A white blood cell classification detection system and method

Abstract

The present invention discloses a leukocyte classification detection system and method, which has a simple chip structure design, low production cost, simple use method, and does not require too much reagent. According to the size and deformability of leukocytes, a suitable microchannel and a height range of a detection structure area are designed, and a front-stage microcolumn gap and a terminal microcolumn gap of a suitable size range are designed. The front-stage microcolumn array and the terminal microcolumn array and the area between them form a front-stage classification counting area, a middle-stage classification counting area, and a terminal classification counting area, which effectively realizes the detection and distinction of lymphocytes, granulocytes, and monocytes; the detection and classification of lymphocytes, granulocytes, and monocytes are realized through the target detection model in the leukocyte classification counting platform, and then the counting function module is used to realize the counting of monocytes in the front-stage classification counting area, the counting of lymphocytes in the middle-stage classification counting area, and the counting of granulocytes in the terminal classification counting area, and finally realizes the three-classification counting of leukocytes.

Description

Technical Field

[0001] The present invention belongs to the technical field of cell detection and classification, and in particular relates to a system and method for detecting white blood cell classification. Background Art

[0002] White blood cells play an important role in the human immune system, and are primarily responsible for preventing organisms from being infected and invaded. The classification and counting of white blood cells is often used to detect whether the concentration of white blood cells in a living body is abnormal, so as to further assist in determining the presence and extent of the disease. At present, manual microscopy, flow cytometers, and blood cell analyzers are usually used to classify and count white blood cells. Manual microscopy requires professionals to observe and classify blood smears by making blood smears. Since it is labor-intensive, it is often used as an auxiliary means; flow cytometers also require professional personnel to operate, and the operation steps are many and complex, the equipment is expensive, and many reagents and consumables are required, and a specific storage environment is required; although the price of a blood cell analyzer is not as expensive as a flow cytometer, the required supporting reagents are similar to those of a flow cytometer.

[0003] In recent years, many scholars have proposed new methods for white blood cell classification and counting to make up for the shortcomings of current classification and counting methods and improve the level of white blood cell detection. In order to reduce the cost of detection instruments, some scholars have simplified the optical structure of existing flow cytometers to a certain extent, or made their own optical path structure, and constructed a white blood cell classification detection system based on microfluidics and multiple imaging methods, but it still requires the use of multiple antibodies, magnetic particles, fluorescent dyes and other reagents and consumables; some scholars use digital holographic microscopes and perform fluid dynamic focusing on white blood cell solutions on microfluidic chips to reconstruct white blood cell images and perform label-free detection and classification of white blood cells, but the white blood cell images obtained by this method are poor, the image features are relatively simple, and the classification effect is poor.

[0004] The invention patent "A red blood cell quality detection system and method" (publication number: CN115468895 A) discloses a method for identifying and counting red blood cells, which simplifies the design and structure of the technical device, and can effectively avoid the influence of cell overlap and achieve accurate counting; the invention mainly divides red blood cells into six different morphological categories based on the characteristics of different morphological characteristics of red blood cells stored for different lengths of time, and identifies and counts red blood cells of these six different morphological categories. The object of the present invention is white blood cells of multiple categories, and there is no obvious difference in morphology between some categories. It is difficult to effectively detect, classify and count white blood cell categories based only on the morphology of white blood cells. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a leukocyte classification detection system and method that solves the problems of existing leukocyte classification and counting methods, such as labor-intensiveness, expensive equipment, large number of reagents and consumables required, many and complex operation steps, and poor classification effect.

[0006] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a leukocyte classification detection system, including a leukocyte classification microfluidic chip, a microscope, a high-speed camera mounted on the microscope, and a leukocyte classification counting platform;

[0007] The leukocyte classification microfluidic chip comprises a base and a cover plate; the cover plate is arranged above the base, a sample inlet and a sample outlet are arranged on the cover plate, a microchannel is arranged on the side of the cover plate facing the base; a detection structure area is arranged in the microchannel; the microchannel connects the sample inlet, the sample outlet and the detection structure area;

[0008] The detection structure area includes a front-end classification and counting area, a middle-end classification and counting area and a terminal classification and counting area; the front-end classification and counting area is composed of a front-end micro-column array and a front-end micro-column gap; the terminal classification and counting area is composed of a terminal micro-column array and a terminal micro-column gap.

