; The collimating lens group 3 and the excitation mirror 4 are used to adjust the direction and focusing position of the light beam, and to focus the light beam after passing through the Powell lens 2 into a light sheet at the sample.

[0032] Specifically, the Powell lens is the core element forming the light sheet; it is a special type of optical prism comprising a two-dimensional aspherical surface. In this embodiment, when the light beam passes through the Powell lens 2, the Powell lens 2 converts the passing light beam into a straight beam. Then, the collimating lens group 3 and the excitation mirror 4 adjust the direction and focusing position of the beam, focusing the uniform straight beam into a light sheet to illuminate a specific plane. It should be noted that the specific plane refers to the focal plane of the detection unit, that is, the area where the light sheet interacts with the sample. Compared with traditional cylindrical prisms, the Powell lens 2 can effectively avoid the central hot spot and edge fading problems of Gaussian beams.

[0033] Furthermore, the beam passing through the Powell lens 2 can be a Gaussian beam, a Bessel beam, an Airy beam, etc., and in this embodiment, a Bessel beam is preferred. A Bessel beam is a non-diffraction beam that maintains its shape and intensity distribution during propagation. Using a Bessel beam and the Powell lens, a high-quality light sheet can be formed, which is beneficial for high-resolution, large-field-of-view three-dimensional imaging.

[0034] Furthermore, the collimating lens group 3 comprises several collimating mirrors, used to convert the diverging light beam into a parallel light beam. Preferably, the collimating lens group includes a first collimating mirror 31 and a second collimating mirror 33, capable of modulating the light beam to generate a collimated beam, thereby improving the quality of the subsequently generated light sheet. Then, the excitation mirror focuses the collimated light beam onto the sample, exciting the target substance in the sample, thereby causing the sample to generate a fluorescence signal.

[0035] Furthermore, the light-gathering unit also includes at least one aperture for shaping the light beam. Preferably, please refer to [reference needed]. Figure 2 The light sheet generation unit includes a first aperture 32 and a second aperture 34. Specifically, the first aperture 32 and the second aperture 34 can obtain different beam spot sizes by switching different window sizes, thereby producing light sheets of different sizes. The beam emitted from the illumination unit 1 passes sequentially through the Powell lens 2, the first collimating lens 31, the first aperture 32, the second collimating lens 33, the second aperture 34, and the excitation mirror 4 before reaching the sample chamber 5.

[0036] Specifically, based on Figure 2 The light spot distribution of the light beam emitted by the illumination unit 1 from the light-generating unit shown is as follows: Figure 3 As shown in (a), the beam, after being modulated by the Powell lens 2, forms light that is uniformly divergent in the z-axis direction and parallel in the y-axis direction; the beam spot distribution at the first collimating lens 31 is as follows. Figure 3 As shown in (b), light parallel to the z-axis and focused in the y-axis direction is obtained after passing through the first collimating lens 31, and its spot distribution is as follows. Figure 3 As shown in (c), the size of the light spot along the z-axis is changed by the first aperture 32; the light spot distribution at the second collimating lens 33 is as follows. Figure 3 As shown in (d), light focused along the z-axis and parallel along the y-axis is obtained after passing through the second collimating lens 33, and its spot distribution is as follows. Figure 3 As shown in (e), the beam size along the y-axis is changed by the second aperture 34. Afterward, the beam exits through the excitation mirror 4 and enters the sample chamber 5. The beam spot distribution at the entrance pupil of the excitation mirror 4 is as follows. Figure 3 As shown in (f), the light emitted from the excitation mirror 4 is parallel to the z-axis and focused in the y-axis direction, thus obtaining a light sheet with uniform intensity distribution in the z-axis direction and a very thin light sheet in the y-axis direction at the focal plane of the objective lens. Furthermore, by rotating the Powell lens 2 by 90°, a light sheet with a direction perpendicular to the direction of the light sheet before rotation can be obtained.

[0037] As one optional embodiment, the light plate rotation unit includes: A rotating stage, on which the Powell lens 2 is mounted; An electric hollow servo motor is used to drive the rotary table to rotate, so that the Powell lens 2 rotates synchronously with the rotary table.

[0038] Specifically, the optical sheet rotation unit includes a rotary table and an electric hollow servo motor. The electric hollow servo motor drives the rotary table to rotate, and the Powell lens 2 is mounted on the rotary table and rotates synchronously with it. By rotating the Powell lens 2, the optical sheet generation unit generates optical sheets in different directions. Exemplarily, this application embodiment selects a high-precision electric hollow servo motor with a repeatability of 0.006° and an absolute accuracy of 0.006° to ensure the stability of the optical system.

