Device for performing three-dimensional imaging and structural analysis on human tumor in-vitro specimen

Abstract

The invention provides a device for carrying out three-dimensional imaging and structural analysis on a human tumor in-vitro specimen. The device comprises a main control system, wherein the main control system is used for generating a plurality of continuous imaging position coordinates according to an imaging range and imaging parameters; and the optical coherence tomography system is used for imaging the section to obtain corresponding three-dimensional body data. And the movable objective table is used for moving the sample to each imaging position coordinate and adjusting the height of the sample after one section is imaged. And the slicing system is used for cutting off the imaged section of the sample to form a new imaging section. And the image processing system is used for splicing the three-dimensional volume data of the imaging position coordinates of the same section to obtain an optical coherence tomography image of a single section, performing three-dimensional volume data reconstruction based on the optical coherence tomography images of all sections to obtain a three-dimensional reconstructed image, and performing structural analysis to obtain blood vessel morphology and distribution data. The optical coherence tomography imaging system overcomes the limitation of the optical coherence tomography imaging system in the aspects of imaging view field and imaging depth.

Description

Technical Field

[0001] The present invention relates to the field of biological imaging technology, and in particular to a device for performing three-dimensional imaging and structural analysis on in vitro human tumor specimens. Background Art

[0002] Biomedical imaging technology plays a vital role in disease diagnosis, treatment, and monitoring. In particular, three-dimensional imaging of ex vivo human tumor specimens can reveal information such as the spatial morphology of cell layers and vascular distribution within tumor tissue. This is crucial for clarifying tumor structural characteristics and tumor-vascular network relationships, enabling accurate pathological diagnosis, grading, and personalized treatment. Traditionally, tumor tissue structure observation has been achieved by sectioning ex vivo tumor specimens, staining the sections, and then imaging them. However, this imaging method requires high technical skills for sectioning and staining, is time-consuming, and labor-intensive. Furthermore, staining captures biological information sensitive to specific dyes, and the production process is prone to problems such as incomplete sectioning and fragmentation during staining and mounting. More importantly, traditional sectioning and staining techniques are typically two-dimensional, failing to capture complete three-dimensional information about the tissue, particularly hindering the investigation of vascular connectivity and the complete morphology of the vascular network. Furthermore, in the field of imaging ex vivo human tumor specimens, there is currently no micron-scale tissue imaging technology capable of imaging several centimeters of ex vivo human tumor specimens in a single imaging field of view.

[0003] With the continuous advancement of imaging technology, optical coherence tomography (OCT) imaging technology can image biological tissues in three dimensions with micron-level imaging resolution, non-contact imaging, and submillimeter tissue penetration depth. However, the limited imaging field of view and imaging depth of OCT imaging systems still make it difficult to fully image large tissue volumes in a single shot. Summary of the Invention

[0004] The present invention provides a device for three-dimensional imaging and structural analysis of ex vivo human tumor specimens, addressing the existing technical flaw of single-shot imaging that makes it difficult to fully capture the entirety of large tissue volumes. This device enables three-dimensional reconstruction and tissue structural analysis of ex vivo human tumor specimens, providing a quantitative basis for accurately assessing the three-dimensional morphology, cell stratification, and vascular distribution of ex vivo human tumor tissue. The technical solutions proposed by the present invention are as follows: In a first aspect, the present invention provides a device for performing three-dimensional imaging and structural analysis of an in vitro human tumor specimen, comprising: A main control system is used to generate a plurality of continuous imaging position coordinates according to the received imaging range and imaging parameters, and to generate an imaging control signal and a movement control signal, and to generate a slicing control signal according to the received slicing parameters; An optical coherence tomography system is used to image a cross section of an in vitro human tumor specimen to be measured according to the imaging position coordinates in the imaging control signal to obtain corresponding three-dimensional volume data; a moving stage, used for moving the in vitro human tumor specimen to be tested to each imaging position coordinate according to the movement control signal, and adjusting the height of the in vitro human tumor specimen to be tested after completing a cross-sectional imaging; a slicing system for cutting off a completed section of the in vitro human tumor specimen to be tested according to the slicing control signal after the in vitro human tumor specimen to be tested reaches a preset position, thereby forming a new imaging section; An image processing system is used to stitch the three-dimensional volume data of the imaging position coordinates of the same section to obtain an optical coherence tomography image of a single section, reconstruct the three-dimensional volume data based on the optical coherence tomography images of all sections to obtain a three-dimensional reconstructed image, and perform structural analysis based on the three-dimensional reconstructed image to obtain vascular morphology and distribution data.

[0005] Optionally, the imaging parameters include imaging sequence, imaging field of view, and imaging redundancy within a single plane; The image processing system is used to splice the three-dimensional volume data of the imaging position coordinates of the same section according to the imaging sequence, imaging field of view and imaging redundancy in a single plane to obtain an optical coherence tomography image of a single section.

[0006] Optionally, the imaging parameters further include the amount of stage elevation, the order of cross-sectional imaging, and the redundancy between the depths of two adjacent cross-sectional images; The image processing system is used to perform depth-wise splicing of the optical coherence tomography images of all sections based on the amount of stage elevation, the order of section imaging, and the depth redundancy between two adjacent section images, until the three-dimensional volume data reconstruction of the in vitro human tumor specimen to be tested is completed to obtain a three-dimensional reconstructed image.

[0007] Optionally, the stage elevation is less than a penetration depth of optical coherence tomography.

[0008] Optionally, the image processing system is used to: measure different cell layers based on the three-dimensional reconstructed image, perform blood vessel segmentation, and quantitatively calculate the morphology and distribution data of the blood vessels.

[0009] Optionally, the in vitro human tumor specimen sample to be tested is obtained by the following method: embedding the in vitro human tumor specimen tissue to be tested, and fixing the embedded sample to obtain the in vitro human tumor specimen sample to be tested.

[0010] Optionally, the in vitro human tumor specimen to be tested is immersed in a buffer solution, and imaging and sectioning are both performed in the buffer solution.

[0011] Optionally, the imaging parameters further include tissue geometric dimensions; The main control system is further configured to determine the number of slices according to the received slice parameters and the geometric dimensions of the tissue.

[0012] In a second aspect, the present invention further provides a method for performing three-dimensional imaging and structural analysis on an in vitro human tumor specimen, which is implemented by the apparatus for performing three-dimensional imaging and structural analysis on an in vitro human tumor specimen as described in the first aspect, comprising: Using the main control system to generate a plurality of continuous imaging position coordinates according to the received imaging range and imaging parameters, and to generate an imaging control signal and a movement control signal, and to generate a slicing control signal according to the received slicing parameters; The mobile stage moves the in vitro human tumor specimen to be tested to each imaging position coordinate according to the movement control signal, and the optical coherence tomography imaging system images the cross section of the in vitro human tumor specimen to be tested according to each imaging position coordinate in the imaging control signal to obtain corresponding three-dimensional volume data; After completing a cross-sectional imaging, the height of the human tumor in vitro specimen to be tested is adjusted using a movable stage; After the human tumor in vitro specimen sample to be tested reaches a preset position, a slicing system is used to cut off a completed section of the human tumor in vitro specimen sample to be tested according to the slicing control signal to form a new imaging section; Using an optical coherence tomography system to image the next section of the human tumor in vitro specimen to be tested, until imaging of all sections is completed; An image processing system is used to stitch the three-dimensional volume data of the imaging position coordinates of the same section to obtain an optical coherence tomography image of a single section. Three-dimensional volume data reconstruction is performed based on the optical coherence tomography images of all sections to obtain a three-dimensional reconstructed image. Structural analysis is performed based on the three-dimensional reconstructed image to obtain vascular morphology and distribution data.