[0009] The leukocyte classification microfluidic chip is mainly used to provide a structure for distinguishing different types of leukocytes and to provide a leukocyte image sampling microenvironment. The microscope and the high-speed camera mounted on the microscope are used to collect leukocyte images. The leukocyte classification and counting platform is used to deploy a target detection model for tracking and detecting different types of leukocytes and a counting function module for counting different types of leukocytes.

[0010] In this scheme, since there are many types of white blood cells, and there is no obvious difference in morphology between some types, it is difficult to effectively detect, classify and count the types of white blood cells based only on the morphology of white blood cells. Therefore, the present invention mainly designs suitable microchannels and detection structure areas based on the size and deformability of white blood cells, which can effectively produce different obstacles to white blood cells such as lymphocytes, granulocytes and monocytes, causing them to pass through the structure at different speeds, thereby achieving the distinction between different types of white blood cells. According to the different sizes and deformabilities of different types of white blood cells, a detection structure area with a specific size range, a front microcolumn array, a front microcolumn gap, a terminal microcolumn array and a terminal microcolumn gap are designed, so that the distinction between different types of white blood cells is achieved from the hardware point of view, creating conditions for the subsequent white blood cell classification and counting platform to classify and count different types of white blood cells.

[0011] Furthermore, the hole radius of the injection port and the output port is 400 μm-600 μm;

[0012] Furthermore, the width of the microchannel is 20 μm-250 μm.

[0013] The beneficial effects of the above further scheme are: the hole radius of the sample inlet and the sample outlet is relatively wide, which is convenient for the passage and outflow of the white blood cell mixture; the appropriate microchannel width range makes it convenient for the microscope to clearly observe the morphology of white blood cells under the appropriate magnification, and is just within the field of view of the microscope. At the same time, when the microchannel height is relatively low, the microchannel will not collapse.

[0014] Furthermore, the heights of the microchannel and the detection structure area are both 10μm-30μm; the heights of the front-stage microcolumn array and the rear-stage microcolumn array are both 10μm-30μm; the front-stage microcolumn gaps formed between the microcolumns in the front-stage microcolumn array are 9μm-12μm; the rear-stage microcolumn gaps formed between the microcolumns in the rear-stage microcolumn array are 6μm-9μm; the front-stage microcolumn array and the rear-stage microcolumn array are both perpendicular to the substrate.

[0015] The beneficial effects of the above further scheme are as follows: the appropriate height range of the microchannel and the detection structure area can ensure that white blood cells of different sizes can pass through the microchannel and the detection structure area smoothly, and effectively avoid the overlapping of white blood cells, thereby avoiding the negative impact on the detection effect of the target detection model; the height of the front microcolumn array and the rear microcolumn array is consistent with the height of the microchannel and the detection structure area, which can effectively avoid the white blood cells from being squeezed through the gaps formed by the microcolumns and the bottom of the microchannel when they are squeezed through the gaps between the microcolumns, thereby avoiding the negative impact on the detection effect of the target detection model; the appropriate width of the front microcolumn gap and the rear microcolumn gap can effectively ensure that white blood cells of different sizes and different deformation degrees pass through the microcolumn gaps of different widths at different speeds, providing conditions for the detection, classification and counting of a certain type of white blood cells of different categories in different classification and counting areas, and effectively ensuring that when the gap is severely blocked, the blockage can be cleared by increasing the appropriate fluid pressure, ensuring that the white blood cell detection can continue smoothly.

[0016] Furthermore, the material of the substrate is glass, and the cover plate is a polydimethylsiloxane detection structure module.

[0017] A method for detecting leukocyte classification comprises the following steps:

[0018] S1. Mixing whole blood with a red blood cell solvent to remove red blood cells from the whole blood, and then filtering the mixture without red blood cells through a cell filter screen to remove larger debris and particles in the mixture to obtain a white blood cell mixture;

[0019] S2, placing the leukocyte classification microfluidic chip on the stage of a microscope, and placing the detection structure area just below the objective lens;

[0020] S3, introducing the mixed leukocyte solution obtained in S1 into the microchannel from the injection port through a micropump;

[0021] S4. The high-speed camera on the microscope is used to collect images passing through the detection structure area, and the collected images are input into the white blood cell classification and counting platform, and one of the different types of white blood cells is detected, classified and counted in the front-end classification and counting area, the middle classification and counting area and the end classification and counting area respectively.

[0022] Furthermore, the S1 is specifically:

[0023] The whole blood and the red blood cell solvent are mixed at a volume ratio of 1:3-20 to dissolve the red blood cells in the whole blood; the pore size of the cell filter screen is 15μm-500μm, which will filter out larger debris and particles in the mixed solution.

[0024] Furthermore, in S2, the magnification of the microscope is 100-1000 times.

[0025] The beneficial effects of the above further scheme are: setting a suitable range of the volume ratio of whole blood and red blood cell solvent is to fully dissolve the red blood cells in the whole blood and avoid the interference of red blood cells on white blood cells during detection; the appropriate pore size range of the cell filter screen can effectively avoid the interference of larger debris and particles in the mixed liquid on white blood cells during detection; the microscope is set to a suitable magnification range in order to clearly observe the morphology of white blood cells.

[0026] Furthermore, in S4, the acquired image will be input into the target detection model in the white blood cell classification and counting platform, and the target detection model will track, detect and classify one of the different types of white blood cells in the front classification and counting area, the middle classification and counting area and the final classification and counting area respectively.

[0027] The beneficial effect of the above further scheme is: tracking detection and classification means that while detecting and classifying white blood cells, the white blood cells in motion are also tracked to ensure that the same white blood cell will not be judged as another cell after the position changes due to movement in a certain classification and counting area, thereby avoiding the same cell being counted repeatedly, thereby causing counting errors.

[0028] Furthermore, in S4, the counting function module in the white blood cell classification and counting platform will count one of the different types of white blood cells in the front-end classification and counting area, the middle-end classification and counting area, and the end-end classification and counting area, respectively, to complete the counting task of three types of white blood cells. The core algorithm is as follows: if WBC_x>Mreg_Lx and WBC_x<Mreg_Rxand WBC_y>Mreg_Txand WBC_y<Mreg_Dx and WBC_type==type0

[0029] WBC0←WBC0+1

[0030] if WBC_x>Freg_Lx and WBC_x<Freg_Rx and WBC_y>Freg_Txand WBC_y<Freg_Dx and WBC_type==type1

[0031] WBC1←WBC1+1

[0032] if WBC_x>Lreg_Lx and WBC_x<Lreg_Rx and WBC_y>Lreg_Tx

[0033] and WBC_y<Lreg_Dx and WBC_type==type2

[0034] WBC2←WBC2+1

[0035] Among them, WBC_x and WBC_y are the x-coordinate and y-coordinate of the center of the white blood cell respectively; Mreg_Lx, Mreg_Rx, Mreg_Tx, and Mreg_Dx are the left and right upper and lower boundary line coordinates of the middle classification counting area respectively; Freg_Lx, Freg_Rx, Freg_Tx, and Freg_Dx are the left and right upper and lower boundary line coordinates of the front classification counting area respectively; Lreg_Lx, Lreg_Rx, Lreg_Tx, and Lreg_Dx are the left and right upper and lower boundary line coordinates of the last classification counting area respectively; WBC_type is the category of the white blood cells detected, type0, type1, and type2 are the preset three categories of white blood cell category values ​​respectively; WBC0 is the number of the first category of white blood cells counted in the middle classification counting area, WBC1 is the number of the second category of white blood cells counted in the front classification counting area, and WBC2 is the number of the third category of white blood cells counted in the last classification counting area.

[0036] The beneficial effects of the present invention are:

[0037] (1) The leukocyte classification detection system provided by the present invention has a simple structural design and low manufacturing cost. According to the different sizes and deformabilities of different types of leukocytes, a detection structure area with a specific size range, a front micro-column array, a front micro-column gap, a terminal micro-column array and a terminal micro-column gap are designed; by setting a suitable wide height range for the microchannel and the detection structure area, the target detection model can achieve a better classification and counting effect of leukocytes; by setting a suitable microscope magnification range, it is beneficial to obtain a suitable observation field of view and further obtain the best sampling image.

[0038] (2) The target detection model and counting function module provided by the present invention are deployed on a white blood cell classification and counting platform. The target detection model is used to track, detect and classify white blood cells in the front, middle and end regions respectively, and the counting function module is used to count a certain type of white blood cells of different categories in the front, middle and end regions respectively. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A diagram of the system for use of the present invention;

[0040] Figure 2 It is a schematic diagram of the structure of the microfluidic chip of the present invention;

[0041] Figure 3 A top view of the microfluidic chip structure of the present invention;

[0042] Figure 4 It is a schematic diagram of the detection structure area in the fluidic chip of the present invention;

[0043] Figure 5 It is a flow chart of the use of the present invention;

[0044] Figure 6 It is the final classification counting effect diagram of the present invention. DETAILED DESCRIPTION

[0045] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0046] Embodiment 1:

[0047] like Figure 1 As shown, in one embodiment of the present invention, a leukocyte classification detection system includes a leukocyte classification microfluidic chip 1, a microscope 2, a high-speed camera 3 mounted on the microscope 2, and a leukocyte classification counting platform 4;

[0048] Among them, Figure 2 As shown, the leukocyte classification microfluidic chip 1 comprises a substrate 11 and a cover plate 12; the cover plate 12 is arranged above the substrate 11, and is provided with an inlet 13 and an outlet 14; a microchannel 15 is arranged on the side of the cover plate 12 facing the substrate 11; a detection structure area 16 is arranged in the microchannel 15; the microchannel 15 connects the inlet 13, the outlet 14 and the detection structure area 16;

[0049] like Figure 3 and Figure 4 As shown, the detection structure area 16 includes a front-end classification and counting area 161, a middle-end classification and counting area 162 and a rear-end classification and counting area 163; the front-end classification and counting area 161 is composed of a front-end micro-column array 1611 and a front-end micro-column gap 1612; the rear-end classification and counting area 163 is composed of a rear-end micro-column array 1631 and a rear-end micro-column gap 1632.

[0050] The leukocyte classification microfluidic chip 1 is mainly used to provide a structure for distinguishing different types of leukocytes and to provide a leukocyte image sampling microenvironment. The microscope 2 and the high-speed camera 3 mounted on the microscope are used to collect leukocyte images. The leukocyte classification and counting platform is used to deploy a target detection model for tracking and detecting different types of leukocytes and a counting function module for counting different types of leukocytes.

[0051] The hole radius of the sample inlet 13 and the sample outlet 14 is 400 μm-600 μm;

[0052] The width of the microchannel 15 is 20 μm-250 μm.

[0053] The heights of the microchannel 15 and the detection structure area 16 are both 10μm-30μm; the heights of the front-section microcolumn array 1611 and the rear-section microcolumn array 1631 are both 10μm-30μm; the front-section microcolumn gaps 1612 formed between the microcolumns in the front-section microcolumn array 1611 are 9μm-12μm; the rear-section microcolumn gaps 1632 formed between the microcolumns in the rear-section microcolumn array 1631 are 6μm-9μm; the front-section microcolumn array 1611 and the rear-section microcolumn array 1631 are both perpendicular to the substrate 11.

[0054] The front-stage microcolumn gaps 1612 formed between each microcolumn in the front-stage microcolumn array 1611 are provided with a specific size range, which is set according to the size and deformability of monocytes, and is used to squeeze monocytes to make them pass slowly, so that monocytes can be distinguished from granulocytes and lymphocytes; the terminal microcolumn gaps 1632 formed between each microcolumn in the terminal microcolumn array 1631 are provided with a specific size range, which is set according to the size and deformability of granulocytes, and is used to squeeze granulocytes to make them pass slowly, so that granulocytes can be distinguished from monocytes and lymphocytes.

[0055] The material of the substrate 11 is glass, and the cover plate 12 is a polydimethylsiloxane detection structure module. The polydimethylsiloxane detection structure module is prepared by the following method:

[0056] (1) Weighing 20 g of polydimethylsiloxane liquid prepolymer with a weighing instrument, and mixing the prepolymer with a curing agent at a mass ratio of 16:1 to obtain a mixture;

[0057] (2) Stir the obtained mixture thoroughly with a glass rod until a large amount of bubbles are generated, and place the mixture in a desiccator and evacuate for 20 minutes;

[0058] (3) The vacuumed mixture was slowly poured onto a silicon wafer mold and heated at 75° C. for 1 hour to complete the preparation of the polydimethylsiloxane detection structure module.

[0059] Embodiment 2:

[0060] This embodiment is directed to a method for using a white blood cell classification detection system.

[0061] like Figure 5 As shown, in this embodiment, the method for using the white blood cell classification detection system includes the following steps:

[0062] S1. Mixing whole blood with a red blood cell solvent to remove red blood cells from the whole blood, and then filtering the mixture without red blood cells through a cell filter screen to remove larger debris and particles in the mixture to obtain a white blood cell mixture;

[0063] S2, placing the leukocyte classification microfluidic chip 1 on the stage of the microscope 2, and placing the detection structure area 16 just below the objective lens;

[0064] S3, the leukocyte mixture obtained in S1 is introduced into the microchannel 15 from the injection port 13 through a micro pump, and flows through the detection structure area. The mixture first flows from the front classification and counting area to the middle classification and counting area, and then flows to the end classification and counting area;

[0065] S4. The high-speed camera 3 on the microscope 2 collects images passing through the detection structure area 16, and the collected images are input into the white blood cell classification and counting platform 4, and the monocytes, lymphocytes and granulocytes are detected, classified and counted in the front classification and counting area 161, the middle classification and counting area 162 and the rear classification and counting area 163 respectively.

[0066] The S1 is specifically:

[0067] The whole blood and the red blood cell solvent are mixed at a volume ratio of 1:10 to dissolve the red blood cells in the whole blood; the pore size of the cell filter screen is 50 μm, which will filter out larger debris and particles in the mixed solution.

[0068] In S2, the magnification of the microscope 2 is 400 times.

[0069] In S4, the acquired image will be input into the target detection model in the white blood cell classification and counting platform 4, and the target detection model will track, detect and classify monocytes, lymphocytes and granulocytes in the front classification and counting area 161, the middle classification and counting area 162 and the end classification and counting area 163 respectively. The target detection model adopts yolov5, and the tracking function is implemented by deepsort.

[0070] In S4, the counting function module in the white blood cell classification counting platform 4 will count the monocytes, lymphocytes and granulocytes in the front classification counting area 161, the middle classification counting area 162 and the rear classification counting area 163 respectively, and complete the counting task of the three types of white blood cells. The core algorithm is as follows:

[0071] if WBC_x>Mreg_Lx and WBC_x<Mreg_Rx and WBC_y>Mreg_Txand WBC_y<Mreg_Dx and WBC_type==type0

[0072] WBC0←WBC0+1

[0073] if WBC_x>Freg_Lx and WBC_x<Freg_Rx and WBC_y>Freg_Txand WBC_y<Freg_Dx and WBC_type==type1

[0074] WBC1←WBC1+1

[0075] if WBC_x>Lreg_Lx and WBC_x<Lreg_Rx and WBC_y>Lreg_Tx

[0076] and WBC_y<Lreg_Dx and WBC_type==type2

[0077] WBC2←WBC2+1

[0078] Among them, WBC_x and WBC_y are the x-coordinate and y-coordinate of the center of the white blood cell respectively; Mreg_Lx, Mreg_Rx, Mreg_Tx, and Mreg_Dx are the left and right upper and lower boundary line coordinates of the middle classification and counting area 162 respectively; Freg_Lx, Freg_Rx, Freg_Tx, and Freg_Dx are the left and right upper and lower boundary line coordinates of the front classification and counting area 161 respectively; Lreg_Lx, Lreg_Rx, Lreg_Tx, and Lreg_Dx are the left and right upper and lower boundary line coordinates of the last classification and counting area 163 respectively; WBC_type is the category of the detected white blood cells, type0, type1, and type2 are the preset white blood cell category values, and type0=0, type1=1, and type2=2 are set; WBC0 is the number of lymphocytes counted in the middle classification and counting area 162, WBC1 is the number of monocytes counted in the front classification and counting area 161, and WBC2 is the number of granulocytes counted in the last classification and counting area 163.

[0079] Embodiment 3:

[0080] This embodiment is an experiment on the use of the white blood cell classification detection system.

[0081] The leukocyte classification microfluidic chip 1, microscope 2, high-speed camera 3, and leukocyte classification and counting platform 4 equipped with a target detection model and counting function module prepared in the above embodiment are used to complete the classification and counting experiment of lymphocytes, granulocytes, and monocytes. To verify the effectiveness of the present invention, the following steps are performed:

[0082] (1) Preparation of mixed leukocyte solution

[0083] Whole blood was mixed with a hemolytic agent for blood cell analysis at a volume ratio of 1:10 to dissolve the red blood cells in the whole blood, and then passed through a cell filter sieve with a pore size of 50 μm to filter out larger debris and particles to obtain a white blood cell mixture.

[0084] (2) Leukocyte Image Acquisition

[0085] The leukocyte classification microfluidic chip 1 prepared in the above-mentioned embodiment is placed on the stage of a microscope 2, the detection structure area 16 is placed directly below the objective lens, and the magnification of the microscope 2 is set to 400 times; the leukocyte mixture is introduced into the microchannel 15 from the injection port 13 through a micropump, and the monocytes will slowly pass through the front classification and counting area 161 due to being squeezed by the front microcolumn gap 1612, and the lymphocytes will smoothly pass through the middle classification and counting area 162 because they are relatively small and will not be squeezed by the front microcolumn gap and the end microcolumn gap, and the granulocytes will slowly pass through the end classification and counting area 163 due to being squeezed by the end microcolumn gap 1632; the image of leukocytes passing through the detection structure area 16 is collected by a high-speed camera 3.

[0086] (3) Input the collected leukocyte image into the leukocyte classification and counting platform 4, and detect the classification and counting effect. Figure 6 The target detection model tracks, detects and classifies monocytes (WBC-1) in the front classification and counting area 161, tracks, detects and classifies lymphocytes (WBC-0) in the middle classification and counting area 162, and tracks, detects and classifies granulocytes (WBC-2) in the rear classification and counting area 163;

[0087] The counting function module only counts the number of monocytes (WBC-1) in the front classification counting area 161, only counts the number of lymphocytes (WBC-0) in the middle classification counting area 162, and only counts the number of granulocytes (WBC-2) in the rear classification counting area 163.

[0088] The beneficial effects of the present invention are as follows: the present invention discloses a leukocyte classification detection system and a method of use, which has a simple chip structure design, low manufacturing cost, simple method of use, and does not require excessive reagents. According to the size and deformability of leukocytes, a suitable height range of the microchannel 15 and the detection structure area 16 is designed, and a front-stage microcolumn gap 1612 and a terminal microcolumn gap 1632 of a suitable size range are designed. The front-stage microcolumn array 1611 and the terminal microcolumn array 1631 and the area between them form a front-stage classification counting area 161, a middle-stage classification counting area 162 and a terminal classification counting area 163, which effectively realize the detection and distinction of lymphocytes, granulocytes and monocytes; the detection and classification of lymphocytes, granulocytes and monocytes are realized through the target detection model in the leukocyte classification counting platform 4, and then the counting function module realizes the counting of monocytes in the front-stage classification counting area 161, the counting of lymphocytes in the middle-stage classification counting area 162, and the counting of granulocytes in the terminal classification counting area 163.

[0089] 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 replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A leukocyte classification detection system, characterized in that: It comprises a leukocyte classification microfluidic chip (1), a microscope (2), a high-speed camera (3) mounted on the microscope (2), and a leukocyte classification and counting platform (4); The leukocyte classification microfluidic chip (1) comprises a base (11) and a cover plate (12); the cover plate (12) is arranged above the base (11); a sample inlet (13) and a sample outlet (14) are arranged on the cover plate (12); a microchannel (15) is arranged on a side of the cover plate (12) facing the base (11); a detection structure area (16) is arranged in the microchannel (15); the microchannel (15) connects the sample inlet (13), the sample outlet (14) and the detection structure area (16); The detection structure area (16) comprises a front-stage classification and counting area (161), a middle-stage classification and counting area (162) and a rear-stage classification and counting area (163); the front-stage classification and counting area (161) is composed of a front-stage micro-column array (1611) and a front-stage micro-column gap (1612); the rear-stage classification and counting area (163) is composed of a rear-stage micro-column array (1631) and a rear-stage micro-column gap (1632); The heights of the microchannel (15) and the detection structure area (16) are both 10 μm-30 μm; the heights of the front-stage microcolumn array (1611) and the rear-stage microcolumn array (1631) are both 10 μm-30 μm; the front-stage microcolumn gaps (1612) formed between the microcolumns in the front-stage microcolumn array (1611) are 9 μm-12 μm; the rear-stage microcolumn gaps (1632) formed between the microcolumns in the rear-stage microcolumn array (1631) are 6 μm-9 μm; the front-stage microcolumn array (1611) and the rear-stage microcolumn array (1631) are both perpendicular to the substrate (11); The leukocyte classification microfluidic chip (1) is mainly used to provide a structure for distinguishing different types of leukocytes and to provide a leukocyte image sampling microenvironment; the microscope (2) and the high-speed camera (3) mounted on the microscope are used to collect leukocyte images; the leukocyte classification and counting platform (4) is used to deploy a target detection model for tracking and detecting leukocytes of different types and a counting function module for counting leukocytes of different types; the target detection model will track, detect and classify a certain type of leukocytes of different types in the front-end classification and counting area (161), the middle-end classification and counting area (162) and the rear-end classification and counting area (163).

