Cell culture method, screening method and mesenchymal stem cell

Abstract

The invention provides a cell culture method, a cell screening method and mesenchymal stem cells, and the cell culture method comprises the following steps: carrying out multiplication culture on the mesenchymal stem cells, and carrying out real-time acquisition on the mesenchymal stem cells to obtain a target cell image; analyzing the target cell image, and determining cell statistical information; and when the cell statistical information meets the threshold requirement of the evaluation standard, screening out qualified mesenchymal stem cells.

Description

Cell culture methods, screening methods, and mesenchymal stem cells Technical Field

[0001] Embodiments of the present disclosure relate to a cell culture method, a cell screening method, and mesenchymal stem cells. Background Art

[0002] Mesenchymal stem cells (MSCs) are a type of cell with the potential for self-renewal and multidirectional differentiation that can be isolated from various tissues. For example, MSCs can be isolated from bone marrow, adipose tissue, and umbilical cord, and have been widely used for tissue and organ regeneration. Currently, researchers in related fields believe that the main biological functions of MSCs are achieved through paracrine effects. Through the secretion of cytokines by MSCs, the purpose of regulating immunity, promoting angiogenesis, and promoting the homing of autologous stem cells can be achieved.

[0003] Mesenchymal stem cells (MSCs) hold great promise for clinical application as biological agents. However, the use of MSCs as biological agents requires not only in vitro culturing of MSCs to achieve the desired cell number, but also comprehensive quality control of the proliferation, passage, and expansion processes to ensure the quality, safety, and reliability of the resulting MSCs, meeting clinical application standards.

[0004] Summary of the Invention

[0005] At least one embodiment of the present disclosure provides a cell culture method, a cell screening method, and a mesenchymal stem cell. The cell culture method acquires target cell images in real time from mesenchymal stem cells during proliferation and culture, obtains a final target image segmentation result, and determines cell statistical information based on the final target image segmentation result. When the cell statistical information meets the threshold requirements of an evaluation standard, qualified mesenchymal stem cells are screened and cultured for the next generation. This avoids the need to culture poor-quality mesenchymal stem cells, thus avoiding ineffective work and improving cell culture efficiency. Specifically, the embodiments of the present disclosure establish a relationship between the morphology and function of mesenchymal stem cells and provide a cell culture method based on cell image segmentation, thereby enabling rapid and non-destructive evaluation of cell quality, thereby saving time and cost. Furthermore, the above operation can obtain image segmentation results without the need for manual data annotation, and the above operation is independent of the environment in which the cell images were captured, thus exhibiting strong robustness.

[0006] At least one embodiment of the present disclosure provides a cell culture method, comprising: proliferating and culturing mesenchymal stem cells, and acquiring target cell images of the mesenchymal stem cells in real time; analyzing the target cell images to determine cell statistical information; and screening qualified mesenchymal stem cells when the cell statistical information meets a threshold requirement of an evaluation standard.

[0007] For example, in the cell culture method provided in at least one embodiment of the present disclosure, analyzing the target cell image includes: obtaining a target cell nuclear map based on the initial image segmentation result of the target cell image, wherein the target cell image is an image corresponding to at least one cell; obtaining a target cell area contour map based on the target cell nuclear map and the initial image segmentation result; obtaining the target cell contour map based on the target cell area contour map and the target cell nuclear map; and obtaining a target image segmentation result based on the target cell contour map and the target cell nuclear map.

[0008] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the cell statistical information includes at least one of the following: cell number, area of ​​a single cell, perimeter of a single cell, ratio of the major axis length of a single cell to the minor axis length of a single cell, cell density, and cell confluence.

[0009] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the threshold requirement for the number of cells is greater than or equal to 50,000, the threshold requirement for the area of ​​a single cell is less than or equal to 800 square microns, the threshold requirement for the perimeter of a single cell is less than or equal to 160 microns, the threshold requirement for the ratio of the major axis length of a single cell to the minor axis length of a single cell is less than or equal to 2.8, the threshold requirement for the cell density is less than or equal to 34, and the threshold requirement for the cell confluence is greater than or equal to 80%.

[0010] For example, the cell culture method provided in at least one embodiment of the present disclosure further includes: determining a quality assessment result of the cells based on the cell statistical information.

[0011] For example, in the cell culture method provided in at least one embodiment of the present disclosure, an initial image segmentation result of the target cell image is obtained based on a threshold segmentation method.

[0012] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the threshold segmentation method includes a maximum inter-class variance method, the initial image segmentation result includes a cell edge map, a cell area map and a cell nucleus area map, and based on the threshold segmentation method, the initial image segmentation result of the target cell image is obtained, including: based on the maximum inter-class variance method, according to the target cell image, obtaining a first grayscale threshold, a second grayscale threshold and a third grayscale threshold, wherein the second grayscale threshold is greater than the third grayscale threshold and less than the first grayscale threshold; binarizing the target cell image according to the first grayscale threshold to obtain the cell edge map; binarizing the target cell image according to the second grayscale threshold to obtain the cell area map; and binarizing the target cell image according to the third grayscale threshold to obtain the cell nucleus area map.

[0013] For example, in the cell culture method provided in at least one embodiment of the present disclosure, a target cell nucleus map is obtained based on the initial image segmentation result of the target cell image, including: obtaining an intermediate cell region map based on the cell region map and the cell edge map; obtaining at least one connected domain cell region map based on the intermediate cell region map and a predetermined foreground region division strategy; and obtaining the target cell nucleus map based on the cell nucleus region map and the at least one connected domain cell region map.

[0014] For example, in the cell culture method provided in at least one embodiment of the present disclosure, an intermediate cell area map is obtained according to the cell area map and the cell edge map, including: processing the cell area map using an expansion operation to obtain a first cell area expansion map; processing the first cell area expansion map using a fill operation to obtain a cell area expansion fill map; processing the cell area expansion fill map using an corrosion operation to obtain a cell area fill map; and removing the target cell edges in the cell area fill map according to the cell area fill map and the cell edge map to obtain the intermediate cell area map.

[0015] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the target cell nucleus map is obtained according to the cell nucleus area map and the at least one connected domain cell area map, including: for the connected domain cell area map in the at least one connected domain cell area map, according to the cell nucleus area map and the connected domain cell area map, obtaining the cell nucleus map corresponding to the connected domain cell area map; and obtaining the target cell nucleus map according to the cell nucleus maps corresponding to each of the at least one connected domain cell area map.

[0016] For example, in the cell culture method provided in at least one embodiment of the present disclosure, a cell nucleus map corresponding to the connected domain cell area map is obtained based on the cell nucleus area map and the connected domain cell area map, including: when it is determined that the area of ​​the connected domain cell area map is greater than or equal to a first predetermined area threshold, processing the connected domain cell area map to obtain an intermediate connected domain cell area map; and, based on the intermediate connected domain cell area map and the cell nucleus area map, obtaining a cell nucleus map corresponding to the connected domain cell area map; and, when it is determined that the area of ​​the connected domain cell area map is less than the first predetermined area threshold, obtaining a cell nucleus map corresponding to the connected domain cell area map based on the cell nucleus area map and the connected domain cell area map.

[0017] For example, in the cell culture method provided in at least one embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the intermediate connected domain cell region map and the cell nucleus region map includes: performing an AND operation on the intermediate connected domain cell region map and the cell nucleus region map to obtain a first intersection cell region map; and obtaining a cell nucleus map corresponding to the connected domain cell region map based on the first intersection cell region map and the intermediate connected domain cell region map.

[0018] For example, in the cell culture method provided in at least one embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the first intersection cell region map and the intermediate connected domain cell region map includes: generating a first template image, wherein the size of the first template image is equal to the size of the intermediate connected domain cell region map, and the pixel value of the pixel of the first template image is a first predetermined pixel value; obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map based on the first template image, the first intersection cell region map and the intermediate connected domain cell region map; and obtaining a cell nucleus map corresponding to the connected domain cell region map based on the first candidate cell nucleus map corresponding to the connected domain cell region map.

[0019] For example, in the cell culture method provided in at least one embodiment of the present disclosure, obtaining the first candidate cell nucleus map corresponding to the connected domain cell region map based on the first template image, the first intersection cell region map and the intermediate connected domain cell region map includes: determining a first contour set corresponding to the connected domain in the first intersection cell region map; and traversing each first predetermined pixel included in the first contour set, modifying the pixel value of the pixel with the same label as the first predetermined pixel in the first template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, and obtaining the first candidate cell nucleus map corresponding to the connected domain cell region map, wherein the label of the first predetermined pixel in the first template image is determined based on the label of the first predetermined pixel in the intermediate connected domain cell region map.

[0020] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the first candidate cell nucleus map corresponding to the connected domain cell region map is obtained based on the first template image, the first intersection cell region map and the intermediate connected domain cell region map, including: traversing each second predetermined pixel in the first template image, and when it is determined that the pixel value of the pixel corresponding to the second predetermined pixel in the first intersection cell region map is the expected pixel value, modifying the pixel value of the pixel with the same label as the second predetermined pixel in the first template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, and the first candidate cell nucleus map corresponding to the connected domain cell region map is obtained, wherein the label of the second predetermined pixel in the first template image is determined based on the label of the second predetermined pixel in the intermediate connected domain cell region map.

[0021] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the cell nucleus map corresponding to the connected domain cell area map is obtained according to the cell nucleus area map and the connected domain cell area map, including: performing an AND operation on the connected domain cell area map and the cell nucleus area map to obtain a second intersection cell area map; and obtaining a cell nucleus map corresponding to the connected domain cell area map according to the second intersection cell area map and the connected domain cell area map; generating a second template image, wherein the size of the second template image is equal to the size of the connected domain cell area map, and the pixel value of the pixel of the second template image is the first predetermined pixel value; obtaining a second candidate cell nucleus map corresponding to the connected domain cell area map according to the second template image, the second intersection cell area map and the connected domain cell area map; and obtaining a cell nucleus map corresponding to the connected domain cell area map according to the second candidate cell nucleus map corresponding to the connected domain cell area map.