[0013] Optionally, the imaging parameters include an imaging sequence within a single plane, an imaging field of view, and an imaging redundancy; and the step of using an image processing system to stitch together the three-dimensional volume data of the imaging position coordinates of the same section to obtain an optical coherence tomography image of a single section includes: The image processing system splices the three-dimensional volume data of the imaging position coordinates of the same section according to the imaging sequence, imaging field of view and imaging redundancy in a single plane to obtain an optical coherence tomography image of a single section.

[0014] Based on the above technical solution, the present invention has the following beneficial effects compared with the prior art: The present invention provides an apparatus for three-dimensional imaging and structural analysis of ex vivo human tumor specimens. By precisely moving a movable stage in the horizontal XY directions, the specimen is divided into multiple small regions for imaging, with each imaging session covering a small region. An image processing system stitches the three-dimensional volume data from these small regions into a complete cross-sectional image. This method covers the entire horizontal extent of the tumor specimen, overcoming the field of view limitations of a single imaging session. After imaging a cross-section, the movable stage is raised to a certain height in the vertical Z direction, and a slicing system automatically excises the imaged tissue portion. This layer-by-layer excision exposes new tissue surfaces, enabling the OCT imaging system to image deeper tissue layers. This imaging and slicing process is repeated, layer by layer, until the entire tumor specimen is imaged. This method effectively extends the imaging depth of the OCT imaging system, enabling it to cover the entire vertical extent of the tumor specimen. The present invention utilizes horizontal XY movement of the movable stage and image stitching technology to cover the entire horizontal extent of the tumor specimen. The slicing system's layer-by-layer excision and cyclic imaging also covers the entire vertical extent of the tumor specimen. By combining optical coherence tomography (OCT) technology with a mobile stage, slicing system, and image processing, the system successfully overcomes the limitations of OCT imaging systems in terms of field of view and depth. The image processing system stitches images from multiple sections into a complete 3D reconstruction and analyzes tissue structure. This system is capable of complete 3D imaging and structural analysis of even large tumor specimens, providing a quantitative basis for accurately assessing the 3D morphology, cell stratification, and vascular distribution of ex vivo human tumor tissue.

[0015] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or understood by practicing the present invention. The purposes and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0016] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 It is a schematic structural diagram of the device provided by the present invention for performing three-dimensional imaging and structural analysis on in vitro human tumor specimens.

[0019] Figure 2 The figure is a schematic diagram of the working process of the device provided by the present invention for performing three-dimensional imaging and structural analysis on in vitro human tumor specimens.

[0020] Figure 3a and Figure 3b The present invention provides imaging of 20 consecutive positions of a piece of human brain tumor tissue; wherein, Figure 3a is the XY plane diagram of each position imaging provided by the present invention, Figure 3b This is a spliced image of one cross-section OCT provided by the present invention.

[0021] Figure 4 This is a three-dimensional reconstructed image of a piece of human brain tumor tissue provided by the present invention.

[0022] Figure 5a 、 Figure 5b and Figure 5c This is a schematic diagram of cell stratification and blood vessel distribution in a piece of human brain tumor tissue calculated by the present invention, wherein: Figure 5a This is a schematic diagram of cell stratification. Figure 5b This is a three-dimensional rendering of the main blood vessels. Figure 5c This is a diagram of tiny blood vessels. DETAILED DESCRIPTION

[0023] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0024] The following describes a device for performing three-dimensional imaging and structural analysis of human tumor specimens in vitro. The device combines optical coherence tomography (OCT), automated control, image processing, and slicing technology.

[0025] Reference Figure 1 As shown, the device includes: an optical coherence tomography (OCT) imaging system 1, a slicing system 2, a moving stage 3, a main control system 4 and an image processing system 5.

[0026] The optical coherence tomography system 1 is used to image a cross-section of an in vitro human tumor specimen based on the imaging position coordinates in the imaging control signal, generating corresponding three-dimensional volume data. The OCT imaging system utilizes optical coherence tomography technology to perform high-resolution imaging of tumor specimen cross-sections. Each image generated is a three-dimensional volume data set containing detailed structural information about the cross-section.

[0027] Prior to imaging, the ex vivo human tumor specimen is embedded and the cross-section is smoothed using a slicing system 2 to ensure image quality. The OCT imaging system images the cross-section of the specimen, generating a three-dimensional volumetric dataset for each imaging pass. By moving the stage 3 horizontally in the XY directions, the OCT imaging system images at multiple locations. The imaging fields of adjacent locations overlap to facilitate subsequent image stitching. After imaging is complete, the data is transmitted to an image processing system 5 for further processing.

[0028] Slicing system 2 is used to slice the in vitro human tumor specimen after it reaches a preset position, according to a slicing control signal, removing the completed cross-section from the in vitro human tumor specimen to form a new imaging section. Specifically, slicing system 2 is used to automatically remove the imaged tissue portion after the specimen reaches the preset position to allow imaging of the next section. This process ensures that subsequent imaging can proceed smoothly, and that each imaging session uses a new, unimaged tissue section.

[0029] The slicing system 2 operates as follows: After the movable stage 3 is raised to a certain height, it automatically slices the specimen, removing the elevated tissue portion. The main control system 4 sets the slice thickness based on imaging parameters, ensuring that the thickness of each excision matches the imaging depth. After slicing, the specimen cross section is re-exposed, ready for imaging the next cross section.

[0030] The mobile stage 3 is used to precisely control the position of the tumor specimen, including horizontal XY movement and vertical Z adjustment. Specifically, the mobile stage 3 is used to move the in vitro human tumor specimen to various imaging coordinates based on the movement control signals. After imaging one section, the mobile stage 3 adjusts the height of the in vitro human tumor specimen to facilitate imaging of the next section. This precise movement ensures imaging continuity and integrity.

[0031] The process flow of moving stage 3: In the horizontal XY directions, mobile stage 3 moves the specimen to the designated position for imaging based on the coordinates generated by main control system 4. After imaging a complete section, mobile stage 3 is raised in the vertical Z direction by a certain height (e.g., the slice thickness). After the elevation, slicing system 2 slices the specimen, removing the raised tissue portion and preparing for imaging the next section. This process is repeated until the entire tumor specimen is imaged.

[0032] The main control system 4 is the core control unit of the entire device, responsible for coordinating the operation of each subsystem. The main control system 4 is used to generate multiple continuous imaging position coordinates based on the received imaging range and imaging parameters. Specifically, the main control system 4 generates multiple continuous imaging position coordinates based on the imaging range (e.g., tumor specimen size) and imaging parameters (e.g., resolution, scanning speed, slice thickness, etc.) input by the user. These coordinates will guide the subsequent imaging process and generate imaging control signals and movement control signals, as well as slicing control signals based on the received slicing parameters. After completing the corresponding imaging, slicing, and movement operations, the optical coherence tomography system 1, slicing system 2, and mobile stage 3 will send a status feedback signal to the main control system 4. After receiving the status feedback signal, the main control system 4 generates the corresponding control signal. For example, after the mobile stage 3 moves the in vitro specimen sample of the human tumor to be tested to each imaging position coordinate, it sends a first state feedback signal to the main control system 4. After receiving the first state feedback signal, the main control system 4 generates an imaging control signal to control the optical coherence tomography system 1 to perform an imaging operation; after the optical coherence tomography system 1 completes the imaging operation, it sends a second state feedback signal to the main control system 4. After receiving the second state feedback signal, the main control system 4 generates a slicing control signal to control the slicing system 2 to cut off the completed section of the in vitro specimen sample of the human tumor to be tested according to the slicing control signal after the in vitro specimen sample of the human tumor to be tested reaches the preset position, so as to form a new imaging section for subsequent imaging operations.

[0033] The main control system 4 operates as follows: the user sets the imaging range and parameters through the system, which then calculates the imaging position coordinates for each section (including the horizontal XY and vertical Z movement distances). The system then transmits this coordinate information to the mobile stage 3 and the optical coherence tomography system 1, ensuring precise imaging and slicing.