2. The white blood cell classification detection system according to claim 1, characterized in that: The hole radius of the sample inlet (13) and the sample outlet (14) is 400 μm-600 μm.

3. The white blood cell classification detection system according to claim 1, characterized in that: The width of the microchannel (15) is 20 μm-250 μm.

4. The white blood cell classification detection system according to claim 1, characterized in that: The material of the substrate (11) is glass, and the cover plate (12) is a polydimethylsiloxane detection structure module.

5. A method for detecting leukocyte classification, using the leukocyte classification detection system according to any one of claims 1 to 4, characterized in that: The following steps are involved: S1. Mixing whole blood with a hemolytic agent for blood cell analysis to remove red blood cells from the whole blood, and then filtering the mixture after removing red blood cells through a cell filter screen to remove larger debris and particles in the mixture to obtain a white blood cell mixture; S2, placing the leukocyte classification microfluidic chip (1) on the stage of a microscope (2), and placing the detection structure area (16) directly below the objective lens; S3, introducing the mixed leukocyte solution obtained in S1 into the microchannel (15) from the injection port (13) via a micro pump; S4. The high-speed camera (3) on the microscope (2) collects images passing through the detection structure area (16), inputs the collected images into the white blood cell classification and counting platform (4), and detects, classifies and counts one of the different types of white blood cells in the front classification and counting area (161), the middle classification and counting area (162) and the rear classification and counting area (163).

6. The method for detecting white blood cell classification according to claim 5, characterized in that: The S1 is specifically: The whole blood is mixed with the hemolytic agent for blood cell analysis at a volume ratio of 1:3-20 to dissolve the red blood cells in the whole blood; the pore size of the cell filter screen is 15μm-500μm, which will filter out larger debris and particles in the mixed solution.

7. The method for detecting white blood cell classification according to claim 5, characterized in that: In S2, the magnification of the microscope (2) is 100-1000 times.

8. The method for detecting white blood cell classification according to claim 5, characterized in that: In S4, the acquired image will be input into the target detection model in the white blood cell classification and counting platform (4), and the target detection model will track, detect and classify a certain category of white blood cells of different categories in the front classification and counting area (161), the middle classification and counting area (162) and the rear classification and counting area (163).

9. The method for detecting white blood cell classification according to claim 5, characterized in that: In S4, the counting function module in the white blood cell classification and counting platform (4) will count one of the different types of white blood cells in the front-end classification and counting area (161), the middle-end classification and counting area (162) and the rear-end classification and counting area (163), respectively, to complete the counting task of the three types of white blood cells. The core algorithm is as follows: if WBC_x>Mreg_Lx and WBC_x<Mreg_Rx and WBC_y>Mreg_Tx and WBC_y<Mreg_Dx and WBC_type==type0 WBC0←WBC0+1 if WBC_x>Freg_Lx and WBC_x<Freg_Rx and WBC_y>Freg_Tx and WBC_y<Freg_Dx and WBC_type==type1 WBC1←WBC1+1 if WBC_x>Lreg_Lx and WBC_x<Lreg_Rx and WBC_y>Lreg_Tx and WBC_y<Lreg_Dx and WBC_type==type2 WBC2←WBC2+1 Among them, WBC_x and WBC_y are the x-coordinate and y-coordinate of the center of the white blood cell, respectively; Mreg_Lx, Mreg_Rx 、 Mreg_Tx 、 Mreg_Dx are respectively the left and right upper and lower boundary line coordinates of the middle classification counting area (162); Freg_Lx, Freg_Rx, Freg_Tx, and Freg_Dx are respectively the left and right upper and lower boundary line coordinates of the front classification counting area (161); Lreg_Lx, Lreg_Rx, Lreg_Tx, and Lreg_Dx are respectively the left and right upper and lower boundary line coordinates of the rear classification counting area (163); WBC_type is the category of the detected white blood cells, and type0, type1, and type2 are respectively the preset three categories of white blood cell category values; WBC0 is the number of the first category of white blood cells counted in the middle classification counting area (162), WBC1 is the number of the second category of white blood cells counted in the front classification counting area (161), and WBC2 is the number of the third category of white blood cells counted in the rear classification counting area (163).

Images