System and method for rapid and label-free imaging of biological tissue based on microscopy with ultraviolet single-plane irradiation

Abstract

A method and system (100) for rapid, label-free, non-destructive imaging of unprocessed biological tissue based on the use of ultraviolet single plane irradiation (MUSI) microscopy. The system (100) or method utilizes the endogenous fluorescence of the biological tissue as imaging contrast, eliminates the out-of-focus fluorescence signal by selective planar illumination, and detects the generated fluorescence signal from an orthogonal direction. This dual-axis configuration decouples the illumination from the detection path, overcoming the inherent contradiction between long depth of field and high spatial resolution. Hematoxylin-eosin (Hamp; Hamp; Hamp; Hamp; Hamp; Hamp; Hamp; Hamp; e) staining), the image generated by the MUSI shows equivalent or even better diagnostic features, and a more potential intraoperative and postoperative diagnostic tool is provided for clinicians, pathologists and surgeons.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Provisional Patent Application Serial No. 63 / 417,682 filed on October 19, 2022, the disclosure of which is incorporated herein by reference in its entirety. Technical Field

[0003] The present invention relates to a system and method for rapid, label-free and non-destructive imaging of untreated biological tissue based on microscopy using ultraviolet single plane irradiation (MUSI). Background Art

[0004] For decades, histopathology has been the gold standard for surgical margin assessment (SMA). However, conventional pathological examination based on formalin-fixed and paraffin-embedded (FFPE) tissue is laborious and time-consuming, and cannot guide surgeons intraoperatively. Although frozen sections can be used as a rapid alternative to FFPE, they still require a 20- to 30-minute turnaround time during surgery. In addition, frozen sections have limited sampling of resection margins and are subject to freezing artifacts in adipose tissue, which will affect the interpretation and diagnostic accuracy of histopathology.

[0005] In recent years, the latest advances in optical microscopy have made it possible to image freshly excised tissues quickly and non-destructively, which has great potential for simplifying the current clinical practice of FFPE histology. Imaging modes based on exogenous fluorophores (such as fluorescence confocal microscopy, light sheet microscopy, structured illumination microscopy, and ultraviolet (UV) surface excitation microscopy) can provide sufficient sampling of large resection margins within the time frame of immediate detection, providing highly specific cell features for diagnosis. However, some new fluorescent contrast agents pose a threat to intraoperative surgical procedures because they may cause toxicity to patients. In addition, the staining process may interfere with subsequent molecular analysis (such as fluorescence in situ hybridization and DNA / RNA sequencing).

[0006] In contrast, imaging techniques based on intrinsic contrast mechanisms are more popular in modern clinical settings. For example, in the fields of ophthalmology and dermatology, optical coherence tomography (OCT) and reflective confocal microscopy (RCM) are used and successfully transformed due to their deep penetration and non-invasive properties. However, these reflected light-based methods provide relatively limited cellular information in visceral organ imaging. In addition, photoacoustic microscopy (PAM) uses the intrinsic spectral absorption properties of different biomolecules to effectively identify and image a variety of biological targets, showing good results in breast cancer screening and vascular imaging. In addition, nonlinear microscopy (NLM) (such as coherent Raman scattering, multiphoton absorption, and second / third harmonic generation) can achieve high-resolution and label-free visualization of a variety of biological processes in an undisturbed and non-destructive manner, showing wide application in oncology research (such as tumor infiltration and growth). However, these methods still face challenges in screening large surgical samples in a short time due to the need for continuous beam scanning. In addition, the lasers used in PAM or NLM are usually bulky and expensive, which increases facility requirements and costs.

[0007] A recently proposed method, computational high-throughput autofluorescence microscopy with patterned illumination (CHAMP), enables rapid and label-free histological imaging of thick and untreated tissues with micrometer-scale resolution. This technique is particularly suitable for intraoperative SMA, providing surgeons with immediate feedback to optimize adjuvant therapy. However, the depth of field (DOF) of CHAMP is limited to less than 80 μm, which is insufficient for rough excised tissues with large surface irregularities, resulting in surgical margins frequently moving in and out of focus during imaging. In addition, CHAMP achieves optical sectioning by exploiting the shallow penetration depth of deep UV light, which may vary greatly among different tissue types, so CHAMP images may deviate from conventional slide-based FFPE histology results.

[0008] Therefore, there is a need for an improved method and system that reduces or eliminates the disadvantages and problems described above. Summary of the invention

[0009] Therefore, a first aspect of the present invention provides a histological imaging and analysis of biological tissues based on microscopy with ultraviolet single plane illumination (MUSI). The present invention MUSI uses the intrinsic fluorescence from biological tissues as imaging contrast, eliminates out-of-focus fluorescence through selective plane illumination, and the generated signal is detected from orthogonal directions. This dual-axis configuration decouples the illumination and detection paths to overcome the inherent trade-off between long DOF and high spatial resolution.

[0010] Unlike other non-destructive optical imaging techniques, such as UV surface excitation microscopy (MUSE), structured illumination microscopy (SIM), and light sheet microscopy (LIM), MUSI does not require fluorescent labeling before imaging. Although some label-free microscopy techniques, such as UV photoacoustic microscopy (UV-PAM) and nonlinear microscopy (NLM), including multiphoton microscopy (MPM), stimulated Raman scattering microscopy (SRS), and second / third harmonic generation (S / THG), can detect the intrinsic response of molecules / cells to excitation irradiation, such as thermoelastic expansion induced by light energy absorption by UV-PAM, intrinsic autofluorescence by MPM, molecular vibration by SRS, and non-centrosymmetric orientation by SHG, in order to achieve cell phenotyping / classification, such point scanning-based techniques are difficult to achieve high-throughput imaging. Although reflection-based imaging techniques such as full-field optical coherence tomography (FF-OCT), reflective confocal microscopy (RCM), and dark-field reflective ultraviolet microscopy (DRUM) are capable of rapidly outputting label-free images when detecting back-reflected light from structures with different refractive indices, the important diagnostic features displayed by such techniques may differ significantly from clinical standard methods due to limited contrast, resolution, and signal-to-noise ratio. In particular, DRUM relies on sensing diffuse reflection, while the present invention based on MUSI relies on the intrinsic absorption of light and uses autofluorescence from the sample as image contrast. Compared with DRUM, MUSI has a longer DOF, so MUSI is more suitable for scanning tissues with irregular surfaces. In addition, the optical sectioning capability of DRUM depends on the different penetration depths of UV light in different tissues, while MUSI has no such constraints, so that MUSI can display stable optical sectioning characteristics in different types of tissues. The present invention may also be applicable to in vitro and in vivo imaging of target tissues, while DRUM can only be used for in vitro imaging.