[0034] Image processing system 5 is responsible for processing and analyzing imaging data, generating 3D reconstructed images, and analyzing tumor tissue structural information. Specifically, it is used to stitch 3D volume data from imaging position coordinates of the same section to produce a single-section optical coherence tomography image, reconstruct 3D volume data from all sections of the optical coherence tomography images to produce a 3D reconstructed image, and perform structural analysis on the 3D reconstructed image to obtain vascular morphology and distribution data.

[0035] By stitching together the 3D volume data from various imaging positions within a single cross-section, a complete optical coherence tomography image of that section can be obtained. Furthermore, based on the optical coherence tomography images from all cross-sections, the image processing system 5 can perform 3D volume data reconstruction to obtain a 3D reconstructed image of the tissue under examination. Finally, based on the 3D reconstructed image, the system can also perform structural analysis to extract key information such as vascular morphology and distribution.

[0036] Image Processing System 5's workflow: Three-dimensional volume data from multiple imaging positions within the same cross-section are stitched together to generate a single-section optical coherence tomography image. During stitching, the overlapping regions of adjacent imaging fields are precisely aligned. Based on the optical coherence tomography images from all cross-sections, 3D volume data is reconstructed to generate a complete 3D reconstructed image. Structural analysis is performed on the reconstructed image to extract vascular morphology and distribution data from tumor tissue, providing important information for medical research.

[0037] The apparatus of the present invention for performing three-dimensional imaging and structural analysis of human tumor specimens in vitro has the following workflow: S110, Specimen Preparation: Embed the in vitro human tumor specimen to ensure its stability and integrity during imaging. Then, use the slicing system 2 to repair the end surface of the sample to make it flat and smooth for subsequent imaging and slicing operations.

[0038] S120, OCT imaging: Start the main control system 4 and input the imaging range and imaging parameters. Move the stage 3 to the first imaging position coordinate. The OCT imaging system images the cross-section of the sample, generating three-dimensional volume data. The stage moves horizontally in the XY direction to the next imaging position coordinate for the next imaging. The imaging fields of adjacent positions should have a certain overlap area to ensure imaging continuity and integrity. Repeat the above steps until a complete cross-section is imaged.

[0039] S130, Z-axis Adjustment and Slicing: After imaging a complete cross-section, the mobile stage 3 is raised vertically in the Z direction. The slicing system 2 automatically slices the sample, removing the imaged tissue. The stage then moves back to the starting position or the next imaging position coordinates to image the next cross-section.

[0040] S140, cyclic imaging: repeat the above imaging, slicing and cyclic steps until the imaging of the entire tissue to be tested is completed.

[0041] S150, Image Processing: Image processing system 5 receives and processes all the 3D volume data obtained through imaging. Through stitching and 3D reconstruction techniques, a 3D reconstructed image of the in vitro human tumor specimen is obtained. Structural analysis is performed based on the 3D reconstructed image, extracting key information such as vascular morphology and distribution.

[0042] The field of view of an OCT imaging system is limited in a single imaging pass, typically covering only a few millimeters to a few centimeters, making it difficult to cover larger tumor specimens in a single pass. Furthermore, the imaging depth of an OCT imaging system is typically submillimeter (1-2 mm), making it difficult to penetrate larger tumor specimens in a single pass. The present invention's device for three-dimensional imaging and structural analysis of ex vivo human tumor specimens successfully overcomes the limitations of OCT imaging systems in terms of field of view and depth by combining optical coherence tomography (OCT) technology, a movable stage 3, a slicing system 2, and an image processing system 5. By precisely moving the movable stage 3 in the horizontal X and Y directions, the tumor specimen is divided into multiple small regions for imaging, with each imaging covering a small region. The image processing system 5 stitches the 3D volume data from these small regions into a complete cross-sectional image. This method covers the entire horizontal extent of the tumor specimen, overcoming the limitations of the field of view of a single imaging pass. After imaging a cross-section, the movable stage 3 is raised to a certain height in the vertical Z direction, and the slicing system 2 automatically removes the imaged tissue portion. Through this layer-by-layer resection method, new tissue surfaces are exposed layer by layer, enabling the OCT imaging system to image deep tissues. The above imaging and slicing process is repeated, imaging and slicing layer by layer, until the imaging of the entire tumor specimen is completed. This method effectively expands the imaging depth of the OCT imaging system, enabling it to cover the vertical range of the entire tumor specimen. By moving the horizontal XY direction of the stage 3 and the image stitching technology, the horizontal range of the entire tumor specimen is covered. By the layer-by-layer resection and cyclic imaging of the slicing system 2, the vertical range of the entire tumor specimen is covered. The images of multiple sections are stitched into a complete three-dimensional reconstructed image through the image processing system 5, and the tissue structure information is analyzed. It can perform complete three-dimensional imaging and structural analysis on large tumor specimens, providing a powerful tool for tumor research and clinical diagnosis.

[0043] A single OCT imaging session can only obtain local 3D volume data, making it difficult to directly generate a complete 3D reconstructed image. The present invention utilizes an image processing system 5 to stitch together the optical coherence tomography images from each section to generate a complete 3D reconstructed image. Based on the 3D reconstructed image, the image processing system 5 can further analyze tumor tissue structural information, such as vascular morphology and distribution data. This approach not only overcomes the field of view and depth limitations of OCT imaging systems but also provides more comprehensive information on tissue structure and pathology.

[0044] Manual operation makes it difficult to achieve high-precision imaging and slicing, and is prone to introducing errors. The present invention utilizes a main control system 4 to automatically generate imaging position coordinates based on the imaging range and parameters set by the user, and controls the operation of the movable stage 3 and slicing system 2. This automated operation ensures high precision and consistency in the imaging and slicing processes. The movable stage 3 achieves micron-level movement accuracy in the horizontal XY and vertical Z directions, ensuring accurate imaging and slicing every time. The coordinated operation of the main control system 4 and the movable stage 3 achieves high-precision imaging and slicing.

[0045] In an optional embodiment, the present invention does not require staining or any other labeling of the ex vivo human tumor tissue specimen. To ensure section quality, the present invention embeds the ex vivo human tumor tissue specimen to be tested, and the cross-section is smoothed using a sectioning system 2. The ex vivo human tumor tissue specimen to be tested is obtained by embedding the ex vivo human tumor tissue specimen to be tested, and fixing the embedded sample to obtain the ex vivo human tumor tissue specimen to be tested.

[0046] The purpose of embedding is to embed the tumor tissue in a supporting medium so that it maintains the stability of its shape and structure during subsequent sectioning and imaging. Embedding uses media such as gelatin, agarose and resin that are more suitable for high-resolution imaging. The embedded tissue can maintain its morphology and structure to avoid deformation or damage during sectioning and imaging. The purpose of fixation is to stabilize the protein and cell structure in the tissue through chemical cross-linking or physical methods to prevent the tissue from degrading or deforming during subsequent processing. The fixed tissue can preserve its morphology and structure for a long time to avoid degradation during sectioning and imaging. Through embedding and fixation, the present invention can enable the tumor tissue to maintain its original morphology and structure to avoid deformation or damage during subsequent processing. The fixed tissue has high chemical and physical stability and is suitable for long-term preservation and multiple imaging. Through the embedding and fixation steps, the in vitro specimen sample of the human tumor to be tested can maintain the stability of its morphology and structure, providing a high-quality sample basis for subsequent high-resolution OCT imaging and structural analysis.

[0047] In an optional embodiment, the in vitro human tumor specimen to be tested is immersed in a buffer solution, and imaging and sectioning are both performed in the buffer solution.

[0048] Buffers (such as phosphate-buffered saline (PBS)) maintain tissue moisture and physiological conditions, preventing desiccation and deformation. The buffer's refractive index is close to that of biological tissue, reducing optical artifacts in OCT imaging and improving image quality. During the sectioning process, the buffer acts as a lubricant and coolant, reducing friction and heat during sectioning and improving section quality.