[0022] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the second candidate cell nucleus map corresponding to the connected domain cell region map is obtained based on the second template image, the second intersection cell region map and the connected domain cell region map, including: determining a second contour set corresponding to the connected domain in the second intersection cell region map; and traversing each third predetermined pixel included in the second contour set, modifying the pixel value of the pixel with the same label as the third predetermined pixel in the second template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, and obtaining the second candidate cell nucleus map corresponding to the connected domain cell region map, wherein the label of the third predetermined pixel in the second template image is determined based on the label of the third predetermined pixel in the connected domain cell region map.

[0023] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the second candidate cell nucleus map corresponding to the connected domain cell area map is obtained based on the second template image, the second intersection cell area map and the connected domain cell area map, including: traversing each fourth predetermined pixel in the second template image, and when it is determined that the pixel value of the pixel corresponding to the fourth predetermined pixel in the second intersection cell area map is the expected pixel value, modifying the pixel value of the pixel with the same label as the fourth predetermined pixel in the second template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, and the second candidate cell nucleus map corresponding to the connected domain cell area map is obtained, wherein the label of the fourth predetermined pixel in the second template image is determined based on the label of the fourth predetermined pixel in the connected domain cell area map.

[0024] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the initial image segmentation result includes a cell edge map and a cell area map; based on the target cell nucleus map and the initial image segmentation result, a target cell area contour map is obtained, including: based on the target cell nucleus map and the cell area map, a primary cell area map is obtained; using an expansion operation to process the initial cell area map to obtain a second cell area expansion map; based on the second cell area expansion map and the primary cell area map, a first intermediate cell area contour map is obtained; based on the first intermediate cell area contour map and the target cell image, a second intermediate cell area contour map is obtained; based on the second intermediate cell area contour map and the target cell nucleus map, a third intermediate cell area contour map is obtained; and based on the third intermediate cell area contour map and the cell edge map, the target cell area contour map is obtained.

[0025] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the second intermediate cell area contour map is obtained based on the first intermediate cell area contour map and the target cell image, including: performing watershed processing on the first intermediate cell area contour map based on the target cell image to obtain a fourth intermediate cell area contour map; and setting the pixel values ​​of the pixels in the background area in the fourth intermediate cell area contour map to the first predetermined pixel value, and setting the pixel values ​​of the pixels in other areas of the fourth intermediate cell area contour map except the background area to the second predetermined pixel value, to obtain the second intermediate cell area contour map.

[0026] For example, in the cell culture method provided in at least one embodiment of the present disclosure, obtaining the third intermediate cell area contour map based on the second intermediate cell area contour map and the target cell nucleus map includes: generating a third template image, wherein the size of the third template image is equal to the size of the second intermediate cell area contour map, and the pixel value of the pixel of the third template image is the first predetermined pixel value; and obtaining the third intermediate cell area contour map based on the third template image, the second intermediate cell area contour map and the target cell nucleus map.

[0027] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the third intermediate cell area contour map is obtained based on the third template image, the second intermediate cell area contour map and the target cell nucleus map, including: determining a third contour set corresponding to the connected domain in the target cell nucleus map; and traversing each fifth predetermined pixel included in the third contour set, modifying the pixel value of the pixel with the same label as the fifth predetermined pixel in the third template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, to obtain the third intermediate cell area contour map, wherein the label of the fifth predetermined pixel in the third template image is determined based on the label of the fifth predetermined pixel in the second intermediate cell area contour map.

[0028] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the third intermediate cell area contour map is obtained according to the third template image, the second intermediate cell area contour map and the target cell nucleus map, including: traversing each sixth predetermined pixel in the third template image, and when it is determined that the pixel value of the pixel corresponding to the sixth predetermined pixel in the target cell nucleus map is the expected pixel value, modifying the pixel value of the pixel with the same label as the sixth predetermined pixel in the third template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, and the third intermediate cell area contour map is obtained, wherein the label of the sixth predetermined pixel in the third template image is determined according to the label of the sixth predetermined pixel in the second intermediate cell area contour map.

[0029] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the target cell contour map is obtained based on the target cell area contour map and the target cell nucleus map, including: subtracting the target cell area contour map and the target cell nucleus map to obtain the intermediate cell contour map; and performing watershed processing on the intermediate cell contour map based on the target cell image to obtain the target cell contour map.

[0030] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the proliferation culture of mesenchymal stem cells includes: adding fetal bovine serum to a mesenchymal stem cell culture medium to form a first culture medium; adding the mesenchymal stem cells to the first culture medium for culture, and when the confluence of the mesenchymal stem cells after proliferation is in the range of 70% to 95%, the mesenchymal stem cells are passaged.

[0031] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the mass percentage of the fetal bovine serum in the first culture medium is 1% to 10%.

[0032] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the seeding density of the mesenchymal stem cells in the first culture medium is 1000 to 30000 / cm 2 .

[0033] For example, in the cell culture method provided in at least one embodiment of the present disclosure, during the process of culturing the mesenchymal stem cells in the first culture medium, the mesenchymal stem cell culture medium is replaced every two days.

[0034] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the passaging operation of the mesenchymal stem cells includes: aspirating the mesenchymal stem cell culture medium in the first cell culture container, and cleaning the cell culture container with a buffer detergent having a pH of 7.2 to 7.4, and aspirating the buffer detergent; adding 7ML to 10ML of pancreatic enzyme to the first cell culture container, and digesting the mesenchymal stem cells with the pancreatic enzyme for 1 to 2 minutes; allowing the mesenchymal stem cells to leave the bottom surface of the first cell culture container, adding the mesenchymal stem cell culture medium to the first cell culture container to terminate the digestion of the mesenchymal stem cells and form a first cell suspension; using a 10ml pipette to collect the cell suspension into a 50ml centrifuge tube, and centrifuging to obtain a cell pellet; adding 3ML to 5ML of the mesenchymal stem cell culture medium to the cell pellet to form a second suspension.

[0035] For example, the cell culture method provided in at least one embodiment of the present disclosure further includes: mixing the second suspension evenly, aspirating the second suspension into a sterile centrifuge tube, adding an amount of dye equal to the aspirated second suspension into the centrifuge tube, and mixing evenly to obtain a third suspension.

[0036] For example, the cell culture method provided by at least one embodiment of the present disclosure further includes: adding the third suspension to a cell counting plate to perform cell counting, then inoculating the second suspension into a second cell culture container according to a predetermined inoculation density, then adding the mesenchymal stem cell culture medium to the second cell culture container, and culturing the next generation of mesenchymal stem cells.

[0037] For example, the cell culture method provided in at least one embodiment of the present disclosure further includes: performing cell activity detection, aging-related β-galactosidase detection, colony formation ability detection, cell cycle detection, in vitro angiogenesis ability detection, peripheral blood mononuclear cell immunosuppression ability detection, mesenchymal stem cell induced differentiation staining test and mesenchymal stem cell surface marker test on the screened mesenchymal stem cells.

[0038] For example, in the cell culture method provided in at least one embodiment of the present disclosure, performing cell activity detection on the screened mesenchymal stem cells includes: adding CCK-8 to the mesenchymal stem cells to secrete an orange-red formazan product, and detecting the specific absorbance of light with a wavelength of 450 nm. The higher the absorbance, the better the activity of the mesenchymal stem cells.

[0039] For example, in the cell culture method provided in at least one embodiment of the present disclosure, performing senescence-related β-galactosidase assay on the screened mesenchymal stem cells includes: using an in situ β-galactosidase staining kit to detect cell senescence, and the more blue staining, the higher the proportion of senescent cells.

[0040] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the colony-forming ability of the screened mesenchymal stem cells is determined by inoculating the cells at a concentration such that one mesenchymal stem cell is located in one well, culturing the cells for 14 days, and observing under a microscope the number of wells containing 50 or more mesenchymal stem cells, which represents the colony-forming ability of the mesenchymal stem cells.

[0041] For example, in the cell culture method provided in at least one embodiment of the present disclosure, performing cell cycle detection on the screened mesenchymal stem cells includes: when the confluence of the mesenchymal stem cells reaches 60% to 70%, obtaining the mesenchymal stem cells, fixing the mesenchymal stem cells with 70% ethanol by volume, staining the mesenchymal stem cells with a cell cycle staining kit, and detecting the cell cycle using flow cytometry.

[0042] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the in vitro angiogenesis ability test of the screened mesenchymal stem cells includes: inoculating the mesenchymal stem cells in a culture flask, replacing the mesenchymal stem cell culture medium when the confluence of the mesenchymal stem cells reaches 60% to 70%, and continuing to culture the mesenchymal stem cells for 24 hours, collecting the supernatant of the mesenchymal stem cell culture medium, centrifuging the mesenchymal stem cell supernatant, and filtering the mesenchymal stem cell supernatant using a filter membrane with a pore size of 0.22 μm; incubating a well plate coated with matrix gel in an incubator for 30 minutes, inoculating human umbilical vein endothelial cells on the matrix gel, adding 100 μl of the culture medium supernatant to the wells of the well plate, and then capturing images of the formed blood vessels to obtain blood vessel images, observing the connection points and total length of the blood vessels. The more connection points and the longer the total length of the blood vessels, the stronger the in vitro angiogenesis ability of the mesenchymal stem cells.

[0043] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the immunosuppressive ability of peripheral blood mononuclear cells of the screened mesenchymal stem cells is tested, including: treating the mesenchymal stem cells with colchicine to inhibit the proliferation of the mesenchymal stem cells, adding the peripheral blood mononuclear cells to a 6-well plate, using 1640 culture medium containing 10% fetal bovine serum by volume, and co-culturing at a temperature of 37° C. and a humidity of 95%, then staining the peripheral blood mononuclear cells with a BrdU staining kit, and then detecting the peripheral blood mononuclear cells with a flow cytometer, wherein the peripheral blood mononuclear cells that are BrdU-positive during the detection are newly proliferated.