[0039] As one optional embodiment, the detection unit includes: The first objective lens 6 and the second objective lens 10 are used to collect the fluorescence signal emitted by the sample, and the optical path collection directions of the first objective lens 6 and the second objective lens 10 are orthogonal. Optical lens group 8 is used to image the fluorescence signals collected by the first objective lens 6 and the second objective lens 10 onto the detection surface of the imaging camera 9. The imaging camera 9 is used to convert the fluorescence signals collected by the first objective lens 6 and the second objective lens 10 into fluorescence images.

[0040] Among them, optical lens group 8 is a tube lens.

[0041] Specifically, the detection unit of this embodiment includes at least two objective lenses and an imaging camera. The two objective lenses are used to collect fluorescence emitted by the sample in different directions. The fluorescence collected by the objective lenses in different directions is imaged on the detection surface of the imaging camera 9 after passing through the optical lens group 8. After fusing and reconstructing multiple images in different directions collected by the imaging camera 9, a three-dimensional image with a large field of view and high resolution is obtained.

[0042] Please see Figure 4 , Figure 4 This is a side view of a detection unit provided in an embodiment of this application.

[0043] As one optional embodiment, the detection unit further includes a linear stage 12, on which a reflector is disposed; when the first objective lens 6 captures a fluorescence signal, the linear stage 12 is driven to a first position so that the reflector reflects the fluorescence signal captured by the first objective lens 6 into the optical lens group 8; when the second objective lens 10 captures a fluorescence signal, the linear stage 12 is driven to a second position so that the fluorescence signal captured by the second objective lens 10 enters the optical lens group 8.

[0044] Specifically, to ensure that the signals captured by the first objective lens 6 and the second objective lens 10 can both enter the tube mirror, the detection unit in this embodiment further includes a linear moving stage 12, on which a reflector is mounted. When the first objective lens 6 captures a signal, the linear moving stage 12 is driven to a first position (Position 1), so that the reflector on the linear moving stage 12 reflects the light collected by the first objective lens 6 into the tube mirror. When the second objective lens 10 captures a signal, the linear moving stage 12 is driven to a second position (Position 2), so that the reflector on the linear moving stage 12 avoids the transmission of the light collected by the second objective lens 10, allowing the light to be directed towards the tube mirror.

[0045] In some embodiments, the imaging camera 9 in the detection unit is a line scan camera. The line scan camera acquires image information by scanning the sample line by line. The line scan camera typically captures only one line at a time, and the line width is usually one pixel. As the sample moves, the line scan camera continuously captures images of the lines, thereby converging the lines into an area and obtaining a high-resolution large sample image.

[0046] Preferably, the line scan camera is a TDI (Time Delay Integration) line scan camera. Based on multiple exposures of the same target, the TDI line scan camera significantly increases light energy collection through delay integration. Compared to ordinary line scan cameras, TDI line scan cameras have higher sensitivity, higher responsiveness, and a wider dynamic range.

[0047] The principle of TDI line scan camera is as follows: Figure 5 As shown, a TDI line scan camera consists of multiple photosensitive row arrays. When the sample moves, the pixels in each row independently capture the light signal. As the sample moves, the signals captured in these row arrays are precisely aligned and accumulated in time, meaning signals from the same location are continuously superimposed, increasing the total signal intensity and ultimately merging to output the data image. By accumulating the signals from multiple rows, the TDI line scan camera effectively increases the signal strength while slowing down noise growth, thus improving the image's signal-to-noise ratio. Figure 5 As shown, assuming the TDI line scan camera uses 5 rows of pixels for superposition, since the 5 rows of pixels sequentially acquire and integrate the same position, the exposure time of the final image is equivalent to 5 times the single-row exposure time, which significantly improves the signal-to-noise ratio of the image.