[0049] The working process of the device of the present invention is: S210. Immerse the embedded and fixed ex vivo tumor specimen in a buffer solution, ensuring complete immersion of the tissue. The buffer should be compatible with the OCT imaging and sectioning process. PBS, saline, or a dedicated tissue preservation solution can be used.

[0050] S220. Place the specimen soaked in buffer solution on the mobile stage 3, and the OCT imaging probe scans the specimen through the buffer solution. The OCT imaging system scans the specimen cross-section based on the coordinates generated by the main control system 4, generating a three-dimensional volume data set for each scan. The mobile stage 3 moves in the horizontal XY directions to ensure that adjacent imaging positions have a certain overlap area. After completing the imaging of one cross-section, the mobile stage 3 is raised in the vertical Z direction to a certain height (less than the penetration depth of the OCT) in preparation for imaging the next cross-section.

[0051] S230: Slice the specimen in a buffer solution. The microtome cuts the imaged section of the specimen in the buffer solution. Specifically, after the mobile stage 3 is raised to a certain height, the slicing system 2 automatically slices the specimen, removing the imaged tissue portion. The slice thickness is set by the main control system 4 based on imaging parameters to match the OCT penetration depth. After slicing, tissue fragments are washed away by the buffer solution to prevent interference with subsequent imaging.

[0052] S240, repeat the above imaging and slicing process until the entire tumor specimen is imaged. After each slicing, a new tissue surface is exposed to the buffer solution and imaged by the OCT imaging system.

[0053] S250: The image processing system 5 stitches and 3D reconstructs the OCT imaging data of each section to generate a complete 3D image. Structural analysis is performed based on the 3D image to extract the vascular morphology and distribution data of the tumor tissue.

[0054] The present invention reduces optical artifacts through a buffer solution, thereby improving the resolution and contrast of OCT imaging. Imaging in a buffer solution avoids tissue drying and deformation, ensuring imaging continuity and data integrity. The lubricating and cooling effects of the buffer solution reduce friction and heat during slicing, improving the quality and consistency of the slices. Slicing in a buffer solution avoids tissue damage and deformation, ensuring the accuracy of subsequent imaging. Imaging and slicing in a buffer solution enables automated operation, improving efficiency and precision. Imaging and slicing in a buffer solution can obtain high-resolution images of tumor tissue structure, helping researchers observe the microscopic features of tumors.

[0055] In an optional embodiment, the imaging parameters further include tissue geometric dimensions (such as the length, width, and height of the tumor specimen); The main control system 4 is further configured to determine the number of slices according to the received slice parameters and the tissue geometric dimensions (such as slice thickness).

[0056] The user manually enters the geometric dimensions of the tumor specimen before the procedure. Alternatively, the specimen can be pre-scanned using an OCT imaging system or other measurement device to automatically obtain its geometric dimensions. Tissue geometry includes length (X-axis), width (Y-axis), and height (Z-axis). Slicing parameters, including slice thickness, are user-defined based on research needs. The slice thickness should be optimized to match the OCT imaging penetration depth, for example, 0.2-1 mm. Slicing is performed vertically (Z-axis) to ensure layer-by-layer resection and imaging.

[0057] The main control system 4 calculates the number of slices as follows: input parameters, including tissue geometric dimensions (length, width, height) and slice thickness (set by the user).

[0058] Number of slices = tissue height (Z axis) / slice thickness. For example, if the tumor specimen is 10 mm tall and the slice thickness is 0.5 mm, the number of slices is 20.

[0059] The workflow of the device of the present invention is as follows: the main control system 4 generates a specific slicing plan based on the number of slices and the slice thickness, including the position coordinates and slicing order of each slice. The main control system 4 combines the slicing plan with the imaging parameters to ensure that the OCT imaging system can image the new tissue surface after each slice. The OCT imaging system images the first section of the tumor specimen and generates a three-dimensional volume data. The mobile stage 3 rises a certain height (equal to the slice thickness) in the vertical Z direction. The slicing system 2 automatically cuts off the imaged tissue part, and the OCT imaging system images the new tissue surface. The above process is repeated until all slicing times are completed. The image processing system 5 splices and reconstructs the OCT imaging data of each section into three dimensions to generate a complete three-dimensional image. Structural analysis is performed based on the three-dimensional image to extract the vascular morphology and distribution data of the tumor tissue.

[0060] The main control system 4 of the present invention accurately calculates the number of slices based on the tissue geometry and slice thickness, ensuring that the thickness of each slice is consistent and avoiding excessive or insufficient slicing. By automating the slicing and imaging cycle, the imaging efficiency is significantly improved, and the time and errors of manual operation are reduced. The slice thickness matches the OCT imaging penetration depth to ensure that each imaging covers the complete tissue section and avoid data loss. Based on the OCT imaging data of all sections, the image processing system 5 can generate high-precision three-dimensional reconstructed images and provide comprehensive tissue structure information. Users can adjust the slice thickness and tissue geometry according to specific needs to adapt to different types of tumor specimens.

[0061] In an optional embodiment, in order to overcome the limitations of the OCT imaging system in terms of imaging field of view and imaging depth, the device adopts a design in which the imaging fields of adjacent imaging position coordinates have overlapping areas.

[0062] When generating imaging position coordinates, the main control system 4 ensures a certain overlap between adjacent imaging positions (e.g., 10%-20% of the imaging field of view). This overlap ensures precise alignment of adjacent imaging regions during subsequent image stitching, preventing image misalignment or loss. The movable stage 3 moves the tumor specimen to multiple imaging positions in the horizontal XY directions, with the imaging field of each position overlapping with adjacent positions. This overlap ensures continuity between adjacent imaging regions, providing critical data support for subsequent image stitching.

[0063] In the horizontal XY directions, the movable stage 3 moves the specimen to the designated position for imaging based on coordinates generated by the main control system 4. Each movement distance is slightly less than the width of the imaging field of view, resulting in overlapping areas. In the vertical Z direction, the movable stage 3 rises to a certain height (e.g., the slice thickness) after imaging a section. The slicing system 2 then removes the imaged tissue portion, preparing for imaging the next section.

[0064] The slice thickness of the slicing system 2 matches the imaging depth of the OCT imaging system, ensuring that the thickness of tissue removed at each time does not affect the imaging quality of adjacent sections. By performing layer-by-layer excision and imaging, the device can achieve continuous three-dimensional imaging in the vertical Z direction while maintaining the overlap between adjacent sections.

[0065] Image processing system 5 utilizes the overlapping regions of adjacent imaging areas to stitch together the 3D volume data from multiple small regions into a complete cross-sectional image. The overlapping regions provide critical information for alignment and matching, ensuring a seamless stitched image. Based on the optical coherence tomography images from all cross-sections, image processing system 5 reconstructs the 3D volume data to generate a complete 3D reconstructed image. Structural analysis is performed on the 3D reconstructed image to extract data on the vascular morphology and distribution of the tumor tissue.

[0066] The overlapping region provides critical alignment information for image stitching, ensuring seamless stitching and avoiding misalignment or omissions. The overlapping region design enables continuous 3D imaging in both horizontal and vertical directions, covering the entire tumor specimen. The overlapping region provides data redundancy, improving the robustness and accuracy of image processing.

[0067] In an optional embodiment, the imaging parameters include an imaging sequence within a single plane, an imaging field of view, and an imaging redundancy. The image processing system 5 is configured to stitch the three-dimensional volume data of the imaging position coordinates of the same section based on the imaging sequence, the imaging field of view, and the imaging redundancy within the single plane to obtain an optical coherence tomography image of the single section.

[0068] Specifically, the imaging parameters are first set: the imaging sequence within a single plane determines the scanning path of the OCT imaging system when imaging the sample. Scanning sequences include line-by-line scanning and spiral scanning. The imaging field of view refers to the area that the OCT imaging system can cover in a single image. It determines the amount of information that can be acquired in a single image. To ensure seamless splicing of information between adjacent imaging positions, a certain overlap area, known as imaging redundancy, is set for each image.

[0069] According to the preset imaging parameters, the OCT imaging system images the cross-section of the sample to be tested. Each imaging will obtain a three-dimensional volume data containing the structural information of the area. The mobile stage 3 moves the sample to the next imaging position according to the preset imaging sequence for the next imaging. The image processing system 5 receives and stores the three-dimensional volume data of all imaging positions. According to the imaging sequence and imaging field of view information, the system can determine the position of each three-dimensional volume data in the overall image. Using the imaging redundancy information, the system splices adjacent three-dimensional volume data through feature matching, image registration and other technologies to ensure the continuity and integrity of the information. After the splicing process, the image processing system 5 finally obtains a complete optical coherence tomography image containing the entire cross-sectional structural information.

[0070] Based on the imaging field of view and imaging redundancy, a phase correlation and global optimal displacement model is used to estimate the optimal displacement change and the global optimal displacement between pairwise imaging data blocks (i.e., the above-mentioned three-dimensional volume data) within a single section, and the optimal displacement estimate and the global optimal displacement estimate values are stored in an .xml file. Based on the optimal displacement estimate and the global optimal displacement estimate, according to the imaging sequence, the Python programming language is used to call the BigStitcher plug-in in the open software Fiji to perform two-dimensional stitching on the imaging data acquired on the same section, and a fusion algorithm (such as alpha fusion, multiband fusion, and maximum fusion) model is used to achieve image fusion between pairwise imaging blocks, completing image fusion and stitching of the entire section. This can achieve two-dimensional image fusion and stitching of imaging data blocks at multiple imaging positions on all sections.

[0071] The above phase correlation and global optimal displacement model are used to estimate the optimal displacement change between two imaging data blocks in a single section and the global optimal displacement as follows: First, the frequency characteristics of the image are analyzed by phase correlation technology to estimate the relative displacement between the two imaging data blocks: Perform a Fourier transform on two adjacent image data blocks to obtain their frequency domain representations. Calculate the cross-power spectrum of the two frequency domain representations, which contains information about the phase difference between the two images. Perform an inverse Fourier transform on the cross-power spectrum to obtain an image in the spatial domain, referred to here as a phase correlation image. Find a peak in the phase correlation image; the location of this peak represents the displacement between the two images. Determine the offset of this peak relative to the image center to obtain a displacement vector. This displacement vector serves as a preliminary estimate of the optimal displacement.

[0072] These displacement estimates are then further optimized using a global optimal displacement model: The displacement vectors between all two image data blocks are organized into a displacement vector matrix. A global optimization algorithm (such as the least squares method or the iterative closest point algorithm) is used to optimize these displacement vectors to form a globally consistent displacement field across the entire cross-section. This optimization yields the globally optimal displacement for each image data block relative to a reference point (the first image data block).

[0073] The following is an example of a specific image stitching method: Each 3D volume data set is preprocessed to extract distinct feature points or regions. These features can be edges, corners, textures, etc. Features in adjacent 3D volume data sets are matched using a feature matching algorithm (such as SIFT, SURF, ORB, etc.). During the matching process, imaging redundancy must be considered to ensure consistency of the matched features in adjacent images. Based on the matched feature points, a transformation matrix is calculated between adjacent 3D volume data sets. This matrix describes the geometric transformation relationship from one image to another. Using the transformation matrix, adjacent 3D volume data sets are stitched together. During the stitching process, methods such as weighted averaging and Laplacian pyramid fusion can be used to reduce stitching gaps and image distortion. The stitched image is post-processed, such as denoising and contrast enhancement, to improve image quality. Through the above process, the image processing system 5 can accurately stitch together the 3D volume data sets at each imaging position coordinate within the same section, obtaining a complete optical coherence tomography image of a single section.

[0074] This invention leverages the high-resolution imaging capabilities of OCT technology to obtain detailed structural information from cross-sections of the sample being tested. By properly configuring imaging parameters and combining image stitching techniques, it is possible to image the entirety of larger tissue volumes. The stitched optical coherence tomography images present the internal structure of the sample being tested in three dimensions, facilitating observation and analysis by doctors and researchers. OCT technology utilizes a non-contact imaging method, avoiding physical damage to the sample being tested.

[0075] In an optional embodiment, the imaging parameters of the apparatus of the present invention encompass not only imaging details within a single plane, such as the imaging sequence, imaging field of view, and imaging redundancy, but also extend to depth-direction control, specifically the stage elevation, the sequence of cross-sectional imaging, and the depth-direction redundancy between adjacent cross-sectional images. These parameters are crucial for the image processing system 5 to stitch together all cross-sectional optical coherence tomography images in the depth direction, thereby completing the three-dimensional reconstruction of the in vitro human tumor specimen.

[0076] The image processing system 5 is configured to stitch all cross-sectional optical coherence tomography images in the depth direction based on the stage elevation, the order of cross-sectional imaging, and the depth redundancy between adjacent cross-sectional images, until the three-dimensional volume data of the in vitro human tumor specimen to be tested is reconstructed to produce a three-dimensional reconstructed image. To accurately stitch adjacent cross-sectional images, the present invention retains a certain overlap region between adjacent cross-sectional images. The size of this overlap region in the Z direction is referred to as the redundancy. Specifically, stitching includes estimating the optimal Z-direction displacement between two adjacent cross-sectional images based on the depth redundancy between the two adjacent cross-sectional images using phase correlation and a global optimal displacement model. Based on this optimal Z-direction displacement estimation, the BigStitcher plug-in in the open software Fiji is invoked in the Python programming language to reconstruct each completed stitched cross-sectional image in the Z direction. Simultaneously, a fusion algorithm (such as alpha fusion, multiband fusion, and maximum fusion) is employed to perform three-dimensional stitching of the Z-direction position information of each stitched cross-sectional image.

[0077] Specifically, perform Fourier transform on two adjacent cross-sectional images (i.e., optical coherence tomography images) and calculate their cross-power spectrum. Perform inverse Fourier transform on the cross-power spectrum to obtain a phase correlation image. Find the peak position in the phase correlation image, which represents the relative displacement of the two sections in the Z direction. Determine the offset of this peak position relative to the center of the image to obtain a displacement vector. Use a global optimization algorithm (such as the least squares method, iterative closest point algorithm, etc.) to optimize these displacement vectors to ensure that they form a consistent and smooth displacement field across the entire data set, thereby obtaining the optimal displacement of the two sections in the Z direction. Save each cross-sectional image in an appropriate format (such as TIFF) and ensure that they are arranged in order in the Z direction.

[0078] Use a Python script to call the BigStitcher plugin through Fiji's command-line interface (CLI) or a scripting interface (such as Jython or Groovy). Configure the BigStitcher plugin's parameters, including the input image file path, output directory, and stitching algorithm. Based on the previously calculated optimal Z-direction displacement, the BigStitcher plugin aligns and stitches each cross-sectional image together to form a 3D image along the Z direction. Select a fusion algorithm, such as alpha fusion, multiband fusion, or maximum fusion. Configure the fusion algorithm parameters in the BigStitcher plugin. Perform the fusion operation, fusing the stitched cross-sectional images in three dimensions along the Z direction to generate the final 3D image.

[0079] The amount of stage elevation determines the vertical distance between the next section to be imaged and the current section after each slicing. This parameter needs to be precisely set based on the thickness of the sample and the required imaging resolution. In the depth direction, the order of section imaging can be from top to bottom. Those skilled in the art can choose according to the specific circumstances of the sample and the imaging requirements. In order to ensure the continuity of information between adjacent sections, each section will have a certain overlapping area in the depth direction, that is, the redundancy in the depth of the two adjacent section images mentioned above. This helps to reduce information loss and dislocation in the subsequent image stitching process.

