Automatic microdissection method for tumor cell sampling, and fully automatic tissue dissection device

Abstract

Disclosed in the present invention are an automatic microdissection method for tumor cell sampling, and a device. The method comprises: acquiring a WSI; identifying all tumor regions and positions thereof on the WSI; using a visual monitoring apparatus to acquire an initial image; obtaining tumor regions and positions thereof on the initial image; identifying the position of a needle tip of a dissection needle on the initial image; performing calculation by means of a path planning algorithm to obtain a motion path of the dissection needle; performing dissection on a glass slide; during the dissection, using the visual monitoring apparatus to acquire a real-time image, and comparing the real-time image with the initial image to obtain a deformation amount of a normal tissue region; when the deformation amount exceeds a threshold value, stopping the dissection, and when the deformation amount does not exceed the threshold value, continuing dissection until the dissection of the current tumor region is completed; and after the dissection of the current tumor region is completed, moving the glass slide, and repeating the steps until the dissection of all the tumor regions on the glass slide is completed. The present invention can identify the positions of tumor regions, thereby realizing high-precision positioning and fully automatic dissection.

Description

Microscopic Automated Dissection Method and Fully Automated Tissue Dissection Equipment for Tumor Cell Sampling Technical Field

[0001] This invention belongs to the field of biomedical technology, specifically relating to a microscopic automated cutting method and a fully automated tissue cutting device for tumor cell sampling. Background Technology

[0002] With the continuous advancement of biotechnology and medical research, biomedicine has progressed to the stage of exploring the pathogenesis of diseases and revealing the essence of life phenomena at the molecular level. In this process, isolating the desired cell population from tissue sections or cell smears is crucial for subsequent molecular biological analysis. Currently, laser capture microdissection and mechanical microdissection are two commonly used cell isolation methods.

[0003] Laser capture microdissection (LCMS) utilizes a laser beam projected vertically downwards from above a slide, with the beam landing on the area to be cut. As the thin film in that area is peeled away, the cells attached to it are also cut off. This technique offers advantages such as precise cutting, flexible operation, good sample integrity, and compatibility with various analytical techniques. However, it also has drawbacks including high equipment costs, demanding operator skills, relatively low cutting throughput, and the risk of thermal damage.

[0004] Mechanical microdissection technology uses specialized micro-blades, blades, or cutting needles to directly cut tissues or cells under a microscope. This technique is simple, intuitive, relatively inexpensive, and easy to understand. However, the cutting process is manual, and the cutting precision is affected by individual operator differences. It is difficult to handle small, delicate, and dispersed cell populations, resulting in low cutting efficiency.

[0005] Furthermore, both laser-guided microdissection and mechanical microdissection present a trade-off between field of view and efficiency. Using a higher magnification reduces the field of view and decreases operational efficiency. Conversely, while a lower magnification expands the field of view, the image magnification is too low, affecting observation accuracy.

[0006] Existing laser-captured microdissection technology has the following problems:

[0007] First, existing laser capture microdissection technology equipment is expensive, requiring significant investment in the instrument itself, laser source, high-precision microscope, and professional control software. Furthermore, subsequent maintenance costs are substantial, limiting its widespread application in laboratories or research institutions with limited budgets.

[0008] Secondly, laser capture microdissection technology has high requirements for operators. They need to be proficient in microscope operation skills, laser parameter settings, and sample processing procedures. They need to undergo professional and long-term training before they can master it. The operation requirements are high.

[0009] Third, laser capture microdissection technology has a relatively low cutting throughput, and can only process samples from a small area at a time. It is inefficient for large-scale, high-throughput sample collection tasks.

[0010] Fourth, laser capture microdissection technology carries a certain risk of thermal damage. When used for a long time and under high intensity, it may cause thermal damage to the sample, affecting the activity of intracellular biomolecules and interfering with subsequent molecular biology detection results.

[0011] Fifth, the mechanical micro-cutting technology involves manual operation, and the cutting accuracy is affected by individual differences in operators. It is difficult to handle small, delicate and dispersed cell groups, resulting in low cutting efficiency.

[0012] Therefore, there is an urgent need for a new type of microdissection technology that can maintain the advantages of high-precision dissection, flexible operation, and good sample integrity while reducing equipment costs and increasing automation, thereby meeting the demand for high-quality biological samples in biomedical research. Summary of the Invention

[0013] To address the aforementioned technical problems, this invention proposes a microscopic automated cutting method for tumor cell sampling and a fully automated tissue cutting device.

[0014] To achieve the above objectives, the technical solution of the present invention is as follows:

[0015] On one hand, this invention discloses a microscopic automated cutting method for tumor cell sampling, comprising:

[0016] Step S1: Scan the slide using a digital microscope to obtain a WSI image;

[0017] Step S2: Using a pre-trained tumor recognition model, identify all tumor regions and their locations on the WSI image;

[0018] Step S3: Place one or more glass slides on and fix them on the platform, and use one or more visual monitoring devices to acquire an initial image I0, which shows the slices and cutting needles.

[0019] Step S4: Using an image registration algorithm, determine the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image, and obtain the tumor region and its location on the initial image I0;

[0020] Step S5: Identify the position of the cutting needle tip on the initial image I0 using a needle tip recognition algorithm;

[0021] Step S6: Using the position of the cutting needle tip on the initial image I0 as the initial position and the position of the tumor area to be cut on the initial image I0 as the target position, the cutting needle movement path is calculated by the path planning algorithm.

[0022] Step S7: The cutting needle moves to the target position according to the obtained cutting needle movement path and falls to cut on the corresponding glass slide;

[0023] During the cutting process, a visual monitoring device is used to acquire real-time images. t and will real-time image I t By comparing with the initial image I0, the deformation of the normal tissue region is obtained.

[0024] When the deformation exceeds the threshold, the cutting stops; otherwise, cutting continues until the current tumor area is completely cut.

[0025] Step S8: After the current tumor area is cut, move the slide and repeat steps S3-S7 until all tumor areas on the slide are cut to achieve tumor cell sampling.

[0026] Based on the above technical solution, the following improvements can be made:

[0027] As a preferred embodiment, step S4 includes:

[0028] Step S4.1: Using the initial image I0 as the reference image and the WSI image as the floating image, the initial image I0 and the WSI image are registered using an image registration algorithm;

[0029] Step S4.2: After image registration, determine the position of the initial image I0 on the WSI image, and obtain the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image;

[0030] Step S4.3: Based on the relative relationship between the two coordinate systems, and based on all tumor regions and their locations on the WSI image, obtain the tumor regions and their locations on the initial image I0.

[0031] As a preferred embodiment, step S5 includes:

[0032] Step S5.1: Use the image of the cutting needle tip as the template image, slide it pixel by pixel on the initial image I0, and use a similarity measurement method to calculate and record the matching degree between the sliding area of ​​the template image and the initial image I0 at each position;

[0033] Step S5.2: Find the position with the highest matching degree, which is the position of the cutting needle in the initial image I0.

[0034] As a preferred embodiment, step S7 includes:

[0035] Step S7.1: During the cutting process, acquire real-time images using a visual monitoring device. t The real-time normal tissue region N is obtained according to the following formula. t N t =I t -T0;

[0036] Where: T0 is the tumor region on the initial image I0;

[0037] Step S7.2: Obtain the initial normal tissue region N0 according to the following formula: N0 = I0 - T0;

[0038] Step S7.3: Calculate the real-time normal tissue region N t The average of the sum of squares of the differences between the pixel values ​​corresponding to the initial normal tissue region N0 and the normal tissue region is used to obtain the deformation of the normal tissue region. This deformation is then compared with a threshold. Cutting is stopped when the deformation exceeds the threshold.