[0011] Thus, in a first aspect, the imaging system comprises:

[0012] a movable sample holding platform including a sample holder and a liquid holder, the liquid holder having a prism-like structure;

[0013] a deep UV excitation source that provides deep UV irradiation;

[0014] a first plurality of optical elements for generating a light sheet from deep UV illumination provided by a deep UV excitation source and directing the light sheet perpendicularly through the sides of the prismatic structure of the liquid holder toward the bottom side of the sample holder to impinge on the bottom surface of the sample at an angle of incidence relative to a vertical axis of the sample holder or an optical axis of the deep UV illumination;

[0015] a second plurality of optical elements for receiving light emissions from the sample excited by the light sheet, including emitted fluorescent signals, at detection angles relative to a vertical axis of the sample holder; and

[0016] an optical detection unit for detecting and processing light emissions received by the second plurality of optical elements;

[0017] Wherein the incident angle and the detection angle are substantially the same in magnitude, and the beam path of the light sheet is perpendicular (or orthogonal) to the beam path of the light emission from the sample.

[0018] In certain embodiments, the movable sample holding platform is connected to a three-dimensional (3D) translation stage.

[0019] In certain embodiments, the sample holder includes a membrane for supporting the sample.

[0020] In certain embodiments, the liquid in the liquid holder of the movable sample holding platform comprises water or a mixture of water having UV transparency.

[0021] In certain embodiments, the first plurality of optical elements includes at least a first filter, a pair of lenses, a slit aperture, and a cylindrical lens.

[0022] In some embodiments, the first filter is a bandpass filter.

[0023] In certain embodiments, the pair of lenses is a pair of UV grade convex lenses.

[0024] In certain embodiments, the slit aperture is an adjustable slit aperture.

[0025] In some embodiments, the cylindrical lens is a UV cylindrical lens for generating a Gaussian light sheet.

[0026] In some embodiments, the light sheet illuminates the bottom surface of the sample at an angle of incidence of 45° relative to the vertical axis of the sample holder, and the average energy flux of the light sheet illuminates the bottom surface of the sample in accordance with the American Conference of Governmental Industrial Hygienists ( ) safety UV radiation threshold specified.

[0027] In certain embodiments, the liquid retainer is disposed below the sample retainer in the movable sample retaining platform.

[0028] In certain embodiments, the sample is supported by a UV transparent film fixed at the base of the sample holder so that the light sheet can reach the bottom surface of the sample held in the sample holder.

[0029] In certain embodiments, the UV transparent film is made of a highly transparent thermoplastic, including but not limited to polyethylene or any UV transparent material.

[0030] In some embodiments, the prism-like structure of the movable sample holding platform includes two UV transparent windows arranged at two opposite sides of the prism-like structure, each UV transparent window allowing a light sheet from a deep UV excitation source to enter the movable sample holding platform, or allowing light emissions (including fluorescence signals) from the sample after being excited by the light sheet to leave the movable sample holding platform.

[0031] In certain embodiments, the sample is biological tissue (such as freshly excised tissue or biopsy tissue from a living organism) containing normal and abnormal tissue.

[0032] In certain embodiments, biological tissues are rich in endogenous fluorophores, including but not limited to reduced nicotinamide adenine dinucleotide (NADH), structural proteins (such as collagen and elastin), aromatic amino acids (such as tryptophan and tyrosine), and heterocyclic compounds (such as flavins and lipofuscin).

[0033] In certain embodiments, the second plurality of optical elements includes at least a UV objective, a second filter, and an infinity-corrected lens.

[0034] In certain embodiments, the UV objective is an achromatic UV objective.

[0035] In some embodiments, the second filter is a long pass filter.

[0036] In some embodiments, the infinity-corrected lens is an infinity-corrected tube lens.

[0037] In some embodiments, the optical detection unit includes a plurality of CMOS image sensors.

[0038] In certain embodiments, the plurality of CMOS image sensors of the optical detection unit include, but are not limited to, scientific complementary metal oxide semiconductor (sCMOS) sensors.

[0039] In some embodiments, the 3D translation stage is a movable sample holding platform, providing at least two scanning directions. First, the sample is translated along a first scanning direction at a constant speed, so that the light sheet can pass through the UV transparent film to reach the bottom surface of the sample. In addition, the sample can also be translated along a transverse axis relative to the first scanning direction toward a secondary scanning direction, so as to scan the entire maximum surface area of ​​the sample to obtain the complete surface morphology of the sample.

[0040] A second aspect of the present invention provides a method for imaging biological tissue in a label-free, unprocessed manner using the imaging system of the present invention according to the first aspect and various embodiments described herein to output a histology-like image showing a two-dimensional or three-dimensional contour of the biological tissue. The method comprises at least the following steps:

[0041] (1a) placing a biological tissue into a sample holder of a movable sample holding platform of an imaging system, wherein a relatively flat side of the biological tissue faces a UV transparent film of the sample holder;

[0042] (1b) immersing a sample holder holding the biological tissue into a liquid holder of a movable sample holding platform filled with liquid;

[0043] (1c) adjusting one or more of the first plurality of optical elements to direct the light sheet thus generated onto the bottom surface of the sample at an incident angle of 45° relative to the vertical axis of the sample holder, or in other words, the incident beam path of the light sheet is perpendicular to the UV transparent window on one side of the prism-like structure at the bottom of the liquid holder;

[0044] (1d) adjusting one or more of the second plurality of optical elements to receive light emissions emitted by the sample at a detection angle of 45° relative to a vertical axis of the sample holder or orthogonal to an incident beam path of the light sheet, the light emissions comprising a fluorescent signal;

[0045] (1e) activating a deep UV excitation source to generate UV radiation toward the first plurality of optical elements, which is then modulated by the first plurality of optical elements to generate a light sheet, and directing the light sheet to a first point of the sample at an incident angle of 45° relative to a vertical axis of the sample holder through a UV transparent window of a prism-like structure at a bottom of the liquid holder;

[0046] (1f) receiving light emissions from the sample including fluorescent signals emitted by the light sheet through a second plurality of optical elements;

[0047] (1g) detecting, by an optical detection unit including a plurality of CMOS image sensors, received light emissions including emitted fluorescent signals from the second plurality of optical elements;

[0048] (1h) moving a movable sample holding platform driven by a 3D translation stage from a first point of the sample to subsequent points on the same plane in the main scanning direction, and repeating steps (1e) to (1g) in each subsequent point until reaching the last point on the same plane in the main scanning direction to form a first image strip;

[0049] (1i) moving the movable sample holding platform transversely relative to the first scanning direction toward a secondary scanning direction, and repeating steps (1e) to (1h) to form subsequent image strips adjacent to the first image strip until the entire maximum surface area of ​​the biological tissue is scanned; and

[0050] (1j) The image data detected by the optical detection unit is processed by an image processing module to reconstruct the geometric shape of the biological tissue, and one or more images representing the overall geometric shape of the reconstructed biological tissue are output.

[0051] In some embodiments, the light sheet has a wavelength of approximately 266 nm.

[0052] In certain embodiments, the penetration depth of the light sheet can be up to about 30 μm from the bottom surface of the biological tissue.

[0053] In some embodiments, the sample is translated through the light sheet at a constant speed of 250 μm / s along the main scanning direction, and an image of each sample in each image strip is recorded at 250 frames / s along the main scanning direction at the optical detection unit with a sampling pitch of 1 μm / pixel.