[0080] The working process of the device of the present invention is as follows: according to the preset imaging parameters, the OCT imaging system images the first section of the sample to be tested and obtains a three-dimensional volume data containing the structural information of the section. The mobile stage 3 moves the sample to the next position according to the preset stage elevation, and the slicing system 2 slices the sample to remove the imaged part. Repeat the above imaging and slicing steps until the imaging of all predetermined sections is completed. The image processing system 5 receives and stores the three-dimensional volume data of all sections. Utilizing the redundant information of adjacent section images in depth, the three-dimensional volume data of adjacent sections are spliced in the depth direction through feature matching, image registration and other technologies. During the splicing process, it is necessary to consider the order of section imaging and the stage elevation to ensure the consistency and continuity of the spliced image in the depth direction. After the splicing process in the depth direction, the image processing system 5 finally obtains a three-dimensional reconstructed image containing the overall structural information of the sample to be tested.

[0081] The following is an example of a specific depth-wise stitching method: The 3D volume data for each section is preprocessed to extract distinct feature points or regions. These features can be edges, corners, textures, etc., and must be stable and distinguishable in the depth direction. Feature matching algorithms (such as SIFT, SURF, and ORB) are used to match features in adjacent sections. During the matching process, redundant information in the depth direction must be fully considered to ensure the consistency of matched features across adjacent sections. Based on the matched feature points, the rigid body transformation matrix between adjacent sections is calculated. This matrix describes the rotation and translation relationship between one section and another in 3D space. Since the stage moves vertically after each sectioning, it can be assumed that the transformation between sections is primarily translational. Using the rigid body transformation matrix, the 3D volume data from adjacent sections are stitched together in the depth direction. During the stitching process, methods such as weighted averaging and Laplacian pyramid fusion can be used to reduce stitching gaps and image distortion. At the same time, it is important to ensure that the stitched image maintains smooth transitions and continuity in the depth direction. The stitched 3D reconstructed images are then post-processed, such as by denoising, contrast enhancement, and smoothing, to improve image quality and visualization. Through the above process, the image processing system 5 is able to accurately stitch all cross-sectional optical coherence tomography images in the depth direction, completing the 3D volume reconstruction of the in vitro human tumor specimen to be tested and obtaining a high-quality 3D reconstructed image.

[0082] In an optional embodiment, the stage elevation is less than the penetration depth of optical coherence tomography (OCT). OCT has a submillimeter penetration depth (1-2 mm), enabling high-resolution imaging of both superficial and shallow structures in biological tissue. The stage elevation, i.e., the elevation of the stage in the vertical Z direction (i.e., the thickness of each slice), is less than the OCT penetration depth and can be set, for example, to 0.2-1 mm.

[0083] Since the amount of elevation of the stage is less than the penetration depth of OCT imaging, after each elevation, the OCT imaging system can still image the surface of the new tissue after resection and a certain depth below it. This setting ensures that the imaging areas between adjacent sections overlap to a certain extent, avoiding information faults caused by slicing. By retaining some overlapping areas, the device can achieve continuous three-dimensional imaging in the vertical direction, ensuring the data integrity of the entire tumor specimen. The overlapping area provides key data support for subsequent image stitching and three-dimensional reconstruction. The image processing system 5 can use the overlapping areas between adjacent sections to accurately stitch the optical coherence tomography images of multiple sections into a complete three-dimensional volume data. This setting significantly improves the accuracy and continuity of the three-dimensional reconstructed image.

[0084] The operating process of the device of the present invention is as follows: the OCT imaging system images the first section of the tumor specimen, generating three-dimensional volume data. Due to the attenuation of light as it passes through tissue, to ensure imaging quality, images within 30% to 60% of the OCT imaging depth are considered effective images. The imaging depth covers the tissue region from the surface to a certain depth (e.g., 1.2 mm, with effective tissue imaging at 0.6 mm). The movable stage 3 is raised in the vertical Z direction to a certain height (e.g., 0.4 mm), which is less than the OCT imaging penetration depth. The slicing system 2 removes the imaged tissue (the thickness of the removal is the same as the amount of stage elevation, e.g., 0.2 mm). The OCT imaging system then images the new tissue surface. Because the stage elevation is less than the OCT imaging penetration depth, the new imaging area overlaps with the imaging area of the previous section to a certain extent (e.g., 0.2 mm). This overlap ensures data continuity and integrity. This process is repeated until the entire tumor specimen is imaged. The image processing system 5 uses the overlapping areas between adjacent sections to stitch the optical coherence tomography images from multiple sections into complete three-dimensional volume data. Structural analysis is performed based on three-dimensional volume data to extract the vascular morphology and distribution data of tumor tissue.

[0085] By setting the stage elevation to be less than the OCT imaging penetration depth, this invention allows imaging of the remaining tissue after each resection, avoiding information loss due to slicing. The design of the overlapping region significantly improves the accuracy of image stitching and 3D reconstruction, ensuring the continuity and accuracy of the final imaging results. By retaining some overlapping regions, the device can fully utilize the imaging capabilities of the OCT imaging system and maximize the extraction of structural information of tumor tissue.

[0086] In an optional embodiment, the image processing system 5 is used to: measure different cell layers based on the three-dimensional reconstructed image, perform blood vessel segmentation, and quantitatively calculate the morphology and distribution data of the blood vessels.

[0087] This cell layer measurement is based on 3D reconstructed images, identifying and measuring the thickness and distribution of different cell layers in tumor tissue. The process is as follows: The 3D reconstructed images are preprocessed with denoising and contrast enhancement to improve the accuracy of subsequent analysis, resulting in a preprocessed 3D reconstructed image. Image segmentation algorithms (such as deep learning-based semantic segmentation) are then used to identify different cell layers (e.g., epithelium, stroma, etc.). The thickness distribution of each cell layer in 3D space is calculated to generate a layered thickness map. This layered thickness map provides detailed thickness information for different cell layers in tumor tissue, helping researchers understand the tumor's tissue structure and pathological characteristics.

[0088] Vascular segmentation extracts and accurately segments vascular structures from 3D reconstructed images. The process is as follows: The 3D reconstructed image is processed using a filtering algorithm (such as a Frangi filter) to enhance the vascular structure and suppress background noise, resulting in a processed 3D reconstructed image. The processed 3D reconstructed image is segmented using a deep learning-based semantic segmentation model (such as U-Net) or a traditional image segmentation algorithm (such as region growing). Morphological operations (such as removing small connected regions and filling holes) are performed on the segmentation results to optimize the segmentation effect. This accurately extracts vascular structures from tumor tissue, providing basic data for subsequent quantitative analysis. Based on the segmented vascular structure, morphological parameters (such as diameter, length, and curvature) and distribution characteristics (such as density and directionality) of the vessels are calculated.

[0089] Specifically, the segmented vessels are skeletonized and their centerlines are extracted. Based on these centerlines, various morphological parameters (such as diameter and length) are calculated. This includes calculating the local diameter along the vessel centerline as the vessel diameter, calculating the total length of the vessel centerline as the vessel length, and calculating the vessel length or volume percentage within a unit volume. The main orientation and distribution of the vessels are analyzed to obtain vascular distribution data.

[0090] This paper takes the U-Net-based blood vessel segmentation and quantitative analysis as an example to illustrate: Step 1: Data Preparation: Obtain 3D reconstructed images and annotate the vascular regions as training data. Perform image normalization and data augmentation (such as rotation and flipping) to increase the diversity of training samples.

[0091] Step 2: U-Net model training: Use the labeled data to train the U-Net model. U-Net is a classic semantic segmentation network with an encoder and decoder architecture, suitable for processing medical images. Use Dice loss or cross-entropy loss to optimize vessel segmentation accuracy. Use the Adam optimizer for model training.

[0092] Step 3: Blood vessel segmentation: Apply the trained U-Net model to the 3D reconstructed image to generate blood vessel segmentation results.

[0093] Step 4: Vessel Skeletonization: Use a skeletonization algorithm (such as the Zhang-Suen algorithm) to extract the centerline of the segmented blood vessels. Smooth the skeletonized results to reduce noise.

[0094] Step 5: Morphological and distribution parameter calculation: Diameter calculation: Calculates local diameter along the vessel centerline using either a distance transform algorithm or cross-section-based diameter measurement.

[0095] Length calculation: Calculate the total length of the blood vessel centerline.

[0096] Curvature Calculation: Calculates the curvature distribution using the derivative of the centerline.

[0097] Density calculation: Calculate the length or volume percentage of blood vessels within a unit volume.

[0098] Directionality analysis: Principal component analysis (PCA) or direction histogram was used to analyze the main directions of blood vessels.

[0099] Step 6: Results visualization and output: Visualize the vessel segmentation, skeletonization, and quantitative analysis results to generate 3D rendered images and statistical charts. Output quantitative analysis data (such as vessel diameter, length, and density) for further research.

[0100] By combining the U-Net segmentation algorithm, skeletonization technology, and quantitative analysis methods, the image processing system 5 is capable of performing cell layer measurement, vascular segmentation, and quantitative calculation of vascular morphology and distribution on 3D reconstructed images. The U-Net-based vascular segmentation algorithm accurately extracts vascular structures in tumor tissue, achieving pixel-level segmentation accuracy. By calculating vascular morphology and distribution parameters, it provides detailed quantitative data, helping researchers understand the abnormal characteristics of tumor vessels. The entire process is automated, significantly improving analysis efficiency and reducing manual errors.

[0101] Reference Figure 2 As shown, the entire working process of the device for performing three-dimensional imaging and structural analysis of human tumor specimens in vitro is as follows: 1. Preparation 1. Start: The process starts and proceeds to the next step.

[0102] 2. Sample preparation and embedding: Human tumor specimens in vitro are pre-processed, such as fixation, dehydration, and embedding, to facilitate subsequent sectioning and imaging.

[0103] 3. Sample end surface flatness and trimming: Check whether the sample end surface is flat. If it is uneven, perform trimming to ensure subsequent imaging quality.

[0104] 2. Imaging Stage 4. Start the main control system 4: Start the main control programs of the imaging system and the slicing system 2, and prepare for imaging and slicing operations.

[0105] 5. Set slicing parameters and imaging parameters: According to the imaging effect and needs, adjust the slicing parameters (such as slice thickness) and imaging parameters (such as imaging depth, resolution, etc.) to optimize the imaging quality.

[0106] 6. Generate multiple continuous imaging position coordinates according to the preset imaging range and imaging parameters.

[0107] 7. Move to slicing position: Move the sample to the first slicing position.

[0108] 8. Start slicing: Slicing system 2 slices the sample.

[0109] 9. Determine whether the current slicing is completed: Determine whether the sample slice has been completed. If completed, proceed to step 10.

[0110] 10. Start imaging.

[0111] 11. Move to the imaging position coordinates (X, Y) and use the OCT (optical coherence tomography) imaging system to image the sliced sample and obtain image data of the section.

[0112] 12. Whether all sections of the current row have been imaged: Determine whether all sections of the current row have been imaged. If not, move to the next slice position and repeat step 11 until all sections of the current row have been imaged.

[0113] 13. Whether imaging of all positions in the current section has been completed: Determine whether imaging of all positions in the current section has been completed. If not, repeat steps 11-12 until imaging of all positions in the current section has been completed.

[0114] 14. Whether imaging of all sections is completed: Determine whether imaging of all sections is completed. If not, repeat steps 7-13 until imaging of all sections is completed.

[0115] 3.3D Reconstruction Stage 15. Complete single cross-sectional image stitching: stitch multiple imaging results of the same cross-section to obtain a complete cross-sectional image.

[0116] 16. Complete cross-sectional stitching based on 3D reconstruction algorithms: Use 3D reconstruction algorithms to process the acquired cross-sectional images and generate a 3D image of each section. All 3D cross-sectional images are stitched together to form a complete 3D image of the ex vivo human tumor specimen.

[0117] 4. Subsequent Analysis Phase 17. Statistical analysis of vascular distribution using ITK-SNAP: Use ITK-SNAP and other software to segment and count blood vessels in 3D reconstructed images and analyze their distribution characteristics. Evaluate cell stratification, observing and analyzing the structure and characteristics of different cell layers in 3D reconstructed images.

[0118] Through the above steps, three-dimensional imaging and structural analysis of human tumor specimens in vitro can be achieved, providing strong support for subsequent medical research and clinical practice.

[0119] The following describes, with reference to specific implementation examples, a device for three-dimensional imaging and structural analysis of ex vivo human tumor specimens. This device provides a label-free, rapid imaging, and automated sectioning method, along with three-dimensional reconstruction and quantitative tissue structure calculation capabilities. This provides a quantitative basis for accurately assessing the three-dimensional morphology, cell stratification, and vascular distribution of ex vivo human tumor tissue specimens. The device comprises the following steps: Step 1): Embed the tissue with 10% gelatin. Then, fix the embedded sample with 4% formaldehyde to increase the degree of cross-linking between the tissue specimen and the embedded sample. After the sample preparation is completed, use 502 glue to fix it to the sample placement table of the experimental pool, and use the sectioning system 2 to flatten the end surface. Step 2): Using the parameter input module of the main control system 4, the user enters the imaging field size of 5mm, the imaging redundancy of a single slice section of 20%, the slice thickness of 0.4mm, the slice speed of 0.5mm / s, and the tissue geometry. Based on the user's input, the main control system 4 automatically generates the coordinates of the respective imaging positions in the XY plane and the number of slices; Step 3): According to the parameters set in step 2), 7 scanning positions per row and 6 scanning positions per column are used, and the column scanning sequence with priority on Y direction scanning is used to perform C-scan imaging at 42 consecutive positions to obtain the XY plane imaging of each position as shown below: Figure 3a As shown, the resolution is 0.01x0.01mm 2 ; Step 4): After completing the imaging of a single slice plane in step 3), move stage 3 to the X, Y position at the time of slicing, and move the stage up 0.4 mm to cut off the current section that has been imaged, and repeat step 3) to complete C-scan imaging of 42 consecutive positions of the new section; Step 5): Repeat steps 3) to 4) until all tissue sections and post-section imaging are completed; Step 6): Perform data analysis on the imaging data obtained in steps 3) to 5), and use a single plane imaging stitching algorithm to complete the OCT stitching of each section, such as Figure 3b As shown; Step 7): Based on the OCT stitching images of all sections stitched in step 6), a three-dimensional image reconstruction algorithm is used to complete the three-dimensional volume reconstruction of all tissue imaging, such as Figure 4 As shown; Step 8): Use ITK-SNAP medical imaging software to view the three-dimensional reconstructed image, and use the "PaintBrush Mode" of the "Segmentation" tool to mark the tumor tissue cell layer and blood vessels, and then obtain information such as cell layer and blood vessel distribution, for example, Figure 5aThe image shows the tissue stratification in the cross-sectional OCT imaging of a certain section of tumor tissue. Figure 5b The three-dimensional rendering of the main blood vessels in the cross-sectional OCT imaging of this tissue is shown. Figure 5c This image shows the distribution of tiny blood vessels within a single slice. Because connective tissue and endothelium may be present within the blood vessels, the diameters of the blood vessels shown in cross-sectional imaging are not completely consistent.

[0120] The following describes the method for performing three-dimensional imaging and structural analysis of an in vitro human tumor specimen provided by the present invention. The method for performing three-dimensional imaging and structural analysis of an in vitro human tumor specimen described below and the device for performing three-dimensional imaging and structural analysis of an in vitro human tumor specimen described above can be used in correspondence with each other.

[0121] The present invention also provides a method for performing three-dimensional imaging and structural analysis on an in vitro human tumor specimen, which is implemented by the apparatus for performing three-dimensional imaging and structural analysis on an in vitro human tumor specimen as described above, comprising: S310, using the main control system 4 to generate a plurality of continuous imaging position coordinates according to the received imaging range and imaging parameters, and to generate an imaging control signal and a movement control signal, and to generate a slicing control signal according to the received slicing parameters; S320: The mobile stage 3 moves the in vitro human tumor specimen to be tested to each imaging position coordinate according to the movement control signal, and the optical coherence tomography imaging system 1 images the cross-section of the in vitro human tumor specimen to be tested according to each imaging position coordinate in the imaging control signal to obtain corresponding three-dimensional volume data; S330, after completing a cross-sectional imaging, using the movable stage 3 to adjust the height of the human tumor in vitro specimen to be tested; S340, after the human tumor in vitro specimen to be tested reaches a preset position, using the slicing system 2 to cut off the completed section of the human tumor in vitro specimen to be tested according to the slicing control signal to form a new imaging section; S350, imaging the next section of the in vitro human tumor specimen to be tested using the optical coherence tomography system 1 until imaging of all sections is completed; S360. Use the image processing system 5 to stitch the three-dimensional volume data of the imaging position coordinates of the same section to obtain an optical coherence tomography image of a single section, reconstruct the three-dimensional volume data based on the optical coherence tomography images of all sections to obtain a three-dimensional reconstructed image, and perform structural analysis based on the three-dimensional reconstructed image to obtain vascular morphology and distribution data.

[0122] In an optional embodiment, the imaging parameters include an imaging sequence within a single plane, an imaging field of view, and an imaging redundancy; and the step S360 described above of using the image processing system 5 to stitch the three-dimensional volume data of the imaging position coordinates of the same section to obtain an optical coherence tomography image of a single section includes: The image processing system splices the three-dimensional volume data of the imaging position coordinates of the same section according to the imaging sequence, imaging field of view and imaging redundancy in a single plane to obtain an optical coherence tomography image of a single section.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A device for three-dimensional imaging and structural analysis of human tumor specimens in vitro, characterized in that: include: A main control system is used to generate a plurality of continuous imaging position coordinates according to the received imaging range and imaging parameters, and to generate an imaging control signal and a movement control signal, and to generate a slicing control signal according to the received slicing parameters; An optical coherence tomography system is used to image a cross section of an in vitro human tumor specimen to be measured according to the imaging position coordinates in the imaging control signal to obtain corresponding three-dimensional volume data; a moving stage, used for moving the in vitro human tumor specimen to be tested to each imaging position coordinate according to the movement control signal, and adjusting the height of the in vitro human tumor specimen to be tested after completing a cross-sectional imaging; a slicing system for cutting off a completed section of the in vitro human tumor specimen to be tested according to the slicing control signal after the in vitro human tumor specimen to be tested reaches a preset position, thereby forming a new imaging section; An image processing system is used to stitch the three-dimensional volume data of the imaging position coordinates of the same section to obtain an optical coherence tomography image of a single section, reconstruct the three-dimensional volume data based on the optical coherence tomography images of all sections to obtain a three-dimensional reconstructed image, and perform structural analysis based on the three-dimensional reconstructed image to obtain vascular morphology and distribution data.

2. The device for performing three-dimensional imaging and structural analysis of human tumor specimens in vitro according to claim 1, characterized in that: The imaging parameters include imaging sequence, imaging field of view, and imaging redundancy within a single plane; The image processing system is used to splice the three-dimensional volume data of the imaging position coordinates of the same section according to the imaging sequence, imaging field of view and imaging redundancy in a single plane to obtain an optical coherence tomography image of a single section.

3. The device for performing three-dimensional imaging and structural analysis of human tumor specimens in vitro according to claim 1, characterized in that: The imaging parameters also include the amount of stage elevation, the order of cross-sectional imaging, and the redundancy between the depths of two adjacent cross-sectional images; The image processing system is used to perform depth-wise splicing of the optical coherence tomography images of all sections based on the amount of stage elevation, the order of section imaging, and the depth redundancy between two adjacent section images, until the three-dimensional volume data reconstruction of the in vitro human tumor specimen to be tested is completed to obtain a three-dimensional reconstructed image.

4. The device for performing three-dimensional imaging and structural analysis of human tumor specimens in vitro according to claim 3, characterized in that: The amount by which the stage is elevated is less than the penetration depth of optical coherence tomography.

5. The device for performing three-dimensional imaging and structural analysis of human tumor specimens in vitro according to claim 1, characterized in that: The image processing system is used to: Based on the three-dimensional reconstructed image, different cell layers are measured, blood vessels are segmented, and the morphology and distribution data of the blood vessels are quantitatively calculated.

6. The device for performing three-dimensional imaging and structural analysis of human tumor specimens in vitro according to claim 1, characterized in that: The human tumor in vitro specimen sample to be tested is obtained by the following method: embedding the human tumor in vitro specimen tissue to be tested, and fixing the embedded sample to obtain the human tumor in vitro specimen sample to be tested.

7. The device for performing three-dimensional imaging and structural analysis of human tumor specimens in vitro according to claim 1, characterized in that: The human tumor in vitro specimen to be tested is immersed in a buffer solution, and imaging and slicing are both completed in the buffer solution.

8. The device for performing three-dimensional imaging and structural analysis of human tumor specimens in vitro according to claim 1, characterized in that: The imaging parameters also include tissue geometry; The main control system is further configured to determine the number of slices according to the received slice parameters and the geometric dimensions of the tissue.

9. A method for performing three-dimensional imaging and structural analysis on an in vitro human tumor specimen, implemented by the apparatus for performing three-dimensional imaging and structural analysis on an in vitro human tumor specimen according to any one of claims 1 to 8, characterized in that: include: Using the main control system to generate a plurality of continuous imaging position coordinates according to the received imaging range and imaging parameters, and to generate an imaging control signal and a movement control signal, and to generate a slicing control signal according to the received slicing parameters; The mobile stage moves the in vitro human tumor specimen to be tested to each imaging position coordinate according to the movement control signal, and the optical coherence tomography imaging system images the cross section of the in vitro human tumor specimen to be tested according to each imaging position coordinate in the imaging control signal to obtain corresponding three-dimensional volume data; After completing a cross-sectional imaging, the height of the human tumor in vitro specimen to be tested is adjusted using a movable stage; After the human tumor in vitro specimen sample to be tested reaches a preset position, a slicing system is used to cut off a completed section of the human tumor in vitro specimen sample to be tested according to the slicing control signal to form a new imaging section; Using an optical coherence tomography system to image the next section of the human tumor in vitro specimen to be tested, until imaging of all sections is completed; An image processing system is used to stitch the three-dimensional volume data of the imaging position coordinates of the same section to obtain an optical coherence tomography image of a single section. Three-dimensional volume data reconstruction is performed based on the optical coherence tomography images of all sections to obtain a three-dimensional reconstructed image. Structural analysis is performed based on the three-dimensional reconstructed image to obtain vascular morphology and distribution data.

10. The method for three-dimensional imaging and structural analysis of human tumor specimens in vitro according to claim 9, characterized in that: The imaging parameters include the imaging sequence, imaging field of view, and imaging redundancy within a single plane; and the optical coherence tomography image of a single section is obtained by splicing the three-dimensional volume data of the imaging position coordinates of the same section using the image processing system, including: The image processing system splices the three-dimensional volume data of the imaging position coordinates of the same section according to the imaging sequence, imaging field of view and imaging redundancy in a single plane to obtain an optical coherence tomography image of a single section.

Images