[0039] Step S7.4: Repeat steps S7.1-S7.3 until the current tumor area is completely cut.

[0040] As a preferred embodiment, step S6 includes:

[0041] Step S6.1: Extract features from the portion of the tumor area to be cut to obtain tumor features;

[0042] Step S6.2: Use the trained tumor complexity judgment model to judge the complexity of the tumor region to be cut, and output the complexity level T and the number of cutting needles n;

[0043] Step S6.3: Select the n cutting needles closest to the tumor area to be cut as the cutting needles to be operated, and assign a corresponding cutting area to each cutting needle to be operated;

[0044] Step S6.4: Using the needle tip position of each cutting needle to be moved on the initial image I0 as its initial position, and the position of the cutting area of ​​the tumor region to be cut on the initial image I0 as its target position, the cutting needle movement path of each cutting needle to be moved is calculated by the path planning algorithm.

[0045] On the other hand, this invention discloses a microscopic automated cutting device for tumor cell sampling, comprising:

[0046] The WSI image acquisition module is used to scan glass slides using a digital microscope to acquire WSI images;

[0047] The tumor region identification module is used to identify all tumor regions and their locations on WSI images using a pre-trained tumor identification model.

[0048] The initial image acquisition module is used to place one or more glass slides on and fix them on the platform, and to acquire an initial image I0 using one or more visual monitoring devices. The initial image I0 shows the slice and cutting needle.

[0049] The image registration module is used to determine the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image through an image registration algorithm, and to obtain the tumor region and its location on the initial image I0.

[0050] The cutting needle recognition module is used to identify the position of the cutting needle tip on the initial image I0 using a needle tip recognition algorithm;

[0051] The path planning module is used to calculate the movement path of the cutting needle by taking the position of the cutting needle tip on the initial image I0 as the initial position and the position of the tumor area to be cut on the initial image I0 as the target position.

[0052] The cutting module is used to move the cutting needle to the target position according to the obtained cutting needle movement path and then fall to cut on the corresponding glass slide;

[0053] The cutting monitoring module is used to acquire real-time images using a visual monitoring device during the cutting process. t and will real-time image I t By comparing with the initial image I0, the deformation of the normal tissue region is obtained.

[0054] When the deformation exceeds the threshold, the cutting stops; otherwise, cutting continues until the current tumor area is completely cut.

[0055] The repeat execution module is used to move the slide after the current tumor area has been cut, and repeat the methods in the initial image acquisition module, image registration module, cutting needle identification module, path planning module, cutting module, and cutting monitoring module until all tumor areas on the slide have been cut, thus achieving tumor cell sampling.

[0056] As a preferred embodiment, the image registration module includes:

[0057] The image registration unit is used to register the initial image I0 and the WSI image as a floating image using an image registration algorithm.

[0058] The coordinate system corresponding unit is used to determine the position of the initial image I0 on the WSI image after image registration, and to obtain the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image.

[0059] The initial image tumor region determination unit is used to obtain the tumor region and its location on the initial image I0 based on the relative relationship between the two coordinate systems and all tumor regions and their locations on the WSI image.

[0060] As a preferred embodiment, the cutting needle recognition module includes:

[0061] The template matching unit is used to slide the image of the cutting needle tip as a template image on the initial image I0 pixel by pixel, and to calculate and record the matching degree between the template image and the sliding area of ​​the initial image I0 at each position using a similarity measurement method.

[0062] The cutting needle position determination unit is used to find the position with the highest matching degree, which is the position of the cutting needle in the initial image I0.

[0063] As a preferred embodiment, the cutting monitoring module includes:

[0064] The real-time normal tissue acquisition unit is used to acquire real-time images using a visual monitoring device during the cutting process. t The real-time normal tissue region N is obtained according to the following formula. t N t =I t -T0;

[0065] Where: T0 is the tumor region on the initial image I0;

[0066] The initial normal tissue acquisition unit is used to obtain the initial normal tissue region N0 according to the following formula: N0 = I0 - T0;

[0067] The deformation calculation unit is used to calculate the real-time normal tissue region N. t The average of the sum of squares of the differences between the pixel values ​​corresponding to the initial normal tissue region N0 and the normal tissue region is used to obtain the deformation of the normal tissue region. This deformation is then compared with a threshold. Cutting is stopped when the deformation exceeds the threshold.

[0068] The repeat execution unit is used to repeat the methods in the real-time normal tissue acquisition unit, the initial normal tissue acquisition unit, and the deformation calculation unit until the current tumor region is completely cut.

[0069] As a preferred option, the path planning module includes:

[0070] The feature extraction unit is used to extract features from the portion of the tumor region to be cut, thereby obtaining tumor features.

[0071] The complexity identification unit is used to judge the complexity of the tumor region to be cut using a trained tumor complexity judgment model, and outputs the complexity level T and the number of cutting needles n.

[0072] The cutting needle selection unit is used to select the n cutting needles closest to the tumor area to be cut as the cutting needles to be operated, and to assign a corresponding cutting area to each cutting needle to be operated;

[0073] The path planning unit is used to calculate the movement path of each cutting needle by taking the needle tip position of each cutting needle to be moved on the initial image I0 as its initial position and the position of the cutting area of ​​the tumor region to be cut on the initial image I0 as its target position, through the path planning algorithm.

[0074] On the other hand, this invention discloses a fully automatic tissue cutting device, comprising: a housing and a cutting assembly. A door is hinged to one side of the housing, and an observation window is provided at the top of the door. A base plate is fixed to the lower end of the interior of the housing, and a bracket is connected to the top of the base plate. A high-definition camera is fixed to the top of the bracket, and a transport assembly is located below the high-definition camera. The cutting assembly is located on the top of the base plate and includes an X-axis electric slide, a Y-axis electric slide, a Z-axis electric slide, a mounting base, an optically coupled precision fine-tuning track, a drive base, a precision linear guide, a high-precision optical axis, a support spring, a connecting cover, and a scraper holder. A [missing information - likely a device name or component] is mounted on the top of the base plate. An X-axis electric slide is provided, and the output end of the X-axis electric slide is connected to a Y-axis electric slide. The output end of the Y-axis electric slide is fixed to a Z-axis electric slide, and the output end of the Z-axis electric slide is connected to a mounting base. One side of the mounting base is provided with an optical coupler lifting precision fine-tuning track, and the output end of the optical coupler lifting precision fine-tuning track is connected to a drive base. One side of the drive base is slidably connected to a precision linear guide, and the middle of the drive base is slidably connected to a high-precision optical axis. Support springs are sleeved on the outer sides of the upper and lower ends of the high-precision optical axis, and the high-precision optical axis is fixedly connected to the precision linear guide. The other side of the drive base is provided with a connecting cover, and a scraper needle seat is embedded inside the connecting cover.

[0075] Furthermore, the transport assembly includes a transport motor, a synchronous pulley set, a transport platform, guide rails, and LED lights. The transport motor is mounted on the top of the base plate, and the output end of the transport motor is connected to the synchronous pulley set. The transport platform is provided on the top surface of the middle part of the synchronous pulley set, and guide rails are slidably connected to both ends of the bottom of the transport platform. LED lights are installed inside the transport platform.

[0076] Furthermore, the number of LED lights is set to four, and the LED lights are evenly distributed along the top of the transport platform.