[0044] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the mesenchymal stem cells screened out are subjected to a mesenchymal stem cell induced differentiation staining test, including: inducing differentiation of the mesenchymal stem cells using an adipogenic and osteogenic differentiation kit, and then detecting the adipogenic, chondrogenic and osteogenic differentiation results by Oil Red O staining, Alcian Blue staining and Alizarin Red staining, respectively.

[0045] For example, in the cell culture method provided in at least one embodiment of the present disclosure, the screened mesenchymal stem cells are subjected to a mesenchymal stem cell surface marker test, which includes: washing the mesenchymal stem cells twice with a buffer detergent having a pH of 7.2 to 7.4, staining with antibodies CD11B-PE, CD19-FITC, CD34-PE, CD45-APC-CY7, CD73-PE, CD90-FITC, CD105-APC, HLA-DR-APC and corresponding isotype control antibodies at room temperature in the dark for 30 minutes, then washing twice with a buffer detergent having a pH of 7.2 to 7.4, and detecting the markers on the surface of the mesenchymal stem cells by flow cytometry analysis.

[0046] At least one embodiment of the present disclosure further provides a mesenchymal stem cell, which is formed using any of the cell culture methods described above.

[0047] At least one embodiment of the present disclosure further provides a cell screening method, which includes: real-time acquisition of the mesenchymal stem cells during proliferation culture to obtain a target cell image; obtaining a target cell nuclear image based on an initial image segmentation result of the target cell image, wherein the target cell image is an image corresponding to at least one cell; obtaining a target cell area contour map based on the target cell area contour map and the target cell nuclear image; obtaining the target cell contour map based on the target cell area contour map and the target cell nuclear image; obtaining a target image segmentation result based on the target cell contour map and the target cell nuclear image; determining cell statistical information based on the target image segmentation result; and screening out qualified mesenchymal stem cells when the cell statistical information meets the threshold requirement of an evaluation standard.

[0048] For example, in the cell screening method provided in at least one embodiment of the present disclosure, the cell statistical information includes at least one of the following: the number of cells, the area of ​​a single cell, the perimeter of a single cell, the ratio of the major axis length of a single cell to the minor axis length of a single cell, the cell density and the cell confluence.

[0049] For example, in the cell screening method provided in at least one embodiment of the present disclosure, the threshold requirement for the number of cells is greater than or equal to 50,000, the threshold requirement for the area of ​​a single cell is less than or equal to 800 square microns, the threshold requirement for the perimeter of a single cell is less than or equal to 160 microns, the threshold requirement for the ratio of the major axis length of a single cell to the minor axis length of a single cell is less than or equal to 2.8, the threshold requirement for the cell density is less than or equal to 34, and the threshold requirement for the cell confluence is greater than or equal to 80%.

[0050] For example, the cell screening method provided in at least one embodiment of the present disclosure further includes: determining a quality assessment result of the cell based on the cell statistical information. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments will be briefly introduced below. Obviously, the drawings in the following description only relate to some embodiments of the present disclosure, rather than limiting the present disclosure.

[0052] FIG1 is a process diagram of a cell culture method provided by at least one embodiment of the present disclosure;

[0053] FIG2 is a system architecture to which the cell image processing method and apparatus provided by at least one embodiment of the present disclosure can be applied;

[0054] FIG3 is a flow chart of a cell image processing method provided by at least one embodiment of the present disclosure;

[0055] FIG4A schematically shows a principle diagram of a cell image processing process according to an embodiment of the present disclosure;

[0056] FIG4B schematically shows a cell nucleus region map and a connected domain cell region map according to an embodiment of the present disclosure;

[0057] FIG5A schematically shows an example diagram of an initial image segmentation result according to an embodiment of the present disclosure;

[0058] FIG5B schematically shows an example schematic diagram of obtaining an intermediate cell area map according to an embodiment of the present disclosure;

[0059] FIG5C schematically shows an example schematic diagram of obtaining a connected domain cell area map according to an embodiment of the present disclosure;

[0060] FIG5D schematically shows an example schematic diagram of obtaining a cell nucleus map corresponding to a connected domain cell region map according to an embodiment of the present disclosure;

[0061] FIG5E schematically shows an example schematic diagram of a cell nucleus map corresponding to a connected domain cell region map according to another embodiment of the present disclosure;

[0062] FIG5F schematically shows an example schematic diagram of obtaining a cell nucleus map corresponding to a connected domain cell region map according to another embodiment of the present disclosure;

[0063] FIG5G schematically shows an example schematic diagram of obtaining a target cell nucleus image according to an embodiment of the present disclosure;

[0064] FIG5H schematically shows an example schematic diagram of obtaining a target cell region contour map according to an embodiment of the present disclosure;

[0065] FIG5I schematically shows an example schematic diagram of obtaining a target cell outline map according to an embodiment of the present disclosure;

[0066] FIG5J schematically shows an example of a target image segmentation result at the early stage of cell culture according to an embodiment of the present disclosure;

[0067] FIG5K schematically shows an example of a target image segmentation result at the late stage of cell culture according to an embodiment of the present disclosure;

[0068] FIG6 is a block diagram of a cell image processing device provided by at least one embodiment of the present disclosure;

[0069] FIG7 schematically shows a block diagram of an electronic device suitable for implementing a cell image processing method according to an embodiment of the present disclosure;

[0070] FIG8 is a flow chart of proliferation and culture of mesenchymal stem cells according to at least one embodiment of the present disclosure;

[0071] FIG9 is an image of mesenchymal stem cells at passages 5 and 10 according to at least one embodiment of the present disclosure;

[0072] FIG10A is a comparison chart of the cell counts of mesenchymal stem cells at passage 5 and passage 10 according to at least one embodiment of the present disclosure;

[0073] FIG10B is a graph comparing the cell confluence of passage 5 mesenchymal stem cells and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure;

[0074] FIG10C is a diagram comparing the cell areas of passage 5 mesenchymal stem cells and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure;

[0075] FIG10D is a comparison diagram of the cell perimeters of mesenchymal stem cells at passage 5 and passage 10 according to at least one embodiment of the present disclosure;

[0076] FIG10E is a comparison diagram of the cell aspect ratios of passage 5 mesenchymal stem cells and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure;

[0077] FIG10F is a diagram comparing the cell density of passage 5 mesenchymal stem cells and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure;

[0078] FIG11 is a graph showing the cell activity of passage 5 mesenchymal stem cells and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure;

[0079] FIG12 is an image of 5th and 10th passage mesenchymal stem cells stained with β-galactosidase according to at least one embodiment of the present disclosure;

[0080] FIG13 is a graph comparing the colony-forming abilities of passage 5 mesenchymal stem cells and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure;

[0081] FIG14 is a diagram comparing the cell cycles of passage 5 mesenchymal stem cells and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure;

[0082] FIG15 is a graph comparing the in vitro angiogenesis capabilities of passage 5 mesenchymal stem cells and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure;

[0083] FIG16 is a graph showing the test results of the immunosuppressive ability of peripheral blood mononuclear cells of the 5th and 10th generation mesenchymal stem cells provided by at least one embodiment of the present disclosure;

[0084] FIG17 is an electron micrograph of three samples of mesenchymal stem cells of passage 5 and passage 10, respectively, induced to differentiate into adipocytes, chondroblasts, and osteoblasts in vitro, according to at least one embodiment of the present disclosure; and

[0085] FIG18 is a flow chart of a cell screening method provided by at least one embodiment of the present disclosure. DETAILED DESCRIPTION

[0086] To make the purpose, technical solutions, and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the described embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0087] Unless otherwise defined, technical or scientific terms used in this disclosure should have the ordinary meanings understood by people with ordinary skills in the field to which this disclosure belongs. The words "first", "second" and similar terms used in this disclosure do not indicate any order, quantity or importance, but are simply used to distinguish different components. The words "include" or "comprising" and similar terms mean that the elements or objects preceding the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0088] When expressions such as “at least one of A, B, and C, etc.” are used, they should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art (e.g., “a system having at least one of A, B, and C” should include but is not limited to systems having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, and C, etc.). When expressions such as “at least one of A, B, or C, etc.” are used, they should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art (e.g., “a system having at least one of A, B, or C” should include but is not limited to systems having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, and C, etc.).

[0089] Cell proliferation, cell surface marker detection, and tri-directional cell differentiation are conventional methods for identifying whether mesenchymal stem cells can be used as biological agents in clinical practice. However, these identification methods cannot determine the biological functions of mesenchymal stem cells. Mesenchymal stem cells are heterogeneous, meaning that the functional properties of mesenchymal stem cells from different donors or different tissues are different. Furthermore, as mesenchymal stem cells are cultured in vitro for extended periods of time and their passages increase, they gradually age. Mesenchymal stem cells from the same donor but at different passages also have different properties. Generally, high-passage mesenchymal stem cells have a decreased proliferation rate, poorer differentiation ability, and reduced biological function. Enzyme-linked immunosorbent assays (ELISAs) can also be used to detect cytokine secretion in mesenchymal stem cells, as can co-culturing with other cells in vitro to directly test their function. The properties of mesenchymal stem cells can also be evaluated by measuring cell cycle, differentiation induction ability, and β-galactosidase staining. However, current methods for testing the biological functions of mesenchymal stem cells are all analytical methods that damage mesenchymal stem cells. Moreover, the test results obtained during the quality control process are not immediate, making quality control more of a quality inspection rather than a control and prediction, which increases the risks of using cell-based products. In addition, during the process of culturing mesenchymal stem cells, confluence is often used as a criterion for evaluating cell proliferation. This evaluation is subjective and can vary between operators, thus affecting the quality of the final cells.