[0054] In some embodiments, after a first image strip is formed along a main scanning direction, the movable sample holding platform is moved laterally relative to the main scanning direction toward a secondary scanning direction so that a subsequent image strip adjacent to the first image strip is formed along the main scanning direction, wherein the movable sample holding platform adopts a serpentine or column-by-column scanning movement path to cover the entire maximum surface area of ​​the biological sample.

[0055] Alternatively, the movement path of the movable sample holding platform for covering the entire maximum surface area of ​​the biological sample can be any other pattern as long as the entire maximum surface area of ​​the biological sample is imaged and complies with the freedom of movement of the 3D translation stage.

[0056] In some embodiments, about 10% or less of the features of each pair of adjacent image strips overlap.

[0057] In some embodiments, the image processing module can be a stand-alone computer processor or a network connected to the imaging system of the present invention as part of the imaging system of the present invention, and the image processing module contains one or more image processing algorithms for processing a series of image strips of detected image data.

[0058] In certain embodiments, distortion in the raw image of the image strip caused by the detection angle being non-parallel to the vertical axis of the sample holder may be corrected by one of the image processing algorithms.

[0059] In some embodiments, the detected surface features in the image data are extracted by another image processing algorithm until all image swaths are processed.

[0060] In certain embodiments, one of the image processing algorithms for correcting distortion in the raw image of the image strip due to the detection angle being non-parallel to the vertical axis of the sample holder is written in MATLAB.

[0061] In some embodiments, one of the image processing algorithms used for surface feature extraction is an extended DOF algorithm

[0062] In certain embodiments, the extended DOF algorithm is introduced into MATLAB via a Fiji plugin.

[0063] In some embodiments, other image processing algorithms for combining the extracted surface features of one image strip to the extracted surface features of an adjacent image strip are introduced by the Fiji mesh stitching plugin.

[0064] A third aspect of the present invention provides a method for in vivo imaging of a biological tissue or structure of a subject using the imaging system of the present invention. The method comprises:

[0065] (2a) placing the subject under anesthesia on the inner surface of the transparent base of the sample holder so that the target biological tissue is as close as possible to the UV transparent film of the sample holder of the movable sample holder platform, and the light beam of the light sheet enters through the UV transparent window of the prism-like structure of the sample holder platform;

[0066] (2b) adjusting one or more of the first plurality of optical elements so that the light sheet generated thereby impinges upon a bottom surface of the sample at an angle of incidence of 45° relative to a vertical axis of the sample holder;

[0067] (2c) adjusting one or more of the second plurality of optical elements so that light emissions including fluorescent signals emitted from the sample are detected at a detection angle of 45° relative to a vertical axis of the sample holder or at a detection angle orthogonal to an incident beam path of the light sheet;

[0068] (2d) adjusting one or more of the second plurality of optical elements to receive light emissions including a fluorescent signal emitted from the sample at a detection angle of 45° relative to a vertical axis of the sample holder or orthogonal to an incident beam path of the light sheet;

[0069] (2e) activating a deep UV excitation source to generate UV radiation toward the first plurality of optical elements, which is then modulated by the first plurality of optical elements to generate a light sheet, and directing the light sheet to a first point of the sample at an incident angle of 45° relative to a vertical axis of the sample holder through a UV transparent window of a prism-like structure at a bottom of the liquid holder;

[0070] (2f) receiving light emissions including emitted fluorescent signals from the sample excited by the light sheet through a second plurality of optical elements;

[0071] (2g) detecting, by an optical detection unit including a plurality of CMOS image sensors, received light emissions from the second plurality of optical elements including emitted fluorescent signals;

[0072] (2h) moving a movable sample holding platform driven by a 3D translation stage from a first point of the sample to subsequent points on the same plane in the main scanning direction, and repeating steps (2e) to (2g) in each subsequent point until reaching a final point on the same plane in the main scanning direction to form a first image strip;

[0073] (2i) moving the movable sample holding platform transversely relative to the first scanning direction toward a secondary scanning direction, and repeating steps (2e) to (2h) to form subsequent image strips adjacent to the first image strip until the entire maximum surface area of ​​the biological tissue is scanned; and

[0074] (2j) Processing the image data detected by the optical detection unit through the image processing module to reconstruct the geometric shape of the biological tissue, and outputting one or more images representing the overall geometric shape of the reconstructed biological tissue.

[0075] In certain embodiments, the removable sample holding platform has an open-top configuration to allow convenient loading of samples of any size and thickness from the top.

[0076] In certain embodiments, subjects include humans and non-human animals.

[0077] In certain embodiments, biological tissues or structures include freshly excised or intact normal and abnormal tissues, as well as partial or entire internal organs.

[0078] Other possible arrangements described in the second aspect that may be applicable may be various embodiments of the method in the third aspect.

[0079] The present invention is provided to introduce some concepts that will be further described in the following detailed description in a simplified form. The present invention is neither intended to identify the key features or essential features of the claimed subject matter, nor is it intended to be used to help determine the scope of the claimed subject matter. As shown in the following examples, other aspects of the present invention are disclosed. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] The same reference numerals in the accompanying drawings represent the same or similarly characterized elements, and the accompanying drawings include diagrams of certain embodiments to further illustrate and clarify the above and other aspects, advantages and features of the present invention. It will be understood that these drawings depict embodiments of the present invention and are not intended to limit the scope of the present invention. The present invention will be described and explained with additional specificity and detail through the use of the accompanying drawings, in which:

[0081] Figure 1AThe inventive system based on MUSI according to some embodiments is schematically depicted.

[0082] Figure 1B Schematically depicted as Figure 1A The structure and working mechanism of the system of the present invention are shown in FIG.

[0083] Figure 1C Shown by Figure 1A An example of an image processing method for a raw image of a tissue sample acquired by a system.

[0084] Figure 1D FIG. 1 shows a result image of the entire tissue sample obtained by processing multiple image strips of the original image and extracting the surface image and stitching them together according to some embodiments, wherein the original image and the extracted surface image are obtained by Figure 1C The image processing method described in was obtained.

[0085] Figure 2A An image showing a side view of illumination from a light sheet generated by the system of the present invention according to some embodiments.

[0086] Figure 2B Shown as Figure 2A The intensity distribution along the lateral direction at the focus of the light sheet is shown in FIG.

[0087] Figure 2C Shown as Figure 2A The beam radius distribution along the axial direction at the focus of the light sheet is shown in FIG.

[0088] FIG. 3A to FIG. 3J Label-free in vitro imaging of freshly excised mouse tissue by the system of the present invention according to certain embodiments is shown, wherein: Figure 3A MUSI images of fresh mouse liver are shown, and a photograph of the sample is shown in the inset at the lower left; Figure 3B Shown as Figure 3A An enlarged area of ​​a region and its corresponding H&E staining image are shown in; Figure 3C Shown as Figure 3A An enlarged area of ​​another region and its H&E staining image are shown in FIG. Figure 3D Shown from Figure 3C Distribution of extracted kernel features, with Wilcoxon rank sum test performed on each group, where n = 50 for each distribution. In all cases, significance was defined as p*≤0.05; Figure 3E MUSI images of fresh mouse brain are shown, and a photograph of the sample is shown at the lower left inset; Figure 3F Shown as Figure 3E An enlarged area of ​​a region and its corresponding H&E staining image are shown in; Figure 3G Shown as Figure 3E An enlarged area of ​​another region and its H&E staining image are shown in FIG. Figure 3H MUSI images of fresh mouse kidneys are shown, and photographs of the specimens are shown at the lower left inset; Fig. 3I Shown as Figure 3H One of the enlarged areas shown in and its corresponding H&E staining image; Figure 3J Shown as Figure 3H The enlarged area of ​​another area and its H&E staining image are shown in FIG. Figure 3A , Figure 3E and Figure 3H The scale bar used in is 1 mm; Figure 3B , Figure 3C , Figure 3F , Figure 3G , Fig. 3I and Figure 3J The scale bar used in the figure is 100 μm; GL denotes granular layer; ML denotes molecular layer; WM denotes white matter.

[0089] FIG. 4A to FIG. 4B Label-free in vivo imaging of intact mouse tissue in vivo by the system of the present invention according to certain embodiments is shown, wherein: Figure 4A MUSI images showing living brain tissue; Figure 4B MUSI images of living kidney tissue are shown.

[0090] Figure 5 Imaging of human lung adenocarcinoma tissue using the system of the present invention according to certain embodiments is shown.

[0091] FIG. 6A to FIG. 6L The imaging of human lung adenocarcinoma tissue by the system of the present invention according to certain embodiments is shown, wherein: Fig. 6A MUSI and H&E staining images of a lung specimen with acinar predominant adenocarcinoma are shown, and a photograph of the specimen is shown at the lower left inset; Figure 6B Shown as Fig. 6A An enlarged picture of an area shown in and its corresponding H&E staining image; Figure 6C Shown as Fig. 6A An enlarged picture of another area shown in FIG. 1 and its corresponding H&E staining image; Fig.6D Shown as Fig. 6A An enlarged picture of another area and its corresponding H&E staining image are shown in FIG. Fig. 6E MUSI and H&E staining images of a papillary adenocarcinoma lung specimen are shown, and a photograph of the specimen is shown at the lower left inset; Fig. 6F Shown as Fig. 6E An enlarged picture of an area shown in and its corresponding H&E staining image; Figure 6G Shown as Fig. 6EAn enlarged picture of another area shown in FIG. 1 and its corresponding H&E staining image; Figure 6H Shown as Fig. 6E An enlarged picture of another area shown in FIG. 1 and its corresponding H&E staining image; Fig.6I MUSI and H&E staining images of a lung specimen with predominantly micropapillary adenocarcinoma are shown, and a photograph of the specimen is shown at the lower left inset; Figure 6J Shown as Fig.6I An enlarged picture of an area shown in and its corresponding H&E staining image; Figure 6K Shown as Fig.6I An enlarged picture of another area in the region and its corresponding H&E staining image are shown in FIG. Figure 6L Shown as Fig.6I FIG. 4 shows an enlarged area of ​​another area in the region and its corresponding H&E staining image, wherein Fig. 6A , Fig. 6E and Fig.6I The scale bar in the figure is 1 mm; FIG. 6B to FIG. 6D , FIG. 6F to FIG. 6H and Figures 6J to 6L The scale bar in the larger panels is 100 μm; Figure 6B , Figure 6C , Fig. 6F and Figure 6G The scale bar in the smaller panels is 50 μm; TN, tumor cell nucleus; LN, lymphocyte nucleus; FC, fibrovascular core; AS, alveolar space; MC, micropapillary cluster; AP, anthrax pigment.

[0092] 7A to 7D The imaging of human skin tissue by the system of the present invention according to certain embodiments is shown, wherein: Fig. 7A MUSI of human skin tissue sample and its corresponding H&E staining image; Figure 7B Shown as Fig. 7A , showing an enlarged picture of one area in the region and its corresponding H&E staining image; Figure 7C Shown as Fig. 7A An enlarged picture of another area shown in FIG. 1 and its corresponding H&E staining image; Fig.7D Shown as Fig. 7A An enlarged picture of another area and its corresponding H&E staining image are shown in FIG. Fig. 7A The scale bar in the figure is 1 mm; FIG. 7B to FIG. 7D The scale bar in is 100 μm.

[0093] Figure 8 is a flow chart outlining the imaging methods of the present invention for in vitro and in vivo imaging of target tissues according to certain embodiments.

[0094] Those skilled in the art will appreciate that elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale. DETAILED DESCRIPTION

[0095] It is obvious to those skilled in the art that modifications including additions and / or substitutions may be made without departing from the scope and spirit of the invention. Specific details may be omitted so as not to obscure the invention; however, this disclosure is written to enable those skilled in the art to practice the teachings herein without undue experimentation.

[0096] The present invention provides a rapid, label-free, non-destructive method and corresponding system for imaging untreated biological tissue in vitro or even in vivo, so that compared with conventional clinical standard methods (i.e., conventional histological evaluation including H&E staining), the method and system can provide medical practitioners with real-time qualitative evaluation means for locating the location of tissue to be removed during surgery and determining the extent of tissue to be removed. In addition to the conventional FFPE approach, the present invention also has great potential as an auxiliary diagnostic tool for pathologists or clinicians to evaluate biopsy histology, thereby helping to determine the patient's disease prognosis.

[0097] The following examples provide some implementation details of the imaging method and corresponding system of the present invention, which should not be considered as limiting the scope of the present invention.

[0098] Example

[0099] (A) Configuration and imaging parameters of the open-top MUSI system:

[0100] Go to Figure 1A to Figure 1B , schematically depicts an open-top system of MUSI according to certain embodiments. System 100 includes a movable sample holding platform 110, which includes a sample (or sample) holder and a liquid holder, wherein a prism-like structure is configured at the bottom side of the liquid holder. The incident and detection (including emission) light beams are allowed to enter and leave the sample holding platform vertically. From Figure 1BAs can be seen from the enlarged view of the sample holding platform shown in , UV transparent quartz windows are installed on both sides of the prism-like structure, so that the incident light beam originating from the deep UV excitation source 120 and modulated by the first plurality of optical elements 130 into a light sheet shape can vertically enter the liquid retainer of the sample holding platform. The incident light beam then reaches the target tissue held in the sample holder and excites the tissue to emit light including fluorescent signals. Ultimately, these light signals leave the liquid retainer and are subsequently received by the second plurality of optical elements 140. The liquid-filled prism-like structure is configured to be disposed below the sample holder or partially serve as the base of the sample holder to mitigate imaging aberrations caused by the 45° incident angle of the light sheet incident light beam and detection at the air-retainer-tissue interface (such as Figure 1B In order to provide design freedom for the prismatic structure, according to some embodiments, a 3D printing method can be used. Other conventional methods (such as injection molding) can also be used to manufacture a polyhedral structure with a liquid retainer, so that the prismatic structure is combined as part of the base or bottom surface of the liquid retainer.

[0101] In order to allow the light sheet to reach the tissue sample held at the sample holder, the bottom of the sample holder is made of a UV transparent material (e.g., a UV transparent polyethylene film). In some embodiments, the UV transparent polyethylene film has a thickness of about 50 μm. In an exemplary embodiment, the liquid is water or a mixture of water and a reagent that can promote visualization of tissue structures under the excitation of the light sheet generated by the deep UV excitation light source. In some embodiments, the light sheet can be a Gaussian light sheet. Figure 2A The experimental profile of illuminating a light sheet in a dilute solution of acridine orange is shown, where the light sheet generated by the present system has an illumination wavelength (λ of approximately 266 nm). illu ) and an effective numerical aperture (NA) of approximately 0.03 eff ). Figure 2B to Figure 2C Shown along Figure 2A The normalized intensity of the light sheet at the focus in the transverse direction and the beam radius distribution in the axial direction are shown in FIG. Figure 2A As can be seen in Figure 1, a 3mm wide, waist radius (w0) of 2.8μm, DOF (2Z R) is a Gaussian beam of about 200 μm: within the first plurality of optical elements, the UV radiation generated by the deep UV excitation source 120 is spectrally filtered by a bandpass filter F1 (FF01-300 / SP-25, Semrock Inc.) and expanded by a pair of convex lenses L1 and L2 (LA4647-UV and LA4874-UV, Thorlabs Inc.), wherein there is a reflector between L1 and L2, which is configured to guide the UV beam filtered by F1 and incident on L1 toward subsequent optical elements including L2. After the beam is expanded by L2, it propagates through an adjustable slit aperture SA (VA100C, Thorlabs Inc.) and outputs a beam having a size of 3 mm×6 mm, which is then focused by a UV cylindrical lens CL (LJ4395RM, f=100 mm, Thorlabs Inc.). Both the incident beam path and the detection (including emission) beam path are vertically focused through two corresponding UV transparent quartz windows (20 mm×20 mm) bonded on the two sides of the water prism. After passing through the first UV transparent quartz window, the generated light sheet will illuminate the bottom surface of the sample through the water prism at a 45° incident angle (relative to the virtual vertical axis of the sample holder) with an average energy flux of 2 mJ / cm 2 , this is The invention relates to a method for treating a tissue sample with a light sheet having an incident angle of 45°. The method further ...

[0102] For the detection side, the second plurality of optical elements 140 is optimized so that it can receive the maximum intensity of the light emission signal from the tissue sample. In this configuration, the achromatic UV objective lens OL (LMU-5X-NUV, NA = 0.12, Thorlabs Inc) is parallel to the optical path of the detection beam from the tissue sample. The long pass filter F2 (BLP01-325R-25, Semrock Inc.) is set to the same orientation as the OL and is set at the rear end of the OL. The detection beam filtered by F2 is then refocused by the infinity-corrected tube lens TL (TTL200-A, Thorlabs Inc.) and is finally imaged by the sCMOS camera 150 (PCO edge4.2, 2048×2048 pixels, PCO Inc.).

[0103] In order to be able to scan the maximum surface area of ​​the tissue sample, the sample holding platform 110 is connected to a 3D translation stage 160 (L-509.20SD00, PI miCos GmbH), which supports translational movement of the sample holding platform 110 along at least two axes. In some embodiments, the maximum surface area of ​​the tissue sample to be scanned can be up to 5 cm×5 cm. The sample can be exposed to a static light sheet at a constant speed of 250 μm / s along the main scanning direction (i.e., Figure 1B and Figure 1C The image can be sampled by a sCMOS camera 150 with a sampling pitch of 1 μm / pixel at 250 frames / second. The image height (h) can be adjusted according to the surface irregularities of the imaged sample, where Figure 1A In the example shown in , the maximum tolerance is about 200 μm. It is worth noting that in the case of in vitro imaging, the use of a high-power UV source can increase the imaging speed by at least an order of magnitude. This is due to the fact that the intrinsic fluorescence signal is stronger under relatively high-power UV radiation / excitation sources (only 2 mJ / cm was used in this example). 2 sCMOS cameras can use higher frame rates because the more intense UV radiation can generate images with sufficient signal-to-noise ratio at short exposure times.

[0104] The tissue sample (e.g. Figure 1C After that, the sample is moved laterally along the y-axis at intervals of 1.8 mm and scanned in a serpentine pattern (e.g., a vertical serpentine pattern). After stitching multiple image strips by a corresponding algorithm for stitching image strips or grids with a large FOV, the following is obtained: Figure 1D The original images were stored at a 16-bit depth and transferred to a local workstation equipped with a high-speed solid-state drive (970EVO Plus, SAMSUNG Inc.) via the CameraLink interface at a streaming rate of 2040 pixels (w) × 200 pixels (h) × 2 bytes × 250 frames / sec = 0.2 GB / sec. Self-designed software in LabVIEW was used to synchronize image acquisition with the sample scanning movement of the 3D translation stage. With the above settings, MUSI was able to obtain images at a resolution of 0.5 mm. 2 The scanning speed of 1.5 μm / s and image resolution of 1.5 μm enable rapid imaging of freshly excised and unprocessed tissues (in vitro or in vivo), providing histology-like images with sufficient diagnostic features for medical practitioners, surgeons and pathologists during intraoperative, postoperative or treatment decisions.

[0105] like Figure 1CAs shown, the original image of the tissue sample is first acquired in the main scanning direction to form an image strip, and then scanned in a serpentine pattern to cover the maximum surface area of ​​the tissue sample to generate multiple image strips. Since the original image is acquired at an incident angle of 45° relative to the tissue surface, the tissue geometry reconstructed directly using the original data set loaded into MATLAB (MathWorks, Inc) will be distorted. To correct this distortion, the original data set is sheared 45° in the xz plane to form an oblique prismatic data set, as shown in Figure 1C After distortion correction, the extended DOF algorithm was applied to the cropped dataset via the Fiji plugin to extract the complete tissue surface for each image strip ( Figure 1D The extracted surface images of adjacent image strips were then registered and stitched using the Fiji grid stitching plugin to form a large area tissue image ( Figure 1D In some embodiments, the cropping of the original image, the correction of distortion caused by cropping, the extraction of the complete surface image from each image strip generated, and the registration and stitching of multiple adjacent image strips to form a large area tissue image are all handled by micro scripts in the MATLAB-Fiji interface. The 1-cm 2 The processing time of the tissue surface is about 10 minutes. In some embodiments, without affecting the axial resolution, the DOF can be further extended by using a digital scanning light sheet system with a non-diffracted beam, thereby further improving the system's tolerance to irregular tissue surfaces. In some embodiments, certain components of the second plurality of optical elements at the detection side / arm of the system of the present invention can be changed. For example, a fixed objective lens can be replaced by a switchable objective lens with different resolutions to achieve multi-scale tissue structure imaging from submicron to macroscopic spatial scales.

[0106] In order to extract the distribution of cell nucleus features in the target tissue, the tissue image can be segmented by a Fiji plug-in (e.g., TrainableWeka Segmentation plug-in), followed by image binarization and analysis to obtain the cross-sectional area and centroid of each cell nucleus. The shortest adjacent distance between adjacent cell nuclei can be calculated by the centroid of each extracted cell nucleus as the intercellular distance. In certain embodiments, a virtual staining model can be trained using a deep learning neural network to generate pseudo-histochemical staining MUSI images, thereby assisting medical practitioners (such as pathologists or surgeons) to perform more efficient and accurate intraoperative or postoperative clinical evaluations.

[0107] (B) In vitro imaging of mouse tissues:

[0108] Go to FIG. 3A to FIG. 3J , the system of the present invention was used to perform in vitro imaging of mouse visceral organs, and the results were compared with the results of conventional FFPE histology (hematoxylin and eosin (H&E) staining) preparation and imaging. First, fresh visceral organs (such as liver, brain, and kidney) were obtained immediately after killing mice (C57BL / 6 type). After extraction, these organs were manually cut into 3mm to 5mm thick tissue slices for imaging. After imaging, all samples were processed using conventional histological methods to obtain H&E staining images ( Figure 3B , Figure 3C , Figure 3F , Figure 3G , Fig. 3I and Figure 3J for comparison.

[0109] Freshly excised mouse liver ( FIG. 3A to FIG. 3C ),brain( Figures 3E to 3G ) and kidney( Figures 3H to 3J ) MUSI images of tissue slices show that the light sheets generated by the system of the present invention have a UV penetration depth of approximately 5 μm to 30 μm, which depends on the microstructure of the tissue, the degree of tissue scattering, and the density / distribution of certain endogenous fluorophores of different types / states of tissue. For example, UV penetrates to different degrees in areas with densely arranged cell nuclei (such as primary tumors), areas with abundant fat droplets (such as subcutaneous tissue), or areas with large intercellular spaces (such as alveoli). For normal mouse tissues (kidney, brain, liver, lung, and spleen), kidney and brain generally exhibit higher intrinsic fluorescence intensity than liver and lung tissues after absorbing UV, while spleen tissue emits the least fluorescence; the intensity of certain metabolic enzymes (such as NADH and FAD) in tumors is generally higher than that in normal tissues, which is promising as tumor-specific biomarkers.

[0110] It is important to note that FFPE thin sections do not completely replicate the surface imaged by MUSI due to different imaging thicknesses and tissue deformation during sample processing. However, their cell morphology is similar. The morphology of hepatocytes in mouse liver is better characterized 10 μm below the tissue surface, although MUSI has difficulty identifying sinusoidal capillaries ( Figure 3B ) structure, but their morphological H&E staining images still have a high degree of similarity.

[0111] Nuclear features (such as cross-sectional area and intercellular distance) were extracted from both MUSI and H&E stained images for comparison. Figure 3C The nuclear features of 50 hepatocytes were extracted from the MUSI and H&E staining images, and the statistical results ( Figure 3D) showed that the cell nucleus features extracted by MUSI were highly consistent with those extracted by clinical standard methods.

[0112] In the mouse cerebellum, both MUSI and H&E staining images clearly show a clear interface between the molecular layer and the granular layer, as well as the Purkinje cells in between. Figure 3F and Figure 3G ). Similarly, MUSI images were able to identify renal tubules in mouse kidneys very well ( Fig. 3I ) and glomerular capsule ( Figure 3J )feature.

[0113] (C) In vivo imaging of mouse tissues:

[0114] In this example, for kidney imaging, a small incision was made on the left side of the mouse abdomen to expose the kidney, which was then secured and laid flat on a holder for imaging from below. For brain imaging, the mouse was secured in a supine position on a holder and a 5mm x 5mm cranial window was opened in the skull to expose the target brain region for imaging. Figure 4A and Figure 4B MUSI images of living brain tissue and living kidney tissue are shown respectively. Structural details of the corresponding tissues (such as blood vessels, membrane lipids and tubules) can be seen from the magnified view. Figure 4A In the image, microvascular structures and cell nuclei on the surface of the cerebral cortex are clearly visible, which helps to detect tumor margins in real time during neurosurgery. Due to the presence of highly fluorescent cytoplasmic lipofuscin and urinary casts, renal tubules can be clearly imaged with a high signal-to-noise ratio ( Figure 4B Although the penetration depth of MUSI is limited by the strong light scattering of tissue, the invention is still a valuable tool for rapid assessment of intact tissues without damage, especially in cases where biopsy is contraindicated.

[0115] (D) In ​​vitro imaging of human tissues:

[0116] In this example, lung adenocarcinoma samples were used to verify the capabilities of the MUSI system of the present invention. Adenocarcinoma is the most common type of non-small cell lung cancer, usually originating from mucosal glands and often occurring in the peripheral areas of the lungs. Human lung cancer tissue samples (n=11) were obtained from patients who underwent radical lung cancer surgery at Queen Mary Hospital. After lobectomy, lung cancer tissue was cut from the resected lung lobe, then fixed with formalin and transported to the laboratory for MUSI imaging. After imaging, all samples were histologically processed according to standard procedures to obtain H&E stained images for comparative analysis. Figure 5 A collection of representative MUSI images of human lung adenocarcinoma samples generated by the system of the present invention and their corresponding H&E staining images are shown. MUSI images can show different subtype microstructures of lung adenocarcinoma (such as lepidic, alveolar, papillary, micropapillary and solid). FIG. 6A to FIG. 6D Showing a predominantly alveolar adenocarcinoma sample, MUSI can identify irregularly shaped glands within a fibrotic stroma ( Figure 6B and Figure 6C ).from Figure 6C In the images, some glands can be observed arranged as solid clusters of tumor cells, and the luminal structure is less obvious. Fig.6D It is shown that tissue fragments generated by cell rupture can be clearly distinguished by MUSI, and their autofluorescence is significantly enhanced.

[0117] FIG. 6E to FIG. 6H A papillary-predominant lung adenocarcinoma specimen was imaged by MUSI and the results were compared with the clinical standard method H&E staining images. Fig. 6E In the figure, the dotted line clearly outlines the interface between the normal area and the cancerous area. Fig. 6F It is shown that the nuclei of tumor cells (TN) are clearly visible at a penetration depth of approximately 10 microns below the surface of the tissue sample. Figure 6G It is shown that both MUSI and H&E staining images show finger-like papillary structures in which tumor cells are arranged on the surface of branching fibrovascular cores (FC). Figure 6H It is shown that in normal areas, the alveolar structure (AS) is a normal morphology with large air cavities.

[0118] FIG. 6I to FIG. 6K The image shows a micropapillary-predominant adenocarcinoma specimen confirmed by MUSI and clinical standard methods (H&E staining). In it, tumor cell clusters (MCs) were found floating and isolated in the alveolar spaces lacking fibrovascular cores ( Figure 6J and Figure 6K ). Figure 6L An area with a large number of tumor-infiltrating lymphocytes (LN) and anthrax pigment (AP) is shown, which is highly consistent with the H&E staining image.

[0119] 7A to 7D Figure 2 shows images of human skin obtained by MUSI and clinical standard methods (H&E staining), in which microstructures (such as sweat glands ( Figure 7B ), the arterial cavity filled with red blood cells ( Figure 7C ) and adipose tissue ( Fig.7D )) are clearly presented by MUSI in a label-free and non-destructive manner, and are highly consistent with the H&E staining images.

[0120] Figure 8The imaging method of the present invention is summarized. First, the target tissue (or object) is placed on the inner surface of the sample holder base formed by the UV transparent film. The sample holder is then immersed in the liquid holder of the movable sample holding platform (801). In order to reduce the scattering effect of light, it is recommended to close the relatively flat surface of the target tissue to the inner surface of the sample holder base. In some embodiments, the target tissue is freshly cut from the object for in vitro imaging by MUSI, or the surface of the target tissue is directly exposed to the light sheet for in vivo imaging. In the embodiment of in vivo imaging, since the target tissue (such as the surface of the internal organ) and its adjacent tissues will directly contact the inner surface of the sample holder base, the sample holder and liquid holder of the movable sample holder platform must be disinfected. Then, the sample holder together with the target tissue is immersed in the liquid held in the liquid holder of the movable sample holding platform to ensure that the surface of the target tissue forms an interface with the UV transparent film. In some embodiments, the liquid held in the liquid holder can be water or a mixture of water with UV transparency. In addition, other liquids with a suitable refractive index (e.g., 1.3 to 1.5) and other liquids with high UV transmittance may also be used to hold the target tissue and fill the interface between it and the sample holder base. Figure 1B According to some embodiments, the system of the present invention adopts an open-top light sheet configuration, allowing the light sheet to pass through the target tissue and exit from the open top of the movable sample holding platform. Long DOF (e.g., up to 200 μm) can be achieved through selective dual-axis planar illumination generated by MUSI. Figure 1B The dual-axis planar illumination configuration separates the illumination path from the detection path.

[0121] Ensure that the incident beam path and the detection beam path are orthogonal to each other and are respectively located at a 45° lower position with respect to the vertical axis of the sample holder. The first plurality of optical elements at the irradiation arm of the system of the present invention includes a reflector, a bandpass filter, a pair of convex lenses, an adjustable slit aperture, and a UV cylindrical lens. The function of the first plurality of optical elements is to guide the deep UV excitation beam generated by a nanosecond UV pulse laser (wavelength is about 266nm) to the bottom of the tissue sample in the sample holder at a specific incident angle, and modulate the beam into a light sheet (803) of several millimeters wide (for example, a 3mm wide Gaussian light sheet is generated, with a beam waist radius (w0) of about 2.8μm and a DOF (2Z R ) is about 200 μm (such as FIG. 2A to FIG. 2C In some embodiments, the light sheet generated is The light sheet is designed to be substantially perpendicular to the UV transparent window on one side of the prism-like structure at the bottom of the liquid retainer. The two UV transparent windows on both sides of the prism-like structure can transmit the incident light beam to the maximum extent and allow for the maximum transmission of the detection light beam (including the fluorescent signal emitted by the tissue sample after excitation). In some embodiments, the two UV transparent windows are UV transparent quartz windows. The light sheet passing through the prism-like structure at the bottom of the liquid retainer has a radiation intensity of 2 mJ / cm 2 The average energy flux is sufficient to excite endogenous fluorophores in the tissue to emit fluorescent signals.

[0122] The detection arm of the system of the present invention is equipped with a second plurality of optical elements, which are adjusted accordingly to receive the maximum intensity of the fluorescent signal from the tissue sample at a detection angle of 45° with the vertical axis of the sample holder (804). The detection beam leaves the liquid holder through a UV transparent quartz window on the other side of the prism-like structure, which is opposite to the window through which the incident beam passes. The other UV transparent quartz window is designed to be substantially perpendicular to the detection beam path to ensure that the second plurality of optical elements can receive the strongest fluorescent signal emitted by the tissue sample. In certain embodiments, the second plurality of optical elements include but are not limited to achromatic UV objective lenses, long pass filters, infinity correction tube lenses and reflectors. The light emission signal received from the tissue sample by the second plurality of optical elements is captured by an optical detection unit (such as a sCMOS camera) (805). By connecting to a 3D translation stage connected to a movable sample holding platform, the tissue sample is translated from one end to the other end in the main scanning direction to obtain the first raw image data. Subsequently, the sample is translated a lateral distance in the secondary scanning direction to obtain subsequent strips of raw image data adjacent to the first strip of raw image data until the entire bottom surface of the tissue sample is covered (806). In some embodiments, multiple strips of raw image data are obtained by scanning the bottom surface of the tissue sample in a serpentine pattern between the main scanning direction and the secondary scanning direction. The present invention also supports other scanning patterns as long as the maximum surface area of ​​the tissue sample can be scanned. At the same time, a larger image height can also be used to accommodate rough tissues with larger surface irregularities. In some embodiments, there is approximately a 10% overlap area between two adjacent raw image data. The multiple strips of raw image data are stored in a memory or workstation of the system of the present invention. Subsequently, the stored raw image data is loaded as a stack (data volume) into an image processing module with different plug-ins to reconstruct tissue geometry from the raw data volume and extract surface features from a large FOV after splicing multiple image strips (807). In some embodiments, the image processing module uses a MATLAB-Fiji interface combined with multiple plug-ins (algorithms) to cut the initial relative to the tissue surface (such as Figure 1CThe original data volume recorded at 45° (in the xz plane in the image) is formed into an oblique prism-shaped data volume to correct image distortion, and multiple image strips of the original data are spliced ​​to form a large FOV, and the spatial distribution of the cell nucleus is determined by extracting its features, thereby training the deep learning model. The reconstructed tissue image can be directly output for medical practitioners or surgeons to interpret the state of the tissue sample (808), or compared with the corresponding H&E staining image (809). In some embodiments, the deep learning model can generate a pseudo-histochemical staining MUSI image to assist medical practitioners or surgeons in interpreting the state of the tissue sample and making intraoperative decisions or postoperative clinical evaluations.

[0123] Although the invention has been described in terms of certain embodiments, other embodiments that are apparent to those of ordinary skill in the art are also within the scope of the invention. Therefore, the scope of the invention is intended to be limited only by the appended claims.

[0124] Industrial Applicability

[0125] The present invention provides a label-free and non-destructive platform and imaging method suitable for imaging of freshly excised and intact living tissue surfaces, and has great potential to become the next generation of clinical standard methods and intraoperative diagnostic tools. The method can provide real-time guidance to surgeons or pathologists to make clinical or medical decisions (such as determining the extent of resection of suspicious tumors during cancer surgery). By combining MUSI and autofluorescence spectroscopy, the function, structure and / or metabolic changes of the target tissue can be more accurately characterized, thereby further improving the accuracy and professionalism of this label-free, non-destructive imaging method. In addition, combined with deep learning neural networks and models, the presence and spatial distribution of different microstructures in the target tissue can be identified, helping pathologists to more efficiently interpret tissue states related to disease prognosis.

Claims

1. An imaging system, comprising: a movable sample holding platform comprising a sample holder and a liquid holder, the liquid holder having a prism-like structure; a deep UV excitation source, which provides deep UV irradiation; a first plurality of optical elements for generating a light sheet from deep UV illumination provided by the deep UV excitation source and directing the light sheet through one side of the prism-like structure of at least one surface of the liquid holder to impinge on a bottom surface of the sample at an angle of incidence relative to a vertical axis of the sample holder; a second plurality of optical elements for receiving light emissions from the sample excited by the light sheet, including emitted fluorescent signals, at detection angles relative to a vertical axis of the sample holder; as well as an optical detection unit for detecting and processing light emissions received by said second plurality of optical elements, The incident angle and the detection angle are substantially the same in magnitude, and a beam path of the light sheet is perpendicular to a beam path of light emission from the sample.

2. The imaging system of claim 1, wherein the movable sample holding platform has an open-top configuration.

3. The imaging system of claim 1, wherein the movable sample holding platform is connected to a three-dimensional (3D) translation stage.

4. The imaging system of claim 1, wherein at least one surface of the sample holder comprises a membrane that supports a sample. 5 . The imaging system of claim 4 , wherein the liquid in the liquid holder of the movable sample holding platform comprises water or a mixture of water having UV transparency.

6. The imaging system of claim 1, wherein the first plurality of optical elements comprises a first filter, a pair of lenses, a slit aperture, and a cylindrical lens. The imaging system of claim 6 , wherein the first filter is a bandpass filter.

8. The imaging system of claim 6, wherein the pair of lenses is a pair of UV grade convex lenses.

9. The imaging system of claim 6, wherein the slit aperture is an adjustable slit aperture.

10. The imaging system of claim 6, wherein the cylindrical lens is a UV cylindrical lens for generating a Gaussian light sheet.

11. The imaging system of claim 1 , wherein the angle of incidence of the light sheet irradiated on the bottom surface of the sample is 45° relative to the vertical axis of the sample holder, and the average energy flux of the light sheet complies with the safe UV radiation threshold specified by the American Conference of Governmental Industrial Hygienists.

12. The imaging system of claim 1, wherein the liquid holder is disposed below the sample holder in the movable sample holding platform.

13. The imaging system of claim 1, wherein the sample is supported by a UV transparent film fixed at a base of the sample holder so that the light sheet can reach a bottom surface of the sample held in the sample holder.

14. The imaging system of claim 13, wherein the UV transparent film is made of highly transparent thermoplastic including polyethylene or any UV transparent material.

15. The imaging system of claim 1 , wherein the prism-like structure of the movable sample holding platform comprises two UV transparent windows disposed at two opposite sides of the prism-like structure, each UV transparent window allowing the light sheet generated from the deep UV excitation source to enter the movable sample holding platform, or allowing light emission including an emitted fluorescence signal from the sample under excitation by the light sheet to leave the movable sample holding platform.

16. The imaging system of claim 1, wherein the sample is a biological tissue including freshly excised tissue, or a living tissue from a living organism, including normal tissue and abnormal tissue.

17. The imaging system of claim 16, wherein the biological tissue is rich in endogenous fluorophores, the endogenous fluorophores comprising reduced nicotinamide adenine dinucleotide (NADH), structural proteins, aromatic amino acids, and heterocyclic compounds.

18. The imaging system of claim 1, wherein the second plurality of optical elements comprises at least a UV objective, a second filter, and an infinity correction lens.

19. The imaging system of claim 18, wherein the UV objective is an achromatic UV objective.

20. The imaging system of claim 18, wherein the second filter is a long pass filter.

21. The imaging system of claim 18, wherein the infinity-corrected lens is an infinity-corrected tube lens.

22. The imaging system of claim 1, wherein the optical detection unit comprises a plurality of CMOS image sensors.

23. The imaging system of claim 22, wherein the plurality of CMOS image sensors of the optical detection unit comprises scientific complementary metal oxide semiconductor (sCMOS) sensors.

24. A method for imaging biological tissue in a label-free, unprocessed manner, the method comprising using the imaging system of any one of claims 1 to 23 to output a histology-like image showing a two-dimensional or three-dimensional outline of the biological tissue, the use of the imaging system of claim 1 comprising the following steps: (a) placing the biological tissue into a sample holder of a movable sample holding platform of the imaging system, wherein a relatively flat side of the biological tissue faces a UV transparent film disposed at the bottom of the sample holder; (b) immersing the sample holder holding the biological tissue into a liquid holder of the movable sample holding platform filled with liquid; (c) adjusting one or more of the first plurality of optical elements to direct the light sheet generated thereby onto the bottom surface of the sample at an incident angle of 45° relative to the vertical axis of the sample holder, or the incident beam path of the light sheet is perpendicular to the UV transparent window on one side of the prism-like structure at the bottom of the liquid holder; (d) adjusting one or more optical elements of the second plurality of optical elements to receive light emissions including fluorescent signals emitted by the sample at a detection angle of 45° relative to a vertical axis of the sample holder or orthogonal to an incident beam path of the light sheet; (e) activating the deep UV excitation source to generate UV radiation toward the first plurality of optical elements, which is then modulated by the first plurality of optical elements to generate a light sheet, and directing the light sheet to a first point of the sample through a UV transparent window of the prism-like structure at the bottom of the liquid holder at an incident angle of 45° relative to a vertical axis of the sample holder; (f) receiving, through the second plurality of optical elements, light emissions from the sample excited by the light sheet, including emitted fluorescent signals; (g) detecting, by an optical detection unit comprising a plurality of CMOS image sensors, received light emissions from the second plurality of optical elements, including emitted fluorescent signals; (h) moving the movable sample holding platform driven by the 3D translation stage from a first point of the sample to subsequent points on the same plane in the main scanning direction, and repeating steps (e) to (g) at each of the subsequent points until reaching the last point on the same plane in the main scanning direction to form a first image strip; (i) moving the movable sample holding platform transversely relative to the first scanning direction toward a secondary scanning direction, and repeating steps (e) to (h) to form subsequent image strips adjacent to the first image strip until the entire maximum surface area of ​​the biological tissue is scanned; as well as (j) processing the image data detected by the optical detection unit through an image processing module to reconstruct the geometric shape of the biological tissue, and outputting one or more images representing the overall geometric shape of the biological tissue after reconstruction.

25. The method of claim 24, wherein the generated light sheet has a wavelength of 266 nm.

26. The method of claim 24, wherein the penetration depth of the light sheet is up to about 30 μm from the bottom surface of the biological tissue.

27. A method according to claim 24, wherein after a first image strip is formed along the main scanning direction, the movable sample holding platform is moved laterally toward the secondary scanning direction relative to the main scanning direction, so that a subsequent image strip adjacent to the first image strip is formed along the main scanning direction, wherein the movable sample holding platform adopts a serpentine or column-by-column scanning movement path to cover the entire maximum surface area of ​​the biological sample.

28. The method of claim 24, wherein the image processing module is a stand-alone computer processor or is network-connected to the imaging system as part of the imaging system, and the image processing module includes one or more image processing algorithms for processing a series of image strips of detected image data.

29. The method of claim 24, wherein distortion of the image stripes in the original image due to a detection angle that is not parallel to the vertical axis of the sample holder is corrected by one of the image processing algorithms.

30. The method of claim 24, wherein the detected surface features in the image data are extracted by another image processing algorithm until all image swaths are processed.

Images