[0077] Furthermore, a connecting component is provided on the outer side of the scraper needle holder, and the connecting component includes an anti-slip groove, a positioning hole, a ball screw, and a foolproof opening. Anti-slip grooves are provided on both sides of the middle part of the scraper needle holder, and a positioning hole is provided on the upper outer side of the scraper needle holder. A ball screw abuts inside the positioning hole, and the ball screw is threadedly connected to the connecting cover. A foolproof opening is provided at the lower part of the connecting cover.

[0078] Furthermore, the two sides of the middle part of the scraper seat are arc-shaped, and the anti-slip grooves are equidistantly distributed along the arc surface of the scraper seat.

[0079] Furthermore, a partition is fixed inside the housing, with a first power source installed at one lower end of the partition and a second power source installed at the other lower end of the partition.

[0080] Furthermore, a control board is fixed to one upper end of the partition plate, and a driver is installed at the other upper end of the partition plate.

[0081] Furthermore, a fan is installed on the upper part of the other side of the enclosure, and an interface is provided at one end of the other side of the enclosure, while a switch is provided on the lower part of the other side of the enclosure.

[0082] This invention discloses a microscopic automated cutting method and device for tumor cell sampling, which has the following beneficial effects:

[0083] First, the present invention can accurately identify the location of the tumor area, achieving high-precision positioning and fully automatic cutting.

[0084] Secondly, this invention uses an image registration algorithm to associate WSI images with images acquired by a visual monitoring device, mapping the tumor region on the WSI image onto the initial image, thus ensuring high precision and full automation of the cutting process.

[0085] Third, the present invention uses a visual monitoring device to monitor the cutting process, based on the deformation of the normal tissue area, to ensure the stability of the cutting process.

[0086] Fourth, the equipment used in this invention is inexpensive, easy to operate, has a high cutting throughput, avoids laser thermal damage, and provides precise cutting.

[0087] Fifth, this invention, through the setting of a high-definition camera, can observe the position of the glass slide on the transport component. At this time, the scraper on the scraper holder can move along the three axes through the X-axis electric slide, Y-axis electric slide and Z-axis electric slide, thereby automatically controlling the scraper to perform related cutting operations on the tissue on the glass slide, improving the automation level of the equipment. Furthermore, the scraper can be finely adjusted through the optical coupler lifting precision fine-tuning track, thereby improving the cutting accuracy. At the same time, the transport platform is set with multiple stations, which can continuously process multiple samples, thus eliminating the need for frequent shutdowns and improving the overall processing efficiency. In addition, an LED light is set under the glass slide to provide supplemental lighting under the tissue, which is beneficial to ensure the effect of high-definition camera shooting.

[0088] Sixth, through the design of the connecting components, the ball screw can use its internal spring to drive the ball to embed into the positioning hole, which can quickly position the scraper holder and the connecting cover. At the same time, the anti-fooling notch ensures that the scraper is fixed in orientation during installation and will not obstruct the microscope's field of view after installation. Furthermore, when it is necessary to replace consumables such as scraper blades, the anti-slip groove can increase the friction between the hand and the scraper holder. Then, by pulling down the scraper holder, the ball of the ball screw can be moved out of the positioning hole. Therefore, no additional tools are needed during disassembly, which improves the convenience of disassembling and replacing the scraper holder. Attached Figure Description

[0089] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0090] Figure 1 is a flowchart of the automatic microscopic cutting method provided in an embodiment of the present invention.

[0091] Figure 2 is one of the structural schematic diagrams of the micro-dissection instrument provided in the embodiment of the present invention.

[0092] Figure 3 is a schematic diagram of tumor region identification on WSI image provided by an embodiment of the present invention.

[0093] Figure 4 is a schematic diagram of the initial image I0 provided in an embodiment of the present invention.

[0094] Figure 5 shows the effect of image registration provided by an embodiment of the present invention.

[0095] Figure 6 is a schematic diagram of needle tip identification of the cutting needle provided in an embodiment of the present invention.

[0096] Figure 7(a) is a schematic diagram of the back-and-forth cutting of the cutting needle provided in an embodiment of the present invention.

[0097] Figure 7(b) is a schematic diagram of the cutting needle cutting in the same direction according to an embodiment of the present invention.

[0098] Figure 8 is a second schematic diagram of the structure of the micro-dissection instrument provided in the embodiment of the present invention.

[0099] Figure 9 is a third schematic diagram of the structure of the micro-dissection instrument provided in the embodiment of the present invention.

[0100] Figure 10 is a schematic diagram of the overall three-dimensional structure of a fully automatic tissue cutting device according to the present invention.

[0101] Figure 11 is a schematic diagram of the internal structure of the box of a fully automatic tissue cutting device according to the present invention.

[0102] Figure 12 is a three-dimensional structural diagram of the transport component of a fully automatic tissue cutting device according to the present invention.

[0103] Figure 13 is a schematic diagram of the connection component structure of a fully automatic tissue cutting device according to the present invention.

[0104] Figure 14 is a schematic diagram of the overall rear view structure of a fully automatic tissue cutting device according to the present invention.

[0105] The components are as follows: A-Visual monitoring device, B-Cutting needle control device, C-Algorithm processing device, D-Supplemental light source, E-Slide placement platform, F-Slide, G-Cutting needle, 1-Box body; 2-Box door; 3-Observation window; 4-Base plate; 5-Bracket; 6-High-definition camera; 7-Transportation component; 701-Transportation motor; 702-Synchronous belt pulley set; 703-Transportation stage; 704-Guide rail; 705-LED light; 8-Cutting component; 801-X-axis electric slide; 802-Y-axis electric slide; 803-Z 804-Mounting base; 805-Optical coupler lifting precision fine-tuning track; 806-Drive base; 807-Precision linear guide; 808-High-precision optical axis; 809-Support spring; 810-Connecting cover; 811-Scraper needle seat; 9-Connecting assembly; 901-Anti-slip groove; 902-Positioning hole; 903-Ball screw; 904-Anti-foolproof notch; 10-Partition plate; 11-First power supply; 12-Second power supply; 13-Control board; 14-Driver; 15-Fan; 16-Interface; 17-Switch. Detailed Implementation

[0106] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0107] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0108] The expression “includes” is an “open-ended” expression, which means that there is a corresponding component or step, and should not be interpreted as excluding additional components or steps.

[0109] To achieve the objectives of this invention, in some embodiments of the automated microsurgical cutting method for tumor cell sampling, as shown in FIG1, the automated microsurgical cutting method includes:

[0110] Step S101: Scan the slide using a digital microscope to obtain a WSI image;

[0111] Step S102: Using a pre-trained tumor recognition model, identify all tumor regions and their locations on the WSI image;

[0112] Step S103: Place a glass slide on and fix it on the platform, and use a visual monitoring device 1 to acquire an initial image I0, which shows a slice and a cutting needle.

[0113] Step S104: Using an image registration algorithm, determine the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image, and obtain the tumor region and its location on the initial image I0;

[0114] Step S105: Identify the position of the cutting needle tip on the initial image I0 using a needle tip recognition algorithm;

[0115] Step S106: Using the position of the cutting needle tip on the initial image I0 as the initial position and the position of the tumor area to be cut on the initial image I0 as the target position, the cutting needle movement path is calculated by the path planning algorithm.

[0116] Step S107: The cutting needle moves to the target position according to the obtained cutting needle movement path and falls to cut on the corresponding glass slide;

[0117] During the cutting process, a visual monitoring device 1 is used to acquire real-time images I. t and will real-time image I t By comparing with the initial image I0, the deformation of the normal tissue region is obtained.

[0118] When the deformation exceeds the threshold, the cutting stops; otherwise, cutting continues until the current tumor area is completely cut.

[0119] Step S108: After the current tumor area is cut, move the slide and repeat steps S103-S107 until all tumor areas on the slide are cut to achieve tumor cell sampling.

[0120] Existing technologies suffer from problems such as high equipment costs, demanding operational requirements, low cutting throughput, risk of thermal damage, and limited cutting precision. Therefore, this invention discloses a microscopic automatic cutting method based on intelligent algorithms, which employs a micro-cutting instrument. As shown in Figure 2, the micro-cutting instrument includes, but is not limited to: a platform, a visual monitoring device A, a cutting needle control device B, and an algorithm processing device C.

[0121] The platform includes a base, a supplementary lighting source D, and a slide placement platform E. The slide placement platform E is used to place and secure slides, and it is movable. The supplementary lighting source D can be an LED light source to provide supplementary lighting for the visual monitoring device A. The platform can move in the X and Y directions to facilitate sample replacement and movement. Simultaneously, it can be fine-tuned in the Z direction to help the visual monitoring device A focus.

[0122] The visual monitoring device A is located directly above the platform and includes, but is not limited to, an industrial camera (e.g., 20 megapixels) and a 10x objective lens to acquire images and monitor the glass slide F and cutting needle G on the platform in real time.

[0123] The cutting needle control device B includes a power supply, a drive motor, and a three-axis motion slider, which controls the cutting needle to move with high precision in the XYZ three-axis directions according to coordinate information for cutting tumor tissue. In some embodiments, a pressure sensor is installed in the Z-axis direction of the cutting needle to prevent the cutting needle from damaging the slide. The cutting needle 7 may, but is not limited to, use a tungsten carbide needle with a diameter of 0.3 mm.

[0124] In some other embodiments, the cutting needle G is mounted on an adapter fixture, and both the adapter fixture and the micro-cutting body are provided with mutually magnetic metal attraction points. Before micro-cutting, the adapter fixture with the cutting needle G is brought close to the micro-cutting body, and the two are magnetically aligned due to magnetic force. The metal attraction points of the adapter fixture and the metal attraction points of the micro-cutting body contact each other, forming a circuit, and the micro-cutting equipment can then operate.

[0125] The algorithm processing device C may include, but is not limited to, workstations equipped with an Intel Core i9 processor, 32GB of RAM, and an NVIDIA RTX 3090 graphics card. The algorithm processing device C has multiple built-in algorithms.

[0126] Each of the above steps will be described in detail below.

[0127] Step S101: Scan the slide using a digital microscope to obtain a whole-slice image, i.e., a WSI image.

[0128] Step S102 uses a pre-trained tumor recognition model to identify all tumor regions and their locations on the WSI image.

[0129] As shown in Figure 3, the tumor recognition model identifies all tumor cells in the WSI image and records the regions where these tumor cells are located as tumor regions. ) Since WSI-based AI-powered tumor recognition is a mature technology, it will not be elaborated upon here.

[0130] In step S103, a glass slide is placed and fixed on the platform, an appropriate amount of paraffin oil is dripped onto the sample, and the position of the platform is finely adjusted. By adjusting the XY direction, the sample is placed in the visual acquisition area of ​​the visual monitoring device. By adjusting the Z axis, the visual monitoring device can focus and acquire a clear initial image I0. The initial image I0 shows slices and cutting needles, as shown in Figure 4.

[0131] Step S104 uses an image registration algorithm to determine the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image, thereby obtaining the tumor region and its location on the initial image I0.

[0132] Specifically, step S104 includes:

[0133] Step S104.1: Using the initial image I0 as the reference image and the WSI image as the floating image, the initial image I0 and the WSI image are registered using an image registration algorithm (e.g., a registration algorithm based on SIFT feature operators), as shown in Figure 5.

[0134] Step S104.2: After image registration, determine the position of the initial image I0 on the WSI image, and obtain the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image;

[0135] Step S104.3: Based on the relative relationship between the two coordinate systems, and based on all tumor regions and their locations on the WSI image, obtain the tumor region and its location on the initial image I0, denoted as T0.

[0136] The image registration algorithms described above are mature technologies, so they will not be elaborated on here.

[0137] Step S105 uses a needle identification algorithm to identify the position of the cutting needle tip on the initial image I0. The needle identification algorithm can employ a template matching method.

[0138] Specifically, step S105 includes:

[0139] Step S105.1: Use the image of the cutting needle tip as the template image, slide it pixel by pixel on the initial image I0, and use a similarity measurement method to calculate and record the matching degree between the sliding area of ​​the template image and the initial image I0 at each position;

[0140] Step S105.2: Find the position with the highest matching degree, which is the position P0 of the cutting needle in the initial image I0, as shown in Figure 6.

[0141] Step S106 takes the position of the cutting needle tip on the initial image I0 as the initial position and the position of the tumor area to be cut on the initial image I0 as the target position, and calculates the cutting needle movement path through a path planning algorithm.

[0142] In step S107, the cutting needle is raised with the current needle position as the zero point. After moving to the target position according to the obtained cutting needle movement path, it falls and cuts on the corresponding glass slide with a cutting depth of 3-5 micrometers.

[0143] Within the T0 region of the tumor, the cutting needle is used to scrape line by line, such as back-and-forth scraping and unidirectional scraping. As shown in Figure 7(a), back-and-forth scraping is more efficient, but may be affected by the clearance of the control screw; as shown in Figure 7(b), unidirectional scraping is less efficient, but effectively avoids errors caused by the clearance of the control screw.

[0144] During the cutting process, a visual monitoring device is used to acquire real-time images. t and will real-time image I t Compared with the initial image I0, the deformation of the normal tissue area is obtained. When the deformation exceeds the threshold, the cutting stops; otherwise, the cutting continues until the tumor area is completely cut.

[0145] Specifically, step S107 includes:

[0146] Step S107.1: During the cutting process, a real-time image I is acquired using a visual monitoring device. t The real-time normal tissue region N is obtained according to the following formula. t N t =I t -T0;

[0147] Where: T0 is the tumor region on the initial image I0;

[0148] Step S107.2: Obtain the initial normal tissue region N0 according to the following formula: N0 = I0 - T0;

[0149] Step S107.3: Calculate the real-time normal tissue region N tThe average of the sum of squares of the differences between the pixel values ​​corresponding to the initial normal tissue region N0 and the normal tissue region is used to obtain the deformation of the normal tissue region. This deformation is then compared with a threshold. Cutting is stopped when the deformation exceeds the threshold.

[0150] Step S107.4: Repeat steps S107.1-S107.3 until the current tumor area is completely cut.

[0151] This invention assesses image deformation by calculating the mean square error (MSE) between the normal tissue region in a real-time image and the normal tissue region in an initial image. If the deformation is normal, segmentation continues; if the difference is significant, segmentation is interrupted to prevent further tissue damage.

[0152] After the current tumor area is cut in step S108, the slide is moved to the next area by moving the platform, and steps S103-S107 are repeated until all tumor areas on the slide are cut, thus achieving tumor cell sampling.

[0153] Collect the excised tumor cell tissue for molecular testing.

[0154] In some other embodiments, as shown in FIG8, multiple cutting needles can be used in coordination to perform cutting.

[0155] Specifically, step S106 includes:

[0156] Step S106.1: Extract features from the tumor region to be cut to obtain tumor features, including: morphological features (such as area, perimeter, density, irregularity of tumor boundaries, etc.), texture features (such as micro-texture features, etc.) and heterogeneity features (such as color differences in multiple regions, uneven distribution of nuclei, etc.).

[0157] Step S106.2: Use the trained tumor complexity judgment model to judge the complexity of the tumor region to be cut, and output the complexity level T and the number of cutting needles n;

[0158] For example, for low complexity, use one cutting pin;

[0159] Medium complexity, using 2 cutting pins;

[0160] High complexity, employing 3 cutting pins;

[0161] Step S106.3: Select the n cutting needles closest to the tumor area to be cut as the cutting needles to be operated, and assign a corresponding cutting area to each cutting needle to be operated;

[0162] Step S106.4: Using the needle tip position of each cutting needle to be moved on the initial image I0 as its initial position, and the position of the cutting area of ​​the tumor region to be cut on the initial image I0 as its target position, the cutting needle movement path of each cutting needle to be moved is calculated by the path planning algorithm.

[0163] Of course, in other embodiments, multiple cutting needles can simultaneously cut multiple tumor regions, using a one-to-one allocation method to improve cutting efficiency. Each cutting needle prioritizes cutting the tumor region closest to it to prevent interference during the cutting process.

[0164] In some other embodiments, as shown in FIG9, to improve the degree of automation, the present invention can cut multiple glass slides at once and can employ multiple visual monitoring devices.

[0165] Similar to the above embodiments, the difference lies in step S103.

[0166] Step S103 involves placing and fixing multiple glass slides on a platform, acquiring images using multiple visual monitoring devices, and stitching the acquired images together to obtain an initial image I0, which shows slices and cutting needles.

[0167] This invention discloses a microscopic automated cutting method for tumor cell sampling, which has the following beneficial effects:

[0168] First, the equipment is inexpensive, requiring only a regular camera and cutting needle, without the need for expensive laser light sources and high-precision microscopes. This significantly reduces equipment investment costs and facilitates its widespread application in more laboratories and research institutions.

[0169] Secondly, it is easy to operate and does not require long-term professional training. Simply place the slide on the platform, and it will automatically identify tumor cells and plan the cutting path. Operators do not need to be proficient in complex laser parameter settings and sample processing procedures.

[0170] Third, it has a high cutting throughput and can automatically cut the entire slice, which greatly improves cutting efficiency and meets the needs of large-scale, high-throughput sample collection.

[0171] Fourth, it avoids laser thermal damage. No laser irradiation is required during the cutting process, which will not cause thermal damage to the sample, thus ensuring the activity of intracellular biomolecules and providing reliable material for subsequent molecular biology detection.

[0172] Fifth, it has high cutting precision. With the help of artificial intelligence algorithms, it can accurately identify tumor cells on the slice and achieve precise cutting through path planning algorithms, minimizing contamination from surrounding irrelevant cells and ensuring sample purity. This makes the results of nucleic acid, protein and other macromolecule detection of the extracted sample closer to the actual state in vivo.

[0173] In other embodiments, the present invention discloses a microscopic automated cutting device for tumor cell sampling, comprising:

[0174] The WSI image acquisition module is used to scan glass slides using a digital microscope to acquire WSI images;

[0175] The tumor region identification module is used to identify all tumor regions and their locations on WSI images using a pre-trained tumor identification model.

[0176] The initial image acquisition module is used to place one or more glass slides on and fix them on the platform, and to acquire an initial image I0 using one or more visual monitoring devices. The initial image I0 shows the slice and cutting needle.

[0177] The image registration module is used to determine the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image through an image registration algorithm, and to obtain the tumor region and its location on the initial image I0.

[0178] The cutting needle recognition module is used to identify the position of the cutting needle tip on the initial image I0 using a needle tip recognition algorithm;

[0179] The path planning module is used to calculate the movement path of the cutting needle by taking the position of the cutting needle tip on the initial image I0 as the initial position and the position of the tumor area to be cut on the initial image I0 as the target position.

[0180] The cutting module is used to move the cutting needle to the target position according to the obtained cutting needle movement path and then fall to cut on the corresponding glass slide;

[0181] The cutting monitoring module is used to acquire real-time images using a visual monitoring device during the cutting process. t and will real-time image I t By comparing with the initial image I0, the deformation of the normal tissue region is obtained.

[0182] When the deformation exceeds the threshold, the cutting stops; otherwise, cutting continues until the current tumor area is completely cut.

[0183] The repeat execution module is used to move the slide after the current tumor area has been cut, and repeat the methods in the initial image acquisition module, image registration module, cutting needle identification module, path planning module, cutting module, and cutting monitoring module until all tumor areas on the slide have been cut, thus achieving tumor cell sampling.

[0184] Furthermore, the image registration module includes:

[0185] The image registration unit is used to register the initial image I0 and the WSI image as a floating image using an image registration algorithm.

[0186] The coordinate system corresponding unit is used to determine the position of the initial image I0 on the WSI image after image registration, and to obtain the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image.

[0187] The initial image tumor region determination unit is used to obtain the tumor region and its location on the initial image I0 based on the relative relationship between the two coordinate systems and all tumor regions and their locations on the WSI image.

[0188] As a preferred embodiment, the cutting needle recognition module includes:

[0189] The template matching unit is used to slide the image of the cutting needle tip as a template image on the initial image I0 pixel by pixel, and to calculate and record the matching degree between the template image and the sliding area of ​​the initial image I0 at each position using a similarity measurement method.

[0190] The cutting needle position determination unit is used to find the position with the highest matching degree, which is the position of the cutting needle in the initial image I0.

[0191] Furthermore, the cutting monitoring module includes:

[0192] The real-time normal tissue acquisition unit is used to acquire real-time images using a visual monitoring device during the cutting process. t The real-time normal tissue region N is obtained according to the following formula. t N t =I t -T0;

[0193] Where: T0 is the tumor region on the initial image I0;

[0194] The initial normal tissue acquisition unit is used to obtain the initial normal tissue region N0 according to the following formula: N0 = I0 - T0;

[0195] The deformation calculation unit is used to calculate the real-time normal tissue region N. t The average of the sum of squares of the differences between the pixel values ​​corresponding to the initial normal tissue region N0 and the normal tissue region is used to obtain the deformation of the normal tissue region. This deformation is then compared with a threshold. Cutting is stopped when the deformation exceeds the threshold.

[0196] The repeat execution unit is used to repeat the methods in the real-time normal tissue acquisition unit, the initial normal tissue acquisition unit, and the deformation calculation unit until the current tumor region is completely cut.

[0197] Furthermore, the path planning module includes:

[0198] The feature extraction unit is used to extract features from the portion of the tumor region to be cut, thereby obtaining tumor features.

[0199] The complexity identification unit is used to judge the complexity of the tumor region to be cut using a trained tumor complexity judgment model, and outputs the complexity level T and the number of cutting needles n.

[0200] The cutting needle selection unit is used to select the n cutting needles closest to the tumor area to be cut as the cutting needles to be operated, and to assign a corresponding cutting area to each cutting needle to be operated;

[0201] The path planning unit is used to calculate the movement path of each cutting needle by taking the needle tip position of each cutting needle to be moved on the initial image I0 as its initial position and the position of the cutting area of ​​the tumor region to be cut on the initial image I0 as its target position, through the path planning algorithm.

[0202] Furthermore, it should be noted that the microscopic automatic cutting device for tumor cell sampling provided in the above embodiments is only illustrated by the division of the above functional modules when cutting tumor cells. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the microscopic automatic cutting device for tumor cell sampling can be divided into different functional modules to complete all or part of the functions described above.

[0203] In some other embodiments, the present invention discloses a fully automatic tissue cutting device, as shown in Figures 10 to 12. The fully automatic tissue cutting device includes a housing 1 and a cutting assembly 8. A door 2 is hinged to one side of the housing 1, and an observation window 3 is provided on the upper part of the door 2. The internal state of the housing 1 can be easily observed through the observation window 3 made of transparent acrylic material. A base plate 4 is fixed to the lower end of the interior of the housing 1, and a bracket 5 is connected to the top of the base plate 4. A high-definition camera 6 is fixed to the top of the bracket 5, and a transport assembly 7 is provided below the high-definition camera 6. The position of the slide on the transport assembly 7 is observed through the high-definition camera 6. The transport assembly 7 includes a transport motor 701, a synchronous pulley group 702, a transport platform 703, a guide rail 704, and an LED light 705. A transport motor 701 is mounted on the top, and the output end of the transport motor 701 is connected to a synchronous pulley set 702. A transport platform 703 is set on the top surface of the middle of the synchronous pulley set 702, and guide rails 704 are slidably connected to the bottom ends of the transport platform 703. The transport motor 701 drives the transport platform 703 to slide along the guide rails 704 through the synchronous pulley set 702. Four LED lights 705 are installed inside the transport platform 703, and the LED lights 705 are evenly distributed along the top of the transport platform 703. The LED lights 705 can provide supplementary lighting under the glass slide, which helps to ensure the effect of shooting by the high-definition camera 6. The cutting component 8 is set on the top of the base plate 4, and the cutting component 8 includes an X-axis electric slide 801 and a Y-axis... The system includes an electric slide 802, a Z-axis electric slide 803, a mounting base 804, an optical coupler lifting precision fine-tuning track 805, a drive base 806, a precision linear guide 807, a high-precision optical axis 808, a support spring 809, a connecting cover 810, and a scraper holder 811. The top of the base plate 4 is equipped with an X-axis electric slide 801, whose output end is connected to a Y-axis electric slide 802. The output end of the Y-axis electric slide 802 is fixed to a Z-axis electric slide 803, whose output end is connected to a mounting base 804. The scraper can move along three axes via the X-axis electric slide 801, Y-axis electric slide 802, and Z-axis electric slide 803, thereby automatically controlling the scraper to perform relevant cutting operations on the tissue on the glass slide. The mounting base 804 has an optical coupler lifting precision fine-tuning track 805 on one side, and the output end of the optical coupler lifting precision fine-tuning track 805 is connected to a drive base 806. The position of the scraper needle can be finely adjusted through the optical coupler lifting precision fine-tuning track 805, thereby improving the cutting accuracy. It should be noted that the optical coupler lifting precision fine-tuning track 805 is an existing precision control device, and this application is not an improvement on this device. Therefore, this application will not elaborate on its specific driving principle. A precision linear guide 807 is slidably connected to one side of the drive base 806, and a high-precision optical shaft 808 is slidably connected to the middle of the drive base 806. Support springs 809 are sleeved on the outer sides of the upper and lower ends of the high-precision optical shaft 808, and the high-precision optical shaft 808 is fixedly connected to the precision linear guide 807.The weight of the clamp and the scraper is counteracted by the different forces of two springs at the top and bottom, creating a near-zero-gravity environment for easy operation. A connecting cover 810 is mounted on the other side of the drive base 806, and a scraper seat 811 is fitted inside the connecting cover 810.

[0204] As shown in Figures 11 to 14, a connecting component 9 is provided on the outer side of the scraper needle holder 811. The connecting component 9 includes an anti-slip groove 901, a positioning hole 902, a ball screw 903, and an anti-fooling slot 904. Anti-slip grooves 901 are provided on both sides of the middle part of the scraper needle holder 811, and positioning holes 902 are provided on both sides of the upper part of the scraper needle holder 811. The middle sides of the scraper needle holder 811 are arc-shaped, and the anti-slip grooves 901 are evenly distributed along the arc surface of the scraper needle holder 811. The anti-slip grooves 901 can improve the stability of the scraper needle holder 811. The friction between the hand and the scraper holder 811, and the ball screw 903 abutting inside the positioning hole 902, with the ball screw 903 threadedly connected to the connecting cover 810, allow the ball screw 903 to use its internal spring to drive the ball to engage in the positioning hole 902. This enables rapid positioning between the scraper holder 811 and the connecting cover 810. Furthermore, the multiple ball screws 903 eliminate the gap between the scraper and the fixture, thereby eliminating the impact of machining errors on the stability of the scraper and improving overall stability. The lower part of the connecting cover 810 is provided with a foolproof opening 904. The foolproof opening 904 makes the placement and orientation of the scraper fixed. After installation in this direction, it will not obstruct the field of view of the microscope above, and the contact point of the scraper is easily identifiable in the field of view. A middle partition 10 is fixed inside the housing 1. A first power supply 11 is installed at one end of the lower part of the middle partition 10, and a second power supply 12 is installed at the other end of the lower part of the middle partition 10. The first power supply 11 and the second power supply 12 supply power to the internal components of the equipment. A control board 13 is fixed at one end of the upper part of the middle partition 10, and a driver 14 is installed at the other end of the upper part of the middle partition 10. The driver 14 can drive the transport component 7 and the cutting component 8 to work. A fan 15 is installed on the upper part of the other side of the housing 1, and an interface 16 is provided at one end of the other side of the housing 1. The fan 15 can improve the heat dissipation effect of the electrical components between the middle partition 10 and the housing 1. A switch 17 is provided on the lower part of the other side of the housing 1. The equipment is started by using the switch 17.

[0205] In summary, when using this fully automatic tissue cutting equipment, firstly, it is connected to the corresponding equipment via interface 16, and then the switch 17 is turned on to supply power to the internal components of the equipment via the first power supply 11 and the second power supply 12. During the operation of the equipment, the fan 15 can improve the heat dissipation effect of the electrical components between the partition 10 and the housing 1. Secondly, the housing door 2 is opened, and the drive motor 701 is controlled by the driver 14 to work. The synchronous pulley group 702 will then drive the transport platform 703 to slide along the guide rail 704, moving the transport platform 703 to the opening, making it convenient for the operator to place the glass slides on the transport platform 703. Next, close the chamber door 2. The internal state of the chamber 1 can be easily observed through the transparent acrylic observation window 3. At this time, the high-definition camera 6 on the bracket 5 will observe the position of the slide on the transport component 7. An LED light 705 is set under the slide to provide supplemental lighting under the tissue, which helps to ensure the effect of the high-definition camera 6 when taking pictures. Then, the X-axis electric slide 801, Y-axis electric slide 802 and Z-axis electric slide 803 are driven by the driver 14 to move the scraper on the scraper holder 811 along the three axes, thereby automatically controlling the scraper to perform relevant cutting operations on the tissue on the slide. This system improves the automation level of the equipment, and the position of the scraper needle can be finely adjusted through the optical coupler lifting precision micro-adjustment track 805, thereby improving cutting accuracy. During this process, the support springs 809 sleeved on the outer sides of the upper and lower ends of the high-precision optical axis 808 can offset the weight of the clamp and scraper needle, creating a near-zero gravity environment for easy driving. At the same time, the transport platform 703 is equipped with multiple workstations, which can continuously process multiple samples without frequent shutdowns, thus improving overall processing efficiency. Finally, when the scraper needle holder 811 needs to be replaced after cutting, the anti-slip groove 901 can improve the hand's grip on the scraper needle holder 811. Friction is applied, and then pulling down the scraper holder 811 will move the ball of the ball screw 903 out of the positioning hole 902. During installation, the scraper holder 811 is directly inserted into the connecting cover 810, and the ball screw 903 can use its internal spring to drive the ball to embed into the positioning hole 902, which can quickly position the scraper holder 811 and the connecting cover 810. At the same time, the anti-fooling slot 904 fixes the placement and orientation of the scraper. After installation in this direction, it will not obstruct the upper microscope field of view. Therefore, no additional tools are needed when disassembling, which helps to improve the convenience of disassembling and replacing the scraper holder 811.

[0206] Furthermore, the fully automated tissue cutting equipment and the microscopic automated cutting method for tumor cell sampling provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0207] This invention discloses a microscopic automated cutting method and a fully automated tissue cutting device for tumor cell sampling, which have the following beneficial effects:

[0208] First, the present invention can accurately identify the location of the tumor area, achieving high-precision positioning and fully automatic cutting.

[0209] Secondly, this invention uses an image registration algorithm to associate WSI images with images acquired by a visual monitoring device, mapping the tumor region on the WSI image onto the initial image, thus ensuring high precision and full automation of the cutting process.

[0210] Third, the present invention uses a visual monitoring device to monitor the cutting process, based on the deformation of the normal tissue area, to ensure the stability of the cutting process.

[0211] Fourth, the equipment used in this invention is inexpensive, easy to operate, has a high cutting throughput, avoids laser thermal damage, and provides precise cutting.

[0212] Fifth, this invention, through the setting of a high-definition camera, can observe the position of the glass slide on the transport component. At this time, the scraper on the scraper holder can move along the three axes through the X-axis electric slide, Y-axis electric slide and Z-axis electric slide, thereby automatically controlling the scraper to perform related cutting operations on the tissue on the glass slide, improving the automation level of the equipment. Furthermore, the scraper can be finely adjusted through the optical coupler lifting precision fine-tuning track, thereby improving the cutting accuracy. At the same time, the transport platform is set with multiple stations, which can continuously process multiple samples, thus eliminating the need for frequent shutdowns and improving the overall processing efficiency. In addition, an LED light is set under the glass slide to provide supplemental lighting under the tissue, which is beneficial to ensure the effect of high-definition camera shooting.

[0213] Sixth, through the design of the connecting components, the ball screw can use its internal spring to drive the ball to embed into the positioning hole, which can quickly position the scraper holder and the connecting cover. At the same time, the anti-fooling notch ensures that the scraper is fixed in orientation during installation and will not obstruct the microscope's field of view after installation. Furthermore, when it is necessary to replace consumables such as scraper blades, the anti-slip groove can increase the friction between the hand and the scraper holder. Then, by pulling down the scraper holder, the ball of the ball screw can be moved out of the positioning hole. Therefore, no additional tools are needed during disassembly, which improves the convenience of disassembling and replacing the scraper holder.

[0214] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the present invention. Various changes and modifications can be made to the present invention without departing from its spirit and scope. All such changes and modifications fall within the scope of the present invention as claimed, which is defined by the appended claims and their equivalents.

Claims

1. A microscopic automated cutting method for tumor cell sampling, characterized in that, include: Step S1: Scan the slide using a digital microscope to obtain a WSI image; Step S2: Using a pre-trained tumor recognition model, identify all tumor regions and their locations on the WSI image; Step S3: Place one or more glass slides on and fix them on the platform, and use one or more visual monitoring devices to acquire an initial image I0, which shows the slices and cutting needles. Step S4: Using an image registration algorithm, determine the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image, and obtain the tumor region and its location on the initial image I0; Step S5: Identify the position of the cutting needle tip on the initial image I0 using a needle tip recognition algorithm; Step S6: Using the position of the cutting needle tip on the initial image I0 as the initial position and the position of the tumor area to be cut on the initial image I0 as the target position, the cutting needle movement path is calculated by the path planning algorithm. Step S7: The cutting needle moves to the target position according to the obtained cutting needle movement path and falls to cut on the corresponding glass slide; During the cutting process, a visual monitoring device is used to acquire real-time images. t and will real-time image I t By comparing with the initial image I0, the deformation of the normal tissue region is obtained. When the deformation exceeds the threshold, the cutting stops; otherwise, cutting continues until the current tumor area is completely cut. Step S8: After the current tumor area is cut, move the slide and repeat steps S3-S7 until all tumor areas on the slide are cut to achieve tumor cell sampling.

2. The automatic microscopic cutting method according to claim 1, characterized in that, Step S4 includes: Step S4.1: Using the initial image I0 as the reference image and the WSI image as the floating image, the initial image I0 and the WSI image are registered using an image registration algorithm; Step S4.2: After image registration, determine the position of the initial image I0 on the WSI image, and obtain the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image; Step S4.3: Based on the relative relationship between the two coordinate systems, and based on all tumor regions and their locations on the WSI image, obtain the tumor regions and their locations on the initial image I0.

3. The automatic microscopic cutting method according to claim 1, characterized in that, Step S5 includes: Step S5.1: Use the image of the cutting needle tip as the template image, slide it pixel by pixel on the initial image I0, and use a similarity measurement method to calculate and record the matching degree between the sliding area of ​​the template image and the initial image I0 at each position; Step S5.2: Find the position with the highest matching degree, which is the position of the cutting needle in the initial image I0.

4. The automatic microscopic cutting method according to claim 1, characterized in that, Step S7 includes: Step S7.1: During the cutting process, acquire real-time images using a visual monitoring device. t The real-time normal tissue region N is obtained according to the following formula. t : N t =I t -T0; Where: T0 is the tumor region on the initial image I0; Step S7.2: Obtain the initial normal tissue region N0 according to the following formula: N0 = I0 - T0; Step S7.3: Calculate the real-time normal tissue region N t The average of the sum of squares of the differences between the pixel values ​​corresponding to the initial normal tissue region N0 and the normal tissue region is used to obtain the deformation of the normal tissue region. This deformation is then compared with a threshold. Cutting is stopped when the deformation exceeds the threshold. Step S7.4: Repeat steps S7.1-S7.3 until the current tumor area is completely cut.

5. The automatic microscopic cutting method according to claim 1, characterized in that, Step S6 includes: Step S6.1: Extract features from the portion of the tumor area to be cut to obtain tumor features; Step S6.2: Use the trained tumor complexity judgment model to judge the complexity of the tumor region to be cut, and output the complexity level T and the number of cutting needles n; Step S6.3: Select the n cutting needles closest to the tumor area to be cut as the cutting needles to be operated, and assign a corresponding cutting area to each cutting needle to be operated; Step S6.4: Using the needle tip position of each cutting needle to be moved on the initial image I0 as its initial position, and the position of the cutting area of ​​the tumor region to be cut on the initial image I0 as its target position, the cutting needle movement path of each cutting needle to be moved is calculated by the path planning algorithm.

6. A microscopic automated cutting device for tumor cell sampling, characterized in that, include: The WSI image acquisition module is used to scan glass slides using a digital microscope to acquire WSI images; The tumor region identification module is used to identify all tumor regions and their locations on WSI images using a pre-trained tumor identification model. The initial image acquisition module is used to place one or more glass slides on and fix them on the platform, and to acquire an initial image I0 using one or more visual monitoring devices. The initial image I0 shows the slice and cutting needle. The image registration module is used to determine the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image through an image registration algorithm, and to obtain the tumor region and its location on the initial image I0. The cutting needle recognition module is used to identify the position of the cutting needle tip on the initial image I0 using a needle tip recognition algorithm; The path planning module is used to calculate the movement path of the cutting needle by taking the position of the cutting needle tip on the initial image I0 as the initial position and the position of the tumor area to be cut on the initial image I0 as the target position. The cutting module is used to move the cutting needle to the target position according to the obtained cutting needle movement path and then fall to cut on the corresponding glass slide; The cutting monitoring module is used to acquire real-time images using a visual monitoring device during the cutting process. t and will real-time image I t By comparing with the initial image I0, the deformation of the normal tissue region is obtained. When the deformation exceeds the threshold, the cutting stops; otherwise, cutting continues until the current tumor area is completely cut. The repeat execution module is used to move the slide after the current tumor area has been cut, and repeat the methods in the initial image acquisition module, image registration module, cutting needle identification module, path planning module, cutting module, and cutting monitoring module until all tumor areas on the slide have been cut, thus achieving tumor cell sampling.

7. The automatic microscopic cutting device according to claim 6, characterized in that, The image registration module includes: The image registration unit is used to register the initial image I0 and the WSI image as a floating image using an image registration algorithm. The coordinate system corresponding unit is used to determine the position of the initial image I0 on the WSI image after image registration, and to obtain the relative relationship between the coordinate system of the initial image I0 and the coordinate system of the WSI image. The initial image tumor region determination unit is used to obtain the tumor region and its location on the initial image I0 based on the relative relationship between the two coordinate systems and all tumor regions and their locations on the WSI image.

8. The automatic microscopic cutting device according to claim 6, characterized in that, The cutting needle identification module includes: The template matching unit is used to slide the image of the cutting needle tip as a template image on the initial image I0 pixel by pixel, and to calculate and record the matching degree between the template image and the sliding area of ​​the initial image I0 at each position using a similarity measurement method. The cutting needle position determination unit is used to find the position with the highest matching degree, which is the position of the cutting needle in the initial image I0.

9. The automatic microscopic cutting device according to claim 6, characterized in that, The cutting monitoring module includes: The real-time normal tissue acquisition unit is used to acquire real-time images using a visual monitoring device during the cutting process. t The real-time normal tissue region N is obtained according to the following formula. t : N t =I t -T0; Where: T0 is the tumor region on the initial image I0; The initial normal tissue acquisition unit is used to obtain the initial normal tissue region N0 according to the following formula: N0 = I0 - T0; The deformation calculation unit is used to calculate the real-time normal tissue region N. t The average of the sum of squares of the differences between the pixel values ​​corresponding to the initial normal tissue region N0 and the normal tissue region is used to obtain the deformation of the normal tissue region. This deformation is then compared with a threshold. Cutting is stopped when the deformation exceeds the threshold. The repeat execution unit is used to repeat the methods in the real-time normal tissue acquisition unit, the initial normal tissue acquisition unit, and the deformation calculation unit until the current tumor region is completely cut.

10. The automatic microscopic cutting device according to claim 6, characterized in that, The path planning module includes: The feature extraction unit is used to extract features from the portion of the tumor region to be cut, thereby obtaining tumor features. The complexity identification unit is used to judge the complexity of the tumor region to be cut using a trained tumor complexity judgment model, and outputs the complexity level T and the number of cutting needles n. The cutting needle selection unit is used to select the n cutting needles closest to the tumor area to be cut as the cutting needles to be operated, and to assign a corresponding cutting area to each cutting needle to be operated; The path planning unit is used to calculate the movement path of each cutting needle by taking the needle tip position of each cutting needle to be moved on the initial image I0 as its initial position and the position of the cutting area of ​​the tumor region to be cut on the initial image I0 as its target position, through the path planning algorithm.

11. A fully automatic tissue cutting device, comprising a housing (1) and a cutting assembly (8), characterized in that, A door (2) is hinged to one side of the housing (1), and an observation window (3) is provided on the upper part of the door (2). A base plate (4) is fixed to the lower part of the interior of the housing (1), and a bracket (5) is connected to the top of the base plate (4). A high-definition camera (6) is fixed to the top of the bracket (5), and a transport component (7) is provided below the high-definition camera (6). The cutting component (8) is located on the top of the base plate (4), and the cutting component (8) includes an X-axis electric slide (801), a Y-axis electric slide (802), a Z-axis electric slide (803), and a mounting base (804). 04) Optical coupler lifting precision fine-tuning track (805), drive seat (806), precision linear guide (807), high precision optical axis (808), support spring (809), connecting cover (810) and scraper seat (811). The top of the base plate (4) is equipped with an X-axis electric slide (801), and the output end of the X-axis electric slide (801) is connected to a Y-axis electric slide (802). The output end of the Y-axis electric slide (802) is fixed with a Z-axis electric slide (803), and the output end of the Z-axis electric slide (803) is connected to a mounting seat (804).

12. The fully automatic tissue cutting device according to claim 11, characterized in that, The transport component (7) includes a transport motor (701), a synchronous pulley set (702), a transport platform (703), a guide rail (704), and an LED light (705). The transport motor (701) is mounted on the top of the base plate (4), and the output end of the transport motor (701) is connected to the synchronous pulley set (702). The transport platform (703) is provided on the top surface of the middle part of the synchronous pulley set (702), and the guide rail (704) is slidably connected to both ends of the bottom of the transport platform (703). An LED light (705) is installed inside the transport platform (703).

13. The fully automatic tissue cutting device according to claim 12, characterized in that, The number of LED lights (705) is four, and the LED lights (705) are evenly distributed along the top of the transport platform (703).

14. The fully automatic tissue cutting device according to claim 11, characterized in that, A connecting component (9) is provided on the outer side of the scraper needle seat (811), and the connecting component (9) includes an anti-slip groove (901), a positioning hole (902), a ball screw (903), and a foolproof opening (904). The scraper needle seat (811) has anti-slip grooves (901) on both sides of the middle part, and a positioning hole (902) is provided on the upper outer side of the scraper needle seat (811). The ball screw (903) is abutted inside the positioning hole (902), and the ball screw (903) is threadedly connected to the connecting cover (810). The lower part of the connecting cover (810) has a foolproof opening (904).

15. The fully automatic tissue cutting device according to claim 14, characterized in that, The middle two sides of the scraper seat (811) are arc-shaped, and the anti-slip grooves (901) are equidistantly distributed along the arc surface of the scraper seat (811).

16. The fully automatic tissue cutting device according to claim 11, characterized in that, The box (1) has a central partition (10) fixed inside, and a first power supply (11) is installed at one end of the lower part of the central partition (10), and a second power supply (12) is installed at the other end of the lower part of the central partition (10).

17. The fully automatic tissue cutting device according to claim 16, characterized in that, A control board (13) is fixed to one upper end of the partition plate (10), and a driver (14) is installed at the other upper end of the partition plate (10).

18. The fully automatic tissue cutting device according to claim 11, characterized in that, A fan (15) is installed on the upper part of the other side of the housing (1), and an interface (16) is provided at one end of the other side of the housing (1). A switch (17) is provided on the lower part of the other side of the housing (1).

19. The fully automatic tissue cutting device according to claim 11, characterized in that, The mounting base (804) is provided with an optical coupler lifting precision fine-tuning track (805) on one side, and the output end of the optical coupler lifting precision fine-tuning track (805) is connected to a drive base (806). A precision linear guide (807) is slidably connected to one side of the drive base (806), and a high-precision optical axis (808) is slidably connected to the middle of the drive base (806).

20. The fully automatic tissue cutting device according to claim 11, characterized in that, The high-precision optical axis (808) is fitted with support springs (809) on the outer sides of its upper and lower ends, and the high-precision optical axis (808) is fixedly connected to the precision linear guide (807). A connecting cover (810) is arranged on the other side of the drive seat (806), and a scraper seat (811) is fitted inside the connecting cover (810).

Images