[0090] The inventors of the present disclosure have noticed that it is possible to collect cell images of mesenchymal stem cells during proliferation, and then segment the cell images to obtain relevant parameters such as cell morphology and cell count. For example, cell morphology includes cell area, perimeter, density, and aspect ratio, and cell count includes cell number and confluence. The quality of mesenchymal stem cells can be judged by the obtained cell morphology and cell count parameters to decide whether to continue to proliferate and culture them or to determine whether they can be used as biological preparations in clinical practice. In addition, the inventors of the present disclosure also used traditional cell biology experiments to detect functional characteristics such as surface markers, cell activity, cell cycle, cell differentiation induction ability, in vitro angiogenesis promotion ability, and in vitro immune cell proliferation inhibition ability of mesenchymal stem cells to further verify the reliability of using cell image segmentation to judge the quality of mesenchymal stem cells. That is, the inventors of the present disclosure established the relationship between the morphology and function of mesenchymal stem cells and provided a method for cell culture based on cell image segmentation, which can quickly and non-destructively evaluate the quality of cells, thereby saving the time and cost of cell culture.

[0091] For example, segmentation of cell images can refer to the use of computer image processing techniques to divide different regions in the cell image, extract valuable parenchymal cell regions, and further separate overlapping cells to obtain the nucleus and cytoplasm of a single cell, providing a basis for subsequent quantitative analysis.

[0092] For example, deep learning models can be used to segment cell images. These models can be trained using sample images. These sample images require manual data labeling. After model training, if the environment in which the cell images were taken changes, the model needs to be retrained. Consequently, deep learning models have limited robustness.

[0093] For example, at least one embodiment of the present disclosure provides a cell culture method, which includes: proliferating and culturing mesenchymal stem cells, and acquiring target cell images of the mesenchymal stem cells in real time; analyzing the target cell images to determine cell statistical information; and screening qualified mesenchymal stem cells when the cell statistical information meets the threshold requirements of an evaluation standard. For example, analyzing the target cell image includes: obtaining a target cell nuclear map based on the initial image segmentation result of the target cell image, wherein the target cell image is an image corresponding to at least one cell; obtaining a target cell area contour map based on the target cell nuclear map and the initial image segmentation result; obtaining a target cell contour map based on the target cell area contour map and the target cell nuclear map; obtaining a target image segmentation result based on the target cell contour map and the target cell nuclear map; determining cell statistical information based on the target image segmentation result; when the cell statistical information meets the threshold requirement of the evaluation standard, screening out qualified mesenchymal stem cells, and culturing the qualified mesenchymal stem cells for the next generation. That is, the embodiment of the present disclosure establishes the relationship between the morphology and function of mesenchymal stem cells, and provides a method based on cell image segmentation, so that cell quality can be evaluated quickly and non-destructively, thereby saving time and cost.

[0094] For example, FIG1 is a process diagram of a cell culture method provided by at least one embodiment of the present disclosure. As shown in FIG1 , the cell culture method includes the following steps.

[0095] Step S101: Proliferating and culturing mesenchymal stem cells, and acquiring target cell images in real time.

[0096] Step S102: obtaining a target cell nucleus image according to an initial image segmentation result of a target cell image, wherein the target cell image is an image corresponding to at least one cell.

[0097] Step S103: Obtaining a target cell region contour map based on the target cell nucleus map and the initial image segmentation result.

[0098] Step S104: obtaining a target cell outline map according to the target cell region outline map and the target cell nucleus map.

[0099] Step S105: Obtaining a target image segmentation result according to the target cell outline image and the target cell nucleus image.

[0100] Step S106: determining cell statistical information according to the target image segmentation result.

[0101] Step S107: When the cell statistical information meets the threshold requirement of the evaluation standard, qualified mesenchymal stem cells are screened out and the qualified mesenchymal stem cells are cultured for the next generation.

[0102] For example, mesenchymal stem cells are cultured for proliferation, and the mesenchymal stem cells during the proliferation culture process are captured in real time to obtain target cell images, and a final target image segmentation result is obtained. Cell statistical information is determined based on the final target image segmentation result. When the cell statistical information meets the threshold requirements of the evaluation criteria, qualified mesenchymal stem cells are screened and cultured for the next generation of qualified mesenchymal stem cells, thereby avoiding the need to continue culturing mesenchymal stem cells of poor quality, thus avoiding ineffective work and improving the efficiency of cell culture. That is, the embodiments of the present disclosure establish the relationship between the morphology and function of mesenchymal stem cells and provide a cell culture method based on cell image segmentation, which can quickly and non-destructively evaluate cell quality, thereby saving time and cost. In addition, the above operation does not require human effort for data annotation to obtain image segmentation results, and the above operation is independent of the environment in which the cell images are taken, and has strong robustness.

[0103] It should be noted that the sequence numbers of the operations in the above method are only used to indicate the operation for the purpose of description and should not be regarded as indicating the order in which the operations should be performed. Unless explicitly stated, the method does not need to be performed in the order shown.

[0104] For example, after obtaining the target image segmentation result, cell statistical information can be determined, and a cell quality assessment result can be determined based on the cell statistical information. For example, the cell statistical information includes at least one of the following: cell number, area of ​​a single cell, perimeter of a single cell, ratio of the major axis length to the minor axis length of a single cell, cell density, and cell confluence.

[0105] For example, assessing cell quality based on image segmentation results can help mitigate errors caused by human factors, standardize cell culture processes, and effectively reduce production costs. In actual production processes, this technology can rapidly analyze continuous target cell images with statistical accuracy exceeding 90%.

[0106] For example, in the embodiments of the present disclosure, the threshold requirement for the number of cells is greater than or equal to 50,000, the threshold requirement for the area of ​​a single cell is less than or equal to 800 square microns, the threshold requirement for the perimeter of a single cell is less than or equal to 160 microns, the threshold requirement for the ratio of the major axis length of a single cell to the minor axis length of a single cell is less than or equal to 2.8, the threshold requirement for the cell density is less than or equal to 34, and the threshold requirement for the cell confluence is greater than or equal to 80%.

[0107] For example, when evaluating cell quality based on image segmentation results, one criterion for determining whether the cell quality meets the requirements may be that the seven evaluation indicators, namely, the number of cells, the area of ​​a single cell, the perimeter of a single cell, the ratio of the major axis length to the minor axis length of a single cell, the cell density, and the cell confluence, are all within the corresponding ranges. Another criterion for determining whether the cell quality meets the requirements may be that six of the seven evaluation indicators, namely, the number of cells, the area of ​​a single cell, the perimeter of a single cell, the ratio of the major axis length to the minor axis length of a single cell, the cell density, and the cell confluence, are within the corresponding ranges. Yet another criterion for determining whether the cell quality meets the requirements may be that five of the seven evaluation indicators, namely, the number of cells, the area of ​​a single cell, the perimeter of a single cell, the ratio of the major axis length to the minor axis length of a single cell, the cell density, and the cell confluence, are within the corresponding ranges. Another criterion for judging whether the quality of cells meets the requirements may be that four of the seven evaluation indicators, namely, the number of cells, the area of ​​a single cell, the perimeter of a single cell, the ratio of the major axis length of a single cell to the minor axis length of a single cell, the cell density and the cell confluence, are within the corresponding ranges.

[0108] For example, Figure 2 shows a system architecture that can be applied to the cell image processing method and apparatus provided in at least one embodiment of the present disclosure. It should be noted that Figure 2 shows only an exemplary system architecture, but the embodiments of the present disclosure are not limited thereto, nor does it mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments, or scenarios. For example, in another embodiment, the exemplary system architecture that can be applied to the cell image processing method and apparatus may include a terminal device, but the terminal device can implement the cell image processing method and apparatus provided in the embodiments of the present disclosure without interacting with a server.

[0109] For example, as shown in FIG2 , the system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, and 103 and the server 105. The network 104 may include various connection types, such as wired and / or wireless communication links.

[0110] For example, a user may use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications may be installed on terminal devices 101, 102, and 103, such as at least one of a shopping application, a web browser application, a search application, an instant messaging tool, an email client, and a social platform software.

[0111] For example, the terminal devices 101 , 102 , and 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0112] For example, server 105 may be any type of server that provides various services. For example, it may be a backend management server that provides support for websites browsed by users using terminal devices 101, 102, and 103. The backend management server may analyze and process received data such as user requests, and feed back the processing results to the terminal device. The processing results may be, for example, web pages, information, or data obtained or generated based on the user requests.

[0113] It should be noted that the cell image processing method provided in the embodiments of the present disclosure can generally be executed by the terminal device 101, 102, or 103. Accordingly, the cell image processing apparatus provided in the embodiments of the present disclosure can also be provided in the terminal device 101, 102, or 103.

[0114] For example, the cell image processing method provided by the embodiment of the present disclosure may also be executed by the server 105. The cell image processing apparatus provided by the embodiment of the present disclosure may be provided in the server 105. The cell image processing method provided by the embodiment of the present disclosure may also be performed by a server or server cluster that is different from the server 105 and that is capable of communicating with the terminal devices 101, 102, 103 and / or the server 105. The cell image processing apparatus provided by the embodiment of the present disclosure may also be provided in a server or server cluster that is different from the server 105 and that is capable of communicating with the terminal devices 101, 102, 103 and / or the server 105.

[0115] It should be noted that the number of terminal devices, networks, and servers in Figure 2 is merely illustrative and any number of terminal devices, networks, and servers may be provided as required.

[0116] For example, Figure 3 is a flow chart of a cell image processing method provided by at least one embodiment of the present disclosure. As shown in Figure 3, the method 200 includes the following steps.

[0117] Step S210 , obtaining a target cell nucleus image according to an initial image segmentation result of a target cell image, wherein the target cell image is an image corresponding to at least one cell.

[0118] Step S220 , obtaining a target cell region contour map based on the target cell nucleus map and the initial image segmentation result.

[0119] Step S230 , obtaining a target cell outline map according to the target cell region outline map and the target cell nucleus map.

[0120] Step S240 , obtaining a target image segmentation result according to the target cell outline image and the target cell nucleus image.

[0121] For example, in embodiments of the present disclosure, cells may include eukaryotic cells and prokaryotic cells. Eukaryotic cells may include a nucleus, cytoplasm, and a cell membrane. In addition, based on the size of the pixel values ​​in the cell image, the cell may be divided into a cell edge region, a cell region, and a cell nuclear region. The cell region may include a cell nuclear region.