[0048] For example, taking a sample with dimensions of 0.72mm × 1mm × 1mm as an example, where L x =720μm, L y =1000μm, L z=1000μm. The imaging camera is a TDI line array camera with a single objective lens. Assuming the objective lens parameters and exposure time are consistent with those in the above embodiment, the objective lens parameters are as follows: magnification M = 25X, numerical aperture NA = 1.0, and field number OFN = 18mm. The sheet thickness 2W0 is 0.5μm, therefore the effective sheet length 2Zr is 0.8μm, and the sheet size is 720μm × 0.5μm × 0.8μm. The length of the sheet magnified to the camera's detection surface by the objective lens is... for: Given that the pixel size p is 5μm, the TDI level of the camera can be set. The exposure time is 4 for a single image. Equivalent to 50ms, the camera's line rate for: Then the moving speed of the displacement table for: Therefore, sample imaging time for: In another example, the imaging camera is a TDI line-scan camera with dual objectives for acquiring the aforementioned sample. Assume the objective parameters and exposure time are consistent with those in the previous embodiment. The sheet thickness 2W0 is 2 μm, therefore the effective sheet length 2Zr is 12.6 μm, and the sheet size is 720 μm × 0.5 μm × 12.6 μm. The length of the sheet magnified to the camera's detection surface by the objective is... for: Given that the pixel size p is 5μm, the TDI level of the camera can be set. The exposure time is 63. Equivalent to 50ms, the camera's line rate for: Then the moving speed of the displacement table for: Therefore, for the above samples, the sample imaging time for: It can be seen that the scheme using a TDI line scan camera and setting dual objective lenses for acquisition has a faster imaging speed than the scheme using a single objective lens, which greatly improves the imaging efficiency.

[0049] It should be noted that the values ​​in the above embodiments are merely illustrative examples, intended to clearly illustrate the technical solution of the present invention and the technical effects it can achieve, and not to limit the present invention. Without departing from the core concept of the present invention, the corresponding values ​​can be adaptively adjusted according to actual circumstances, and such adjustments are all considered to be within the scope of protection of the present invention.

[0050] As one optional embodiment, the detection unit further includes: A dichroic mirror 21 is disposed between the optical lens group 8 and the imaging camera 9 to transmit the first band spectrum and reflect the second band spectrum, so as to split the light beam emitted by the optical lens group 8 into two light beams with different propagation paths. The detection unit is equipped with only one imaging camera 9, which is used to simultaneously image the first band spectrum and the second band spectrum to obtain fluorescence images corresponding to the two bands; the imaging camera 9 is a time-delay integral linear array camera.

[0051] In this embodiment, the TDI linear array camera achieves high-resolution and high-sensitivity imaging through single-line imaging. However, it can only capture image data from one spectral channel at a time. To capture images from multiple spectral channels, multiple scans or filter changes are required. To further improve the acquisition speed, this embodiment places a dichroic mirror 21 between the tube lens and the TDI linear array camera. After the spectrum is separated by the dichroic mirror 21, the camera can simultaneously capture image data from two spectral channels, achieving dual-color synchronous imaging and improving imaging efficiency.

[0052] Specifically, such as Figure 4 As shown, the dichroic mirror 21 is positioned behind the tube mirror. A portion of the light beam emitted from the tube mirror passes through the dichroic mirror 21 and is reflected sequentially by the third reflecting mirror 22 and the fifth reflecting mirror 24 before entering the imaging camera 9. The other portion of the light beam is reflected by the dichroic mirror 21 and then by the fourth reflecting mirror 23 before entering the imaging camera 9, thus achieving dual-color synchronous imaging.

[0053] In this embodiment, based on the emission wavelength of the marker and the spectral image data of the required band in each experiment, a long-pass dichroic mirror with a suitable cutoff wavelength is selected. This allows shorter wavelength signals to be reflected to the second path (Path 2), while longer wavelength signals are transmitted through the dichroic mirror 21 and propagate along the first path (Path 1). This embodiment utilizes this optical structure to reduce image acquisition time to half that of traditional imaging methods, minimizes the impact of photobleaching on the sample, and preserves the quality of the fluorescence signal over time. Traditional dual-color synchronous imaging systems typically require two imaging cameras, each acquiring image data from one spectral channel. The data from the two cameras are then fused for imaging; however, data synchronization between the two sensors is challenging. Therefore, this embodiment uses only one imaging camera for dual-color image capture, significantly reducing system complexity and cost.

[0054] As one of the optional embodiments, the installation positions of each optical element in the detection unit are determined according to preset constraint conditions; The constraints include: the distance from the first objective lens 6 to the optical lens group 8 is equal to the distance from the second objective lens 10 to the optical lens group 8; the size of the two beams emitted from the dichroic mirror 21 is less than half the size of the sensor of the imaging camera 9; the distance between the central axes of the two beams emitted from the dichroic mirror 21 is greater than half the size of the sensor of the imaging camera 9; and the beam entering the imaging camera 9 is perpendicular to the sensor plane of the imaging camera 9.

[0055] Specifically, to optimize optical performance, the position and angle of the optical elements are crucial and require precise placement. The optical elements include a first objective lens 6, a first reflecting mirror 7, an optical lens group 8, a second objective lens 10, a second reflecting mirror 11, a linear stage 12, a dichroic mirror 21, a third reflecting mirror 22, a fourth reflecting mirror 23, and a fifth reflecting mirror 24. In this embodiment, the position of each optical element is determined based on preset constraints. The constraints are as follows: First, the two optical paths between the two objectives and the optical lens group 8 are of the same length to ensure a sharp focus is formed on the sensor of the imaging camera 9, that is, the distance from the first objective 6 to the tube lens is equal to the distance from the second objective 10 to the tube lens. For Figure 4 The optical path shown is constrained as follows: c + d + e = a + b + e, where c is the optical path length from the first objective lens 6 to the first reflecting mirror 7, d is the optical path length from the first reflecting mirror 7 to the first position of the linear moving stage 12, a is the optical path length from the second objective lens 10 to the second reflecting mirror 11, b is the optical path length from the second reflecting mirror 11 to the first position of the linear moving stage 12, and e is the optical path length from the first position of the linear moving stage 12 to the tube mirror.

[0056] Secondly, optical characteristics need to be considered when determining the path, such as... Figure 6 As shown, due to the thickness of the dichroic mirror 21, there is a lateral displacement of light in the first path. For example, if a 1mm thick dichroic mirror with a refractive index of 1.53 is applied at 45°, then according to Snell's law: in, The refractive index of light in air. Let be the refractive index of light in a dichroic mirror. Angle of incidence Let be the angle of refraction, then: like Figure 6 As shown, according to trigonometric functions: , The thickness of the dichroic mirror , The lateral displacement can be obtained by finding the distance between points A and C. for: Further, please refer to Figure 7 The distance between the central axes of the first path and the second path It must be larger than the sensor size To avoid signal interference, half of the value is used, as expressed by the formula below: Furthermore, beam size It must be smaller than the sensor size Half of: At the same time, the distance between the central axes and beam size The sum must be less than the sensor size. To ensure complete capture of the dual-channel signal: .

[0057] Furthermore, the constraints also include: the light beams incident on the imaging camera 9 are all perpendicular to the sensor plane of the imaging camera 9, and the installation of the third reflector 22 and the fifth reflector 24 in the first path does not block the light path of the second path.

[0058] Based on the above constraints, assuming a maximum beam size D0, the installation positions of each optical element are calculated to determine the space requirements under extreme conditions. The possible maximum beam size is equal to the diameter of the inlet pupil of the tube lens. For example, assuming a maximum beam size of 20mm, and using an imaging camera 9 with a sensor size of 45360μm × 1280μm, the following constraints are satisfied: This ensures complete capture of the dual-channel signals.

[0059] Furthermore, please refer to Figure 4 Calculate the optical paths a, b, c, and d to satisfy the constraint: c + d + e = a + b + e, and the optical specifications of the first objective lens 6, the second objective lens 10, and the tube lens. Calculate the lateral displacement d. CD Determine the lateral position of optical path g in the first path. Based on the optical specifications of the tube lens, first determine the lengths of optical paths f, j, and k in the second path, ensuring they are focused on the sensor plane of the imaging camera 9. Then, determine the distance D between the central axes of the first and second paths based on the constraint relationship between the beam size and the sensor size. c After determining D c After f, j, k, the lengths of the light paths g, h, i in the first path are calculated using trigonometric functions according to the constraints, as well as the tilt angle A1 between the third mirror 22 and the fifth mirror 24, where the third mirror 22 and the fifth mirror 24 are parallel.

[0060] Finally, optical performance simulations were performed to confirm the feasibility of the system, such as... Figure 7 As shown. The positions of the optical elements in the detection unit need to be precisely calculated to optimize optical performance and avoid signal interference or physical beam cutoff during transmission. Simultaneously, the distance between the first and second paths is crucial, ensuring that light from each path can be clearly focused onto the sensor plane.

[0061] As one optional embodiment, the line-scan light-sheet imaging microscope system further includes a controller; the controller is configured to: When the detection unit detects the first fluorescence image, the sample is moved at a constant speed to obtain the image stack corresponding to the first direction from the detection unit; wherein, the first fluorescence image is the image of the fluorescence signal emitted after the light sheet in the first direction illuminates the sample and is displayed on the imaging camera 9. After obtaining the image stack corresponding to the first direction, the light sheet rotation unit is controlled to make the light sheet generation unit generate a light sheet in the second direction, wherein the first direction and the second direction are orthogonal; when the detection unit detects the second fluorescence image, the sample is moved at a constant speed, and the image stack corresponding to the second direction is obtained from the detection unit; wherein, the second fluorescence image is the imaging of the fluorescence signal emitted after the light sheet in the second direction illuminates the sample on the imaging camera 9. The image stack corresponding to the first direction is superimposed and fused with the image stack corresponding to the second direction to obtain a three-dimensional image.

[0062] Specifically, in this embodiment, the Bessel beam emitted by the illumination unit passes through the Powell lens 2, the collimating lens group 3, and the excitation mirror 4 to form a light sheet in the first direction. The light sheet in the first direction illuminates the sample, and the fluorescence emitted by the sample is captured by the first objective lens 6. The fluorescence captured by the first objective lens 6 is imaged on the detection surface of the imaging camera 9 after passing through a series of optical elements. Then, the image stack in the first direction is obtained by moving the sample at a constant speed.

[0063] After obtaining the image stack in the first direction, the electric hollow servo motor is driven to rotate the rotary table by 90°, causing the Powell lens 2 to rotate by 90° accordingly. This allows the light sheet generation unit to generate a light sheet in the second direction, which is orthogonal to the first direction. Similarly, the light sheet in the second direction illuminates the sample, and the fluorescence emitted by the sample is captured by the second objective lens 10. The fluorescence captured by the second objective lens 10 is imaged on the detection surface of the imaging camera 9 after passing through a series of optical elements. The image stack in the second direction is obtained by moving the sample at a constant speed.

[0064] Finally, the image stacks in the first and second directions are superimposed and fused to obtain a large field-of-view, high-resolution 3D imaging image. When the image stack in the first direction is an image with a resolution of a×a×b and the image stack in the second direction is an image with a resolution of a×b×a, the superimposed and fused images yield a high-resolution 3D image with a resolution of a×a×a.

[0065] In this system, the frame rate of the imaging camera 9 is matched to the sample's moving speed. The essence of a light-sheet microscope is to illuminate the sample with a thin light sheet, capturing only the thin layer illuminated by the light sheet. The illuminated sample area emits fluorescence, and the imaging lens only captures the signal from this illuminated area. Because the effective length of the light sheet is limited, to obtain the entire sample structure, the sample needs to move at a constant speed. During this movement, the light sheet sequentially illuminates each cross-section of the sample, and the detection unit simultaneously acquires two-dimensional images of each cross-section. These consecutive two-dimensional images are stacked to ultimately form the complete three-dimensional structure of the sample.

[0066] This embodiment employs multi-angle acquisition imaging to ensure consistent resolution across the x, y, and z axes. Furthermore, this embodiment can utilize a relatively thicker light sheet with a larger field of view, thereby improving acquisition speed and reducing acquisition time.

[0067] For example, for a sample with dimensions of 0.72mm × 1mm × 1mm, dual-objective imaging was used. The imaging objective parameters and single-frame exposure time were consistent with those in the previous example. Assuming the light sheet thickness was 2μm, the effective light sheet length was 12.6μm, 15 times the original. The single imaging field of view of this light sheet was 720μm × 12.6μm, with an axial step size of 2μm. Therefore, the imaging time for the entire sample was approximately 1.16 hours. By changing the direction of the light sheet and setting up dual-objective fluorescence acquisition, the acquisition time was significantly reduced from 36.5 hours to 1.16 hours, effectively improving the scanning efficiency. During the imaging process, the sample is scanned in two directions. First, the light sheet is controlled to illuminate the sample along the first direction and scanned to obtain an image with a resolution of 0.305μm × 0.305μm × 2μm. Then, the direction of the light sheet illuminating the sample is adjusted so that it illuminates the sample along the second direction and scans to obtain an image with a resolution of 0.305μm × 2μm × 0.305μm. The two images are superimposed to reconstruct a high-resolution three-dimensional image with a resolution of 0.305μm × 0.305μm × 0.305μm. This embodiment not only improves the imaging speed but also reduces the effects of photobleaching and phototoxicity.

[0068] It should be noted that the values ​​in the above embodiments are merely illustrative examples, intended to clearly illustrate the technical solution of the present invention and the technical effects it can achieve, and not to limit the present invention. Without departing from the core concept of the present invention, the corresponding values ​​can be adaptively adjusted according to actual circumstances, and such adjustments are all considered to be within the scope of protection of the present invention.

[0069] As one optional embodiment, the controller is further configured to: The imaging mode of the system is determined based on the preset imaging accuracy requirements and sample structural properties. When the system is in the first imaging mode, both objectives in the detection unit acquire full scan data of the sample, forming two complementary image stacks, and registering, deconvolving and weighting the pixels of the two image stacks to generate a fused three-dimensional image. When the system is in the second imaging mode, the two objective lenses in the detection unit acquire the full scan data and partial scan data of the sample respectively, forming two asymmetric image stacks, and use a deep learning algorithm to fuse and reconstruct the two image stacks to generate a three-dimensional image. When the system is in the third imaging mode, the detection unit uses two objectives with different magnifications to acquire the scanning data of the sample. The low-magnification objective is used to scan the entire sample, and the high-magnification objective is used to scan the region of interest of the sample, forming two asymmetric image stacks. The two image stacks are then fused and reconstructed using compressed sensing to generate a three-dimensional image.

[0070] Specifically, in this embodiment, the area of ​​the sample to be scanned can be determined according to the set sampling requirements, and a suitable objective lens combination can be selected for fluorescence acquisition to obtain high-resolution and accurate three-dimensional images, greatly improving the system flexibility. Based on the set imaging accuracy requirements and sample structural properties, if high imaging accuracy is required and time is not a concern, the system uses a first imaging mode for scanning and image generation, reconstructing a high-resolution true three-dimensional image from the full scan dataset; if the sample is a homogeneous and dense sample, the system uses a second imaging mode for scanning and image generation, thereby obtaining a three-dimensional image result that combines quality and efficiency; if the sample is a sparse sample, or for targeted research on regions of interest, the system uses a third imaging mode for scanning and image generation, thereby effectively improving efficiency while ensuring imaging quality.

[0071] Furthermore, in the first imaging mode, fluorescence is acquired using a first objective lens 6 and a second objective lens 10 of identical specifications, and the resulting datasets are fused computationally. The first objective lens 6 and the second objective lens 10 provide two complementary datasets from two different perspectives, representing the unique viewpoints of the sample. After obtaining the image data from the two perspectives, algorithms such as registration, deconvolution, and weighted averaging are used to fuse and reconstruct the image data, generating a three-dimensional image. Because the overlapping areas in the datasets provide redundancy, allowing the system to cross-check data consistency, the processed three-dimensional image has high accuracy.

[0072] The purpose of registration is to align datasets from two different perspectives. Two objectives acquire data from two different perspectives, specifically in the XZ and YZ directions. Then, the two sets of data are mapped to the same three-dimensional coordinate system, ensuring that the same physical location of the sample corresponds to the same pixel coordinates in both sets of data. Specifically, common spatial features are extracted from the two datasets as registration feature points. Then, the spatial transformation relationship between the two sets of data is calculated based on the extracted feature points, aligning one set of data to the other, thus achieving registration.

[0073] The purpose of deconvolution is to eliminate blur and restore the true structure of the sample. Specifically, the point spread function of two objectives is obtained separately, and then deconvolution is performed on the respective datasets using their respective point spread functions to restore the clear structure from the perspective of each objective. After deconvolution, the blur artifacts in both sets of data are eliminated.

[0074] Registration and deconvolution ensure the two sets of data meet alignment and sharpness requirements, allowing for the evaluation of signal quality for each pixel and the determination of its corresponding weight. Then, each pixel in the 3D dataset is weighted and summed to obtain the fused pixel value. After fusing all pixels, each layer of 2D images is stacked along the z-axis, ultimately forming a complete, high-precision 3D dataset that includes complementary details from both perspectives and eliminates the deficiencies of a single perspective through weight balancing.

[0075] Furthermore, in the second imaging mode, an asymmetric overlay image dataset is obtained using the same two objectives. One objective scans the entire sample to obtain a global view of the sample, but its image quality is relatively poor due to scattering or shadows; the other objective scans a partial area to obtain a clearer local view with sharper details, but its coverage is limited. A deep learning algorithm is then used to analyze, enhance, and reconstruct the asymmetric image overlay into a high-resolution 3D image. This embodiment utilizes a reliable full-sample global view dataset, a corresponding high-quality local view dataset, and a historical real 3D model reconstruction dataset to train a network model, such as U-Net (a convolutional neural network for image segmentation) or GAN (generative adversarial network), to obtain a deep learning model that can be used for image fusion reconstruction in the second imaging mode. Exemplarily, advanced algorithms such as Self-Net (self-supervised network), SSAI-3D (self-supervised isotropic 3D reconstruction), and CycleGAN (recurrent generative adversarial network) can also be used to generate more accurate and reliable 3D results. The network model in this embodiment can effectively remove noise and other artifacts in the image, improve the resolution of the global view, and use high-quality local views as a reference to infer and supplement the unscanned details in the global view, making the overall image more complete and clear, and improving the acquisition efficiency while ensuring imaging accuracy.

[0076] Furthermore, in the third imaging mode, for sparse samples, asymmetric image superposition obtained from objectives of different magnifications is reconstructed using compressed sensing. This mode employs two objectives with different magnifications: a low-magnification objective for rapid scanning of the entire sample, and a high-magnification objective for fine scanning of the region of interest. Compressed sensing technology is then used for image fusion and reconstruction, effectively improving efficiency.

[0077] Specifically, the key to compressed sensing technology is leveraging certain prior knowledge of the signal, such as its sparsity. For sparse target samples, the signal of interest can be represented in a suitable mathematical domain with a small number of non-zero elements, such as wavelet transform or Fourier transform, where most of the information is concentrated in a few important coefficients. In this embodiment, the low-magnification objective captures a sampled full-sample image stack, which is fast but lacks fine detail, while the high-magnification objective provides a dense, detailed subset as a patch. The low-magnification image is converted to a sparse domain, where most coefficients are small or zero, while the high-magnification image, through a constraint algorithm, defines how sparse features should appear at full resolution. This approach improves resolution to near-high-magnification quality without scanning the entire sample at high magnification, saving time and reducing illumination. It is suitable for low-sampled, low-magnification data, enabling faster data acquisition while reducing the number of measurements, and flexibly adapting to any critical high-magnification subset without requiring a full double scan.

[0078] Compared to existing technologies, this application provides a line-scanning light-sheet imaging microscope system. Its advantages lie in the following: by rotating the Powell prism in the light-sheet generation unit using a light-sheet rotation unit, the direction of the light sheet is changed, enabling multi-angle acquisition of fluorescence signals emitted by the sample. The images acquired from multiple angles are then stacked and fused together by an imaging camera, thereby improving the imaging quality of the three-dimensional image. Furthermore, since this application can use a relatively thicker light sheet with a larger field of view, the acquisition speed is greatly improved. By combining a time-delay integral linear array camera with dual-color imaging technology, it achieves rapid acquisition speed and minimal time displacement, further enhancing imaging efficiency. This system is particularly suitable for studying rapid cellular processes and is also beneficial for analyzing studies involving cell dynamics or interactions between biomolecules. The line-scanning light-sheet imaging microscope system of this application combines the advantages of line scanning technology, thin-sheet microscopy, multi-viewpoint method, and simultaneous dual-color imaging, simultaneously meeting the requirements for high-quality imaging and rapid image acquisition. In this microscope system, acquisition time is minimized, while reducing photobleaching and phototoxicity to the sample.

[0079] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications are also considered to be within the scope of protection of this application.

Claims

1. A line-scanning light-sheet imaging microscope system, characterized in that, include: The sample chamber is used to hold samples; An illumination unit is used to provide excitation light to excite the sample to produce fluorescence, wherein the optical axis of the excitation light is perpendicular to the sample. A light sheet generation unit is used to shape the excitation light to form a light sheet; A light sheet rotation unit is used to change the direction of the light sheet; The detection unit is used to collect fluorescence signals generated by the sample excited by the light sheet from different directions, thereby obtaining image stacks corresponding to the light sheet from different directions, and fusing and reconstructing different image stacks to obtain a three-dimensional image; wherein the direction in which the detection unit collects fluorescence signals is orthogonal to the incident direction of the excitation light and orthogonal to the focal plane of the sample.

2. The line-scanning light-sheet imaging microscope system as described in claim 1, characterized in that, The optical sheet generation unit includes: A Powell lens is used to convert the light beam emitted by the illumination unit into a uniform linear light beam; The collimating lens group and the excitation mirror are used to adjust the direction and focusing position of the light beam, and to focus the light beam after passing through the Powell lens into a light sheet at the sample.

3. The line-scanning light-sheet imaging microscope system as described in claim 2, characterized in that, The optical sheet rotation unit includes: A rotating stage, on which the Powell lens is mounted; An electric hollow servo motor is used to drive the rotary table to rotate, so that the Powell lens rotates synchronously with the rotary table.

4. The line-scanning light-sheet imaging microscope system as described in claim 1, characterized in that, The detection unit includes: The first objective lens and the second objective lens are used to collect the fluorescence signal emitted by the sample, and the optical path collection directions of the first objective lens and the second objective lens are orthogonal. An optical lens group is used to image the fluorescence signals collected by the first objective lens and the second objective lens onto the detection surface of the imaging camera. The imaging camera is used to convert the fluorescence signals acquired by the first objective lens and the second objective lens into fluorescence images.

5. The line-scanning light-sheet imaging microscope system as described in claim 4, characterized in that, The detection unit also includes: A dichroic mirror is disposed between the optical lens group and the imaging camera to transmit a first band spectrum and reflect a second band spectrum, thereby splitting the light beam emitted by the optical lens group into two beams with different propagation paths. The detection unit is equipped with only one imaging camera, which is used to simultaneously image the first band spectrum and the second band spectrum to obtain fluorescence images corresponding to the two bands; the imaging camera is a time-delay integral linear array camera.

6. The line-scanning light-sheet imaging microscope system as described in claim 4, characterized in that, The detection unit further includes a linear moving stage, on which a reflector is disposed; when the first objective lens captures a fluorescence signal, the linear moving stage is driven to a first position so that the reflector reflects the fluorescence signal captured by the first objective lens into the optical lens group; When the second objective lens captures a fluorescence signal, the linear stage is driven to a second position so that the fluorescence signal captured by the second objective lens enters the optical lens group.

7. The line-scanning light-sheet imaging microscope system as described in claim 5, characterized in that, The installation positions of each optical element in the detection unit are determined by solving according to preset constraint conditions; The constraints include: the distance from the first objective lens to the optical lens group is equal to the distance from the second objective lens to the optical lens group; the size of the two beams emitted from the dichroic mirror is less than half the size of the sensor of the imaging camera; the distance between the central axes of the two beams emitted from the dichroic mirror is greater than half the size of the sensor of the imaging camera; and the beam entering the imaging camera is perpendicular to the sensor plane of the imaging camera.

8. The line-scanning light-sheet imaging microscope system as described in claim 1, characterized in that, The line-scan light-sheet imaging microscope system also includes: A sample displacement stage, used to adjust the position of the sample, includes an x-axis displacement stage, a y-axis displacement stage, and a z-axis displacement stage.

9. The line-scanning light-sheet imaging microscope system according to any one of claims 1 to 8, characterized in that, The line-scan light-sheet imaging microscope system also includes a controller; the controller is configured to: When the detection unit detects the first fluorescence image, the sample is moved at a constant speed to obtain the image stack corresponding to the first direction from the detection unit; wherein, the first fluorescence image is the image of the fluorescence signal emitted after the light sheet in the first direction illuminates the sample and is displayed on the imaging camera; After obtaining the image stack corresponding to the first direction, the light sheet rotation unit is controlled to make the light sheet generation unit generate a light sheet in the second direction, wherein the first direction and the second direction are orthogonal; when the detection unit detects the second fluorescence image, the sample is moved at a constant speed, and the image stack corresponding to the second direction is obtained from the detection unit; wherein, the second fluorescence image is the imaging of the fluorescence signal emitted after the light sheet in the second direction illuminates the sample on the imaging camera. The image stack corresponding to the first direction is superimposed and fused with the image stack corresponding to the second direction to obtain a three-dimensional image.

10. The line-scanning light-sheet imaging microscope system as described in claim 9, characterized in that, The controller is also configured to: The imaging mode of the system is determined based on the preset imaging accuracy requirements and sample structural properties. When the system is in the first imaging mode, both objectives in the detection unit acquire full scan data of the sample, forming two complementary image stacks, and registering, deconvolving and weighting the pixels of the two image stacks to generate a fused three-dimensional image. When the system is in the second imaging mode, the two objective lenses in the detection unit acquire the full scan data and partial scan data of the sample respectively, forming two asymmetric image stacks, and use a deep learning algorithm to fuse and reconstruct the two image stacks to generate a three-dimensional image. When the system is in the third imaging mode, the detection unit uses two objectives with different magnifications to acquire the scanning data of the sample. The low-magnification objective is used to scan the entire sample, and the high-magnification objective is used to scan the region of interest of the sample, forming two asymmetric image stacks. The two image stacks are then fused and reconstructed using compressed sensing to generate a three-dimensional image.

Images