[0122] For example, in an embodiment of the present disclosure, the target cell image may be acquired using an image acquisition device. For example, the image acquisition device may include a visual sensor and a microscope. The microscope may include at least one of the following: an optical microscope and an electron microscope. The cell image may be acquired by microscopic imaging of the cell image acquired using a visual sensor. The microscopic imaging may be acquired by placing a cell smear of cells under a microscope and adjusting the relative distance between the microscope carrier stage and the microscope objective lens via the focusing system of the microscope. The target cell image may include an image of at least one cell. The target cell image may be a cell grayscale image. The cell grayscale image may refer to a cell image having grayscale values. The target cell image may refer to a cell image that requires image segmentation.

[0123] For example, in embodiments of the present disclosure, morphological operations can be used to extract relevant information from an image using structural elements, i.e., morphological methods. Morphological methods can be used to effectively suppress noise and simplify an image while maintaining its basic shape characteristics. By continuously moving structural elements within an image, the relationships between various image components can be determined, thereby achieving the purpose of image analysis.

[0124] For example, in the embodiments of the present disclosure, a morphological operation may refer to an operation of connecting adjacent pixels or separating adjacent pixels into independent pixels. The morphological operation may include at least one of the following: a basic morphological operation and other morphological operations. Other morphological operations may be obtained based on the basic morphological operations. For example, the basic morphological operations may include at least one of the following: a dilation operation and an erosion operation. Other morphological operations may include at least one of the following: an opening operation, a closing operation, a top-hat operation, a bottom-hat operation, a morphological gradient operation, and a filling operation. A structuring element may be a basic morphological operator. The reasonable selection of a structuring element plays an important role in improving the effect of image processing using morphological methods. For example, the structuring element may be determined based on at least one of its size, position, orientation, and shape. For example, the shape of the structuring element may include at least one of the following: a line, a square, a rhombus, a disk, and a sphere.

[0125] For example, in an embodiment of the present disclosure, the initial image segmentation result may refer to an image segmentation result in which the segmentation accuracy of the target cell image is less than or equal to a predetermined segmentation accuracy threshold. The predetermined segmentation accuracy threshold can be set according to actual needs and is not limited in the embodiments of the present disclosure. The initial cell segmentation result may include at least one of the following: a cell edge map, a cell region map, and a cell nucleus region map.

[0126] For example, in an embodiment of the present disclosure, the target cell nuclear image may be a cell nuclear image of at least one cell segmented from the target cell image. The cell nuclear images included in the target cell nuclear image may be non-adhesive to each other, that is, the cell nuclear images included in the target cell nuclear image may be independent of each other.

[0127] For example, in an embodiment of the present disclosure, a cell may have a cell outline image corresponding to the cell. The target cell region outline image may refer to a cell region outline image of a cell region segmented from the target cell image. The cell region may contain cells with an adhesion relationship. The target cell region outline image may contain cell outline images with an adhesion relationship.

[0128] For example, in an embodiment of the present disclosure, the target cell outline image may include a cell outline image of at least one cell. The cell outline images included in the target cell outline image may be non-adhesive to each other, that is, the cell outline images included in the target cell outline image are independent of each other.

[0129] For example, in an embodiment of the present disclosure, a target image segmentation result may refer to an image segmentation result in which the segmentation accuracy of a target cell image is greater than a predetermined segmentation accuracy threshold. The target image segmentation result may refer to the segmentation results of cells and cell nuclei included in the cell image. The cell segmentation result may be determined based on a target cell outline map. The cell nucleus segmentation result may be determined based on a target cell nucleus map.

[0130] For example, in an embodiment of the present disclosure, a target cell image corresponding to at least one cell can be obtained. An initial image segmentation result can be obtained based on the target cell image corresponding to the at least one cell using an image segmentation method. The image segmentation method can include at least one of the following: a threshold segmentation method, a region segmentation method, and a clustering segmentation method. The region segmentation method can include at least one of the following: a region growing-based segmentation method and a region splitting and merging-based segmentation method.

[0131] For example, in an embodiment of the present disclosure, the initial image segmentation result and the target cell image can be processed to obtain a target cell region map. The target cell region map and the initial image segmentation result can be processed to obtain a target cell nucleus map. The target cell nucleus map, the target cell region map, and the target cell image can be processed to obtain a target cell region outline map. The target cell region outline map, the target cell nucleus map, and the target cell image can be used to obtain a target cell outline map. The target cell outline map and the target cell nucleus map are used as the target image segmentation result.

[0132] For example, in an embodiment of the present disclosure, the target cell region map, the target cell nucleus map, the target cell region contour map, and the target cell contour map may be binary images.

[0133] For example, in an embodiment of the present disclosure, the operation of segmenting a cell image to obtain a target image segmentation result can obtain an image segmentation result without spending manpower on data labeling, and the above operation is independent of the environment in which the cell image is taken and has strong robustness.

[0134] For example, in another embodiment of the present disclosure, the above-mentioned cell image processing method may further include the following operation: processing the original cell image using an image preprocessing method to obtain a target cell image.

[0135] For example, in an embodiment of the present disclosure, an image preprocessing method can be used to reduce irrelevant information in the original cell image and restore useful information. The image preprocessing method may include at least one of the following: an image enhancement method and a noise filtering method. The image enhancement method may include at least one of the following: an illumination compensation method, an edge enhancement method, a contour enhancement method, a texture enhancement method, a target area enhancement method, and a contrast enhancement method. The noise filtering method may include at least one of the following: a Gaussian filtering method, an adaptive filtering denoising method, a median filtering denoising method, a mean filtering denoising method, and an image denoising method based on wavelet transform.

[0136] For example, the original cell image can be processed using a Gaussian filter method to obtain a target cell image, thereby reducing the noise of the target cell image. Alternatively, the original cell image can be processed using an illumination compensation method to obtain a target cell image, thereby making the edge area of ​​the target cell image uniformly illuminated.

[0137] For example, in an embodiment of the present disclosure, an image preprocessing method is used to process an original cell image to obtain a target cell image, thereby improving the quality of the target cell image and facilitating subsequent operations.

[0138] For example, in an embodiment of the present disclosure, the cell image processing method may further include the following operation: obtaining an initial image segmentation result based on the target cell image based on a threshold segmentation method. The initial image segmentation result may include a cell edge map, a cell region map, and a cell nucleus region map.

[0139] For example, in an embodiment of the present disclosure, a threshold segmentation method may refer to a method for segmenting an image into a background area and at least one foreground area based on a grayscale threshold. For example, the threshold segmentation method may include one of the following: a threshold segmentation method based on a global threshold, a threshold segmentation method based on a local threshold, and a threshold segmentation method based on a dynamic threshold. The threshold segmentation method based on a global threshold may include at least one of the following: a fixed threshold segmentation method, a histogram bimodal method, an iterative threshold segmentation method, a maximum inter-class variance method, a gray level co-occurrence matrix method, and a polynomial fitting method.

[0140] For example, in an embodiment of the present disclosure, a cell edge map may refer to an image of an area outside the cell edge in a target cell image where the grayscale is greater than or equal to a predetermined grayscale threshold. A cell region map may refer to an image of a cell region in a target cell image. A cell nucleus region map may refer to an image of a cell nucleus region in a target cell image.

[0141] For example, in an embodiment of the present disclosure, a predetermined threshold can be determined based on the grayscale value of the target cell image. Image segmentation is performed on the target cell image based on the predetermined threshold and the grayscale value of the target cell image to obtain an initial image segmentation result. The predetermined threshold can be used to divide the target cell image into a background region and at least one foreground region.

[0142] For example, in an embodiment of the present disclosure, by performing image segmentation on the target cell image using a threshold-based segmentation method, an initial image segmentation result is obtained, thereby achieving preliminary segmentation of the target cell image.

[0143] For example, in an embodiment of the present disclosure, the threshold segmentation method may include a maximum between-class variance method (ie, OTSU).

[0144] For example, in an embodiment of the present disclosure, obtaining an initial image segmentation result based on a target cell image based on a threshold segmentation method may include the following operations: obtaining a first grayscale threshold, a second grayscale threshold, and a third grayscale threshold based on the target cell image. The second grayscale threshold is greater than the third grayscale threshold and less than the first grayscale threshold. Binarizing the target cell image according to the first grayscale threshold to obtain a cell edge map. Binarizing the target cell image according to the second grayscale threshold to obtain a cell region map. Binarizing the target cell image according to the third grayscale threshold to obtain a cell nucleus region map.

[0145] For example, in an embodiment of the present disclosure, the maximum inter-class variance method divides an image into a foreground region and a background region based on the grayscale characteristics of the image. At least one inter-class variance between the foreground region and the background region may be determined, a maximum inter-class variance may be determined from the at least one inter-class variance, and a threshold corresponding to the maximum inter-class variance may be determined as the grayscale threshold.

[0146] For example, in an embodiment of the present disclosure, the target cell image can be threshold segmented within a first predetermined grayscale range based on the maximum inter-class variance method to obtain a first grayscale threshold. The target cell image can be threshold segmented within a second predetermined grayscale range based on the maximum inter-class variance method to obtain a second grayscale threshold. The second predetermined grayscale range can be determined based on the first grayscale threshold. The target cell image can be threshold segmented within a third predetermined grayscale range based on the maximum inter-class variance method to obtain a third grayscale threshold. The third predetermined grayscale range can be determined based on the second grayscale threshold. The first predetermined grayscale range can be configured according to actual business needs and is not limited here.

[0147] For example, the first predetermined grayscale range may be a grayscale range greater than or equal to 0 and less than or equal to 255. The second predetermined grayscale range may be a grayscale range greater than or equal to 0 and less than or equal to the first grayscale threshold. The third predetermined grayscale range may be a grayscale range greater than or equal to 0 and less than or equal to the second grayscale threshold.

[0148] For example, the target cell image can be threshold segmented within a first predetermined range based on the maximum inter-class variance method to determine a first grayscale threshold. The target cell image can be binarized according to the first grayscale threshold to obtain a cell edge map. For example, the pixel value of a pixel in the target cell image whose pixel value is greater than the first grayscale threshold can be set to a second predetermined pixel value. The pixel value of a pixel in the target cell image whose pixel value is less than or equal to the first grayscale threshold can be set to a first predetermined pixel value. The first predetermined pixel value and the second predetermined pixel value can be configured according to actual business needs and are not limited here. For example, the first predetermined pixel value can be 0. The second predetermined pixel value can be 255.

[0149] For example, in an embodiment of the present disclosure, the target cell image can be threshold segmented within a second predetermined range based on the maximum inter-class variance method to determine a second grayscale threshold. The target cell image can be binarized according to the second grayscale threshold to obtain a cell region map. For example, the pixel values ​​of pixels in the target cell image whose pixel values ​​are less than the second grayscale threshold can be set to the second predetermined pixel value. The pixel values ​​of pixels in the target cell image whose pixel values ​​are greater than or equal to the second grayscale threshold can be set to the first predetermined pixel value.

[0150] For example, in an embodiment of the present disclosure, the target cell image can be threshold segmented within a third predetermined range based on the maximum inter-class variance method to determine a third grayscale threshold. The target cell image can be binarized according to the third grayscale threshold to obtain a cell region map. For example, the pixel values ​​of pixels in the target cell image whose pixel values ​​are less than the third grayscale threshold can be set to the second predetermined pixel value. The pixel values ​​of pixels in the target cell image whose pixel values ​​are greater than or equal to the third grayscale threshold can be set to the first predetermined pixel value.

[0151] For example, in an embodiment of the present disclosure, the initial image segmentation result includes a cell edge map, a cell region map, and a cell nucleus region map.

[0152] For example, in an embodiment of the present disclosure, obtaining a target cell nucleus map based on an initial image segmentation result of a target cell image may include the following operations: obtaining an intermediate cell region map based on a cell region map and a cell edge map; obtaining at least one connected domain cell region map based on the intermediate cell region map and a predetermined foreground region segmentation strategy; and obtaining a target cell nucleus map based on the cell nucleus region map and the at least one connected domain cell region map.

[0153] For example, in an embodiment of the present disclosure, an intermediate cell region map may refer to a cell region map in which the holes inside the cell region and the depressions at the cell edges are filled and the cell edge contours are relatively clear. A predetermined foreground region division strategy may refer to a strategy for how to divide the intermediate cell region map into foreground regions. For example, a predetermined foreground region division strategy may refer to a strategy for dividing the intermediate cell region map into foreground regions based on the area of ​​the connected domain region and at least one predetermined area threshold to obtain at least one connected domain cell region map. The connected domain cell region map may be extracted from the intermediate cell region map.

[0154] For example, in an embodiment of the present disclosure, at least one predetermined area range can be determined according to a predetermined foreground area division strategy. At least one connected domain cell region map is obtained based on the connected domain area in the intermediate cell region map and at least one predetermined area range. Each predetermined area range can have a connected domain cell region map corresponding to the predetermined area range. For example, the N predetermined area ranges can include a predetermined area range S1, a predetermined area threshold S2, ..., a predetermined area threshold S N-1 and the predetermined area threshold S N . N can be an integer greater than or equal to 1.

[0155] For example, in an embodiment of the present disclosure, an intermediate cell region map is obtained based on a cell region map and a cell edge map, and a target cell nucleus map is obtained based on a cell nucleus region map and at least one connected domain cell region map.

[0156] For example, the morphological operation may include at least one of the following: an erosion operation, a dilation operation, and a fill operation. For example, in an embodiment of the present disclosure, based on a morphological method, obtaining an intermediate cell region map according to a cell region map and a cell edge map may include the following operations: processing the cell region map using an expansion operation to obtain a first cell region dilation map. Processing the first cell region dilation map using a fill operation to obtain a cell region dilation-filled map. Processing the cell region dilation-filled map using an erosion operation to obtain a cell region fill map. Obtaining an intermediate cell region map based on the cell region fill map and the cell edge map.

[0157] For example, in an embodiment of the present disclosure, an expansion operation can be used to obtain a first cell area expansion map based on the first structure element and the cell area map. The first structure element can be configured according to actual business needs, and the embodiment of the present disclosure does not limit this. For example, the first structure element can be a first matrix of M*M. M can be an integer greater than 1. The element values ​​of the elements of the first matrix can be set according to actual needs, and the embodiment of the present disclosure does not limit this. For example, M=2. The first matrix can be

[0158] For example, in an embodiment of the present disclosure, a filling operation may be used to fill the hole areas in the cell region expansion filling map to obtain the cell region expansion filling map.

[0159] For example, in the embodiment of the present disclosure, the corrosion operation can be used to obtain the cell area filling map according to the second structure element and the cell area expansion filling map. The second structure element can be set according to actual needs, and the embodiment of the present disclosure does not limit this. For example, the second structure element can be the same as the first structure element. For example, the second matrix can be

[0160] For example, in an embodiment of the present disclosure, after obtaining the cell region filling map, an intermediate cell region map in which the edges of target cells are removed can be obtained based on the cell region filling map and the cell edge map.

[0161] For example, in an embodiment of the present disclosure, by expanding the cell region map to obtain a first cell region expansion map, filling the first cell region expansion map to obtain a cell region expansion filling map, and then corroding the cell region expansion filling map to obtain a cell region filling map, it is possible to effectively fill the smaller holes inside the cell region and the depressions at the edges, eliminate the elongated gaps and narrow discontinuities, and smooth the image contours. This improves the cell integrity after image segmentation. On this basis, by subtracting the cell edge map from the cell region filling map to obtain an intermediate cell region map, the cell edge contours of the intermediate cell region map are made clearer, which is beneficial to improving the accuracy of the target image segmentation results.

[0162] For example, in an embodiment of the present disclosure, obtaining the intermediate cell area map based on the cell area filling map and the cell edge map may include the following operations: based on the cell area filling map and the cell edge map, removing the target cell edges in the cell area filling map to obtain the intermediate cell area map.

[0163] For example, in an embodiment of the present disclosure, the target cell edge may refer to a cell edge that is incorrectly identified in a cell edge map. The incorrectly identified cell edge may include at least one of the following: a cell edge that is not a cell edge but is identified as a cell edge, and a cell edge that is a cell edge but is not identified as a cell edge.

[0164] For example, in an embodiment of the present disclosure, a subtraction operation can be performed on the cell region fill map and the cell edge map to obtain an intermediate cell region map. For example, the cell region fill map is subtracted from the cell edge map to obtain an intermediate cell region map.

[0165] For example, in an embodiment of the present disclosure, the target cell edges of the incorrectly identified cell edges in the cell area filling map are removed to obtain an intermediate cell area map, so that the cell edge contours of the intermediate cell area map are clearer, which is beneficial to improving the accuracy of the target image segmentation results.

[0166] For example, in an embodiment of the present disclosure, obtaining a target cell nucleus map based on a cell nucleus region map and at least one connected domain cell region map may include the following operations: for a connected domain cell region map in the at least one connected domain cell region map, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the cell nucleus region map and the connected domain cell region map. Obtaining a target cell nucleus map based on the cell nucleus maps corresponding to each of the at least one connected domain cell region maps.

[0167] For example, in an embodiment of the present disclosure, for each connected domain cell region map in at least one connected domain cell region map, a connected domain cell region map and a cell nucleus region map can be processed using a morphological method to obtain a cell nucleus map corresponding to the connected domain cell region map. Thus, a cell nucleus map corresponding to each of the at least one connected domain cell region maps can be obtained. The cell nucleus maps corresponding to each of the at least one connected domain cell region maps can be merged to obtain a target cell nucleus map. For example, the cell nucleus maps corresponding to each of the at least one connected domain cell region maps can be added together to obtain a target cell nucleus map.

[0168] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the cell nucleus region map and the connected domain cell region map may include the following operations: when it is determined that the area of ​​the connected domain cell region map is greater than or equal to a first predetermined area threshold, processing the connected domain cell region map based on a morphological method to obtain an intermediate connected domain cell region map. Obtaining a cell nucleus map corresponding to the connected domain cell region map based on the intermediate connected domain cell region map and the cell nucleus region map. When it is determined that the area of ​​the connected domain cell region map is less than the first predetermined area threshold, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the cell nucleus region map and the connected domain cell region map.

[0169] For example, in an embodiment of the present disclosure, the first predetermined area threshold can be used as a basis for determining whether a morphological operation needs to be performed on the connected domain cell area map. The first predetermined area threshold can be configured according to actual business needs and is not limited here.

[0170] For example, in an embodiment of the present disclosure, for each connected domain cell region map in at least one connected domain cell region map, the area of ​​the connected domain cell region map can be determined. When it is determined that the area of ​​the connected domain cell region map is greater than or equal to the first predetermined area threshold, it can be explained that the degree of cell adhesion in the connected domain cell region map is high. In this case, the corrosion operation can be used to obtain an intermediate connected domain cell region map based on the third structural element corresponding to the connected domain cell region map and the connected domain cell region map. The third structural element can be configured according to actual business needs and is not limited here. For example, the third structural element corresponding to the connected domain cell region map can be determined based on the area of ​​the connected domain cell region map. The larger the area of ​​the connected domain cell region, the larger the size of the third structural element corresponding to the connected domain cell region map. After obtaining the intermediate connected domain cell region map, the cell nucleus map corresponding to the connected domain cell region map can be obtained based on the intermediate connected domain cell region map and the cell nucleus region map.

[0171] For example, in an embodiment of the present disclosure, when it is determined that the area of ​​the connected domain cell region map is less than the first predetermined area threshold, a cell nucleus map corresponding to the connected domain cell region map can be obtained based on the connected domain cell region map and the cell nucleus region map.

[0172] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the intermediate connected domain cell region map and the cell nucleus region map may include the following operations: performing an AND operation on the intermediate connected domain cell region map and the cell nucleus region map to obtain a first intersection cell region map. Obtaining a cell nucleus map corresponding to the connected domain cell region map based on the first intersection cell region map and the intermediate connected domain cell region map.

[0173] For example, in an embodiment of the present disclosure, the first intersection cell region map may refer to an image corresponding to the intersection region between the intermediate connected domain cell region map and the cell nucleus region map. Since the intermediate connected domain cell region map may be obtained by performing an erosion operation on the connected domain cell region map corresponding to the intermediate connected domain cell region map, there is a situation in the intermediate connected domain cell region map where a cell is divided into at least two cells. Since the cell nucleus region map is an image including the cell nucleus, the region intersecting with the cell nucleus region map can be retained by performing an AND operation on the intermediate connected domain cell region map and the cell nucleus region map. Since the cell nucleus region map is an image including the cell nucleus, the probability that the first intersection cell region map includes the cell nucleus can be increased, thereby reducing the noise interference on the first intersection cell region map.

[0174] For example, in an embodiment of the present disclosure, after obtaining the first intersection cell region map, a first candidate cell nucleus map corresponding to the connected domain cell region map can be obtained based on the first intersection cell region map and the intermediate connected domain cell region map. Based on the first candidate cell nucleus map corresponding to the connected domain cell region map, a cell nucleus map corresponding to the connected domain cell region map can be obtained. For example, based on the first intersection cell region map and the intermediate connected domain cell region map, the connected domain in the intermediate connected domain cell region map that is located in the first intersection cell region map can be retained, and other regions can be removed to obtain the first candidate cell nucleus map corresponding to the connected domain cell region map.

[0175] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to a connected domain cell region map based on a first intersection cell region map and an intermediate connected domain cell region map may include the following operations: generating a first template image. The size of the first template image is equal to the size of the intermediate connected domain cell region map. The pixel values ​​of pixels in the first template image are first predetermined pixel values. Based on the first template image, the first intersection cell region map, and the intermediate connected domain cell region map, obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map. Based on the first candidate cell nucleus map corresponding to the connected domain cell region map, obtaining a cell nucleus map corresponding to the connected domain cell region map.

[0176] For example, in an embodiment of the present disclosure, the first model image may be an image having a pixel value of a first predetermined pixel value and a size equal to that of the intermediate connected domain cell region map. The first template image may be processed based on the first intersection cell region map and the intermediate connected domain cell region map to obtain a first candidate cell nucleus map corresponding to the connected domain cell region map.

[0177] For example, in an embodiment of the present disclosure, after obtaining a first candidate cell nucleus map, the first target connected domain can be deleted from the first candidate cell nucleus map to obtain a first candidate cell nucleus map after deleting the first target connected domain. The first candidate cell nucleus map after deleting the first target connected domain is determined as the cell nucleus map. The first target connected domain may refer to a connected domain whose area satisfies a first predetermined area condition. For example, the first target connected domain may refer to a connected domain whose area is less than a second predetermined area threshold.

[0178] For example, in an embodiment of the present disclosure, obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map based on the first template image, the first intersection cell region map, and the intermediate connected domain cell region map may include the following operations: determining a first contour set corresponding to the connected domain in the first intersection cell region map. Traversing each first predetermined pixel included in the first contour set, modifying the pixel value of the pixel with the same label as the first predetermined pixel in the first template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, and obtaining the first candidate cell nucleus map corresponding to the connected domain cell region map. The label of the first predetermined pixel in the first template image is determined based on the label of the first predetermined pixel in the intermediate connected domain cell region map.

[0179] For example, in an embodiment of the present disclosure, a label can be used to characterize a connected domain. Each connected domain has a label corresponding to the connected domain. The connected domain can have a one-to-one correspondence with the label. A pixel of the first template image has a label corresponding to the pixel. The labels of pixels belonging to the same connected domain in the first template image are the same. A pixel of the intermediate connected domain cell region map has a label corresponding to the pixel. The labels of pixels belonging to the same connected domain in the intermediate connected domain cell region map are the same. The label can be associated with the position of the pixel.

[0180] For example, in an embodiment of the present disclosure, the contours corresponding to all connected domains included in the first intersection cell area map can be determined to obtain a first contour set. The first contour set may include at least one first contour. Each first contour may have a plurality of pixels corresponding to the first contour. Each first contour may have a first predetermined pixel corresponding to the first contour. The first predetermined pixel corresponding to the first contour may be determined from the plurality of pixels corresponding to the first contour. The first predetermined pixel may be selected according to actual business needs and is not limited here. For example, the first predetermined pixel corresponding to the first contour may be the first pixel among the plurality of pixels corresponding to the first contour. Alternatively, the first predetermined pixel corresponding to the first contour may be a pixel located in the center area among the plurality of pixels corresponding to the first contour.

[0181] For example, in an embodiment of the present disclosure, for each first contour in the first contour set, a first predetermined pixel corresponding to the first contour is determined. Based on the position of the first predetermined pixel and the association between the label and the position of the pixel, the label corresponding to the first predetermined pixel of the first contour is determined. Based on the label corresponding to the first predetermined pixel of the first contour and the association between the label and the position of the pixel, the positions of the plurality of pixels in the first template image that have the same label as the first predetermined pixel of the first contour are determined. Based on the positions of the plurality of pixels that have the same label as the first predetermined pixel of the first contour, the pixel values ​​of the plurality of pixels that have the same label as the first predetermined pixel of the first contour are modified from the first predetermined pixel value to the second predetermined pixel value. Based on the above method, the first contour set is traversed to obtain the first candidate cell nucleus map corresponding to the connected domain cell region map.

[0182] For example, in an embodiment of the present disclosure, since the cell nucleus region map is an image including the cell nucleus, if the connected domain of the intermediate connected domain cell region map is a valid connected domain, the intermediate connected domain cell region map needs to include the connected domain in the cell nucleus region map. Therefore, by performing an AND operation on the intermediate connected domain cell region map and the cell nucleus region map, a first intersection cell region map that intersects with the cell nucleus region map is obtained. By traversing each first predetermined pixel included in the first contour set in the first intersection cell region map, the pixel value of the pixel with the same label as the first predetermined pixel in the first template image is modified from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, the probability that the first candidate cell nucleus map includes the cell nucleus can be increased, and the noise interference on the first candidate cell nucleus map can be reduced.

[0183] For example, in an embodiment of the present disclosure, obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map based on the first template image, the first intersection cell region map, and the intermediate connected domain cell region map may include the following operations: traversing each second predetermined pixel in the first template image, and when determining that the pixel value of the pixel corresponding to the second predetermined pixel in the first intersection cell region map is the expected pixel value, modifying the pixel value of the pixel with the same label as the second predetermined pixel in the first template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, thereby obtaining the first candidate cell nucleus map corresponding to the connected domain cell region map. The label of the second predetermined pixel in the first template image is determined based on the label of the second predetermined pixel in the intermediate connected domain cell region map.

[0184] For example, in the embodiments of the present disclosure, the expected pixel value can be configured according to actual business needs and is not limited here. For example, the expected pixel value can be a second predetermined pixel value. The second predetermined pixel can be any pixel in the first template image.

[0185] For example, in an embodiment of the present disclosure, for each second predetermined pixel in the first template image, it is determined whether the pixel value of the pixel corresponding to the second predetermined pixel in the first intersection cell area map is the expected pixel value. In the case of determining that the pixel value of the pixel corresponding to the second predetermined pixel in the first intersection cell area map is the expected pixel value, the positions of the plurality of pixels in the first template image that have the same label as the second predetermined pixel can be determined based on the label corresponding to the second predetermined pixel and the association between the label and the position of the pixel. Based on the positions of the plurality of pixels in the first template image that have the same label as the second predetermined pixel, the pixel values ​​of the plurality of pixels in the first template image that have the same label as the second predetermined pixel are modified from the first predetermined pixel value to the second predetermined pixel value.

[0186] For example, in an embodiment of the present disclosure, by traversing each second predetermined pixel in the first template image, if it is determined that the pixel value of the pixel corresponding to the second predetermined pixel in the first intersection cell region image is the expected pixel value, the pixel value of the pixel with the same label as the second predetermined pixel in the first template image is modified from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, the probability that the first candidate cell nucleus image includes a cell nucleus can be increased, and the noise interference of the first candidate cell nucleus image can be reduced. In addition, the cell nucleus morphology can be restored to make it close to the actual cell nucleus morphology.

[0187] For example, in an embodiment of the present disclosure, the first candidate cell nucleus map may include at least one first candidate connected domain. The cell nucleus map may include at least one connected domain.

[0188] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the first candidate cell nucleus map corresponding to the connected domain cell region map may include the following operations: for each first candidate connected domain in the first candidate cell nucleus map corresponding to the connected domain cell region map, when it is determined that the area of ​​the first candidate connected domain is greater than or equal to a second predetermined area threshold, determining the first candidate connected domain as a connected domain in the cell nucleus map corresponding to the connected domain cell region map.

[0189] For example, in the embodiment of the present disclosure, the second predetermined area threshold can be configured according to actual business needs and is not limited here. For example, the second predetermined area threshold can be 20.

[0190] For example, in an embodiment of the present disclosure, a connected domain in a cell nucleus map is a connected domain whose area is greater than or equal to a second predetermined area threshold. Therefore, connected domains with smaller areas in the cell nucleus map are effectively reduced. Connected domains with smaller areas may interfere with image segmentation. Therefore, the first candidate connected domain whose area is greater than or equal to the second predetermined area threshold is used as the connected domain in the cell nucleus map, which can improve the accuracy of the image segmentation result.

[0191] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the cell nucleus region map and the connected domain cell region map may include the following operations: performing an AND operation on the connected domain cell region map and the cell nucleus region map to obtain a second intersection cell region map; and obtaining a cell nucleus map corresponding to the connected domain cell region map based on the second intersection cell region map and the connected domain cell region map.

[0192] For example, in an embodiment of the present disclosure, the second intersection cell region map may refer to an image corresponding to the intersection region between the connected domain cell region map and the cell nucleus region map.

[0193] For example, in an embodiment of the present disclosure, after obtaining the second intersection cell region map, a second candidate cell nucleus map corresponding to the connected domain cell region map can be obtained based on the second intersection cell region map and the connected domain cell region map. Based on the second candidate cell nucleus map corresponding to the connected domain cell region map, a cell nucleus map corresponding to the connected domain cell region map is obtained. For example, based on the second intersection cell region map and the connected domain cell region map, the connected domain in the connected domain cell region map that is located in the second intersection cell region map can be retained, and other regions can be removed to obtain a second candidate cell nucleus map corresponding to the connected domain cell region map.

[0194] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the second intersection cell region map and the connected domain cell region map may include the following operations: generating a second template image. The size of the second template image is equal to the size of the connected domain cell region map. The pixel values ​​of the pixels of the second template image are first predetermined pixel values. Based on the second template image, the second intersection cell region map, and the connected domain cell region map, obtaining a second candidate cell nucleus map corresponding to the connected domain cell region map. Based on the second candidate cell nucleus map corresponding to the connected domain cell region map, obtaining a cell nucleus map corresponding to the connected domain cell region map.

[0195] For example, in an embodiment of the present disclosure, the second model image may be an image having a pixel value of a first predetermined pixel value and a size equal to that of the connected domain cell region map. The second template image may be processed based on the second intersection cell region map and the connected domain cell region map to obtain a second candidate cell nucleus map corresponding to the connected domain cell region map.

[0196] For example, in an embodiment of the present disclosure, after obtaining the second candidate cell nucleus map, the second target connected domain can be deleted from the second candidate cell nucleus map to obtain a second candidate cell nucleus map after deleting the second target connected domain. The second candidate cell nucleus map after deleting the second target connected domain is determined as the cell nucleus map. The second target connected domain may refer to a connected domain whose area satisfies a second predetermined area condition. For example, the second target connected domain may refer to a connected domain whose area is less than a third predetermined area threshold.

[0197] For example, in an embodiment of the present disclosure, obtaining a second candidate cell nucleus map corresponding to the connected domain cell region map based on the second template image, the second intersection cell region map, and the connected domain cell region map may include the following operations: determining a second contour set corresponding to the connected domain in the second intersection cell region map. Traversing each third predetermined pixel included in the second contour set, modifying the pixel value of the pixel with the same label as the third predetermined pixel in the second template image from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, thereby obtaining a second candidate cell nucleus map corresponding to the connected domain cell region map. The label of the third predetermined pixel in the second template image is determined based on the label of the third predetermined pixel in the connected domain cell region map.

[0198] For example, in an embodiment of the present disclosure, a label can be used to characterize a connected domain. Each connected domain has a label corresponding to the connected domain. The connected domain can have a one-to-one correspondence with the label. A pixel of the second template image has a label corresponding to the pixel. The labels of pixels belonging to the same connected domain in the second template image are the same. A pixel of a connected domain cell region map has a label corresponding to the pixel. The labels of pixels belonging to the same connected domain in a connected domain cell region map are the same. The label can be associated with the position of the pixel.

[0199] For example, in an embodiment of the present disclosure, the contours corresponding to all connected domains included in the second intersection cell area map can be determined to obtain a second contour set. The second contour set may include at least one second contour. Each second contour may have a plurality of pixels corresponding to the second contour. Each second contour may have a third predetermined pixel corresponding to the second contour. The third predetermined pixel corresponding to the second contour may be determined from the plurality of pixels corresponding to the second contour. The third predetermined pixel may be selected according to actual business needs and is not limited here. For example, the third predetermined pixel corresponding to the second contour may be the first pixel among the plurality of pixels corresponding to the second contour. Alternatively, the third predetermined pixel corresponding to the second contour may be a pixel located in the center area among the plurality of pixels corresponding to the second contour.

[0200] For example, in an embodiment of the present disclosure, for each second contour in the second contour set, a third predetermined pixel corresponding to the second contour is determined. Based on the position of the third predetermined pixel and the association between the label and the position of the pixel, the label corresponding to the third predetermined pixel of the second contour is determined. Based on the label corresponding to the third predetermined pixel of the second contour and the association between the label and the position of the pixel, the positions of the plurality of pixels having the same label as the third predetermined pixel of the second contour in the second template image are determined. Based on the positions of the plurality of pixels having the same label as the third predetermined pixel of the second contour, the pixel values ​​of the plurality of pixels having the same label as the third predetermined pixel of the second contour are modified from the first predetermined pixel value to the second predetermined pixel value. Based on the above method, the second contour set is traversed to obtain a second candidate cell nucleus map corresponding to the connected domain cell region map.

[0201] For example, in an embodiment of the present disclosure, since the cell nucleus region map is an image including the cell nucleus, if the connected domain of the connected domain cell region map is a valid connected domain, the connected domain cell region map needs to include the connected domain in the cell nucleus region map. Therefore, by performing an AND operation on the connected domain cell region map and the cell nucleus region map, a second intersection cell region map that intersects with the cell nucleus region map is obtained. By traversing each third predetermined pixel included in the second contour set in the second intersection cell region map, the pixel value of the pixel with the same label as the third predetermined pixel in the second template image is modified from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, the probability that the second candidate cell nucleus map includes the cell nucleus can be increased, and the noise interference on the second candidate cell nucleus map can be reduced. In addition, the cell nucleus morphology can be restored to be close to the real cell nucleus morphology.

[0202] For example, in an embodiment of the present disclosure, obtaining a second candidate cell nucleus map corresponding to the connected domain cell region map based on the second template image, the second intersection cell region map, and the connected domain cell region map may include the following operations: traversing each fourth predetermined pixel in the second template image, and when determining that the pixel value of the pixel corresponding to the fourth predetermined pixel in the second intersection cell region map is the expected pixel value, modifying the pixel value of the pixel with the same label as the fourth predetermined pixel in the second template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, thereby obtaining the second candidate cell nucleus map corresponding to the connected domain cell region map. The label of the fourth predetermined pixel in the second template image is determined based on the label of the fourth predetermined pixel in the connected domain cell region map.

[0203] For example, in the embodiments of the present disclosure, the expected pixel value can be configured according to actual business needs and is not limited here. For example, the expected pixel value can be the second predetermined pixel value. The fourth predetermined pixel can be any pixel in the second template image.

[0204] For example, in an embodiment of the present disclosure, for each fourth predetermined pixel in the second template image, it is determined whether the pixel value of the pixel corresponding to the fourth predetermined pixel in the second intersection cell area map is the expected pixel value. In the case of determining that the pixel value of the pixel corresponding to the fourth predetermined pixel in the second intersection cell area map is the expected pixel value, the positions of the plurality of pixels in the second template image that have the same label as the fourth predetermined pixel can be determined based on the label corresponding to the fourth predetermined pixel and the association between the label and the position of the pixel. Based on the positions of the plurality of pixels in the second template image that have the same label as the fourth predetermined pixel, the pixel values ​​of the plurality of pixels in the second template image that have the same label as the fourth predetermined pixel are modified from the first predetermined pixel value to the second predetermined pixel value.

[0205] For example, in an embodiment of the present disclosure, by traversing each fourth predetermined pixel in the second template image, when it is determined that the pixel value of the pixel corresponding to the fourth predetermined pixel in the second intersection cell area map is the expected pixel value, the pixel value of the pixel with the same label as the fourth predetermined pixel in the second template image is modified from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, thereby increasing the probability that the second candidate cell nucleus map includes the cell nucleus and reducing the noise interference on the second candidate cell nucleus map.

[0206] For example, in an embodiment of the present disclosure, the second candidate cell nucleus map may include at least one second candidate connected domain. The cell nucleus map may include at least one connected domain.

[0207] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the second candidate cell nucleus map corresponding to the connected domain cell region map may include the following operations: for each second candidate connected domain in the second candidate cell nucleus map corresponding to the connected domain cell region map, when it is determined that the area of ​​the second candidate connected domain is greater than or equal to a third predetermined area threshold, determining the second candidate connected domain as a connected domain in the cell nucleus map corresponding to the connected domain cell region map.

[0208] For example, in the embodiment of the present disclosure, the third predetermined area threshold can be configured according to actual business needs and is not limited here. For example, the third predetermined area threshold can be 70.

[0209] For example, in an embodiment of the present disclosure, a connected domain in a cell nucleus map is a connected domain whose area is greater than or equal to a third predetermined area threshold. Therefore, connected domains with smaller areas in the cell nucleus map are effectively reduced. Connected domains with smaller areas may interfere with image segmentation. Therefore, the second candidate connected domain whose area is greater than or equal to the third predetermined area threshold is used as a connected domain in the cell nucleus map, which can improve the accuracy of the image segmentation result.

[0210] For example, in an embodiment of the present disclosure, the initial image segmentation result may include a cell edge map and a cell area map.

[0211] For example, in an embodiment of the present disclosure, based on a morphological method, a target cell region contour map is obtained according to a target cell nucleus map and an initial image segmentation result, which may include the following operations: obtaining a primary cell region map according to the target cell nucleus map and the cell region map. Processing the initial cell region map using an expansion operation to obtain a second cell region expansion map. Obtaining a first intermediate cell region contour map according to the second cell region expansion map and the primary cell region map. Obtaining a second intermediate cell region contour map according to the first intermediate cell region contour map and the target cell image. Obtaining a third intermediate cell region contour map according to the second intermediate cell region contour map and the target cell nucleus map. Obtaining a target cell region contour map according to the third intermediate cell region contour map and the cell edge map.

[0212] For example, in an embodiment of the present disclosure, the target cell nucleus map and the cell region map can be added together to obtain a primary cell region map. After obtaining the primary cell region map, an expansion operation can be performed to obtain a second cell region expansion map based on the fourth structure element and the primary cell region map. The fourth structure element can be configured according to actual business needs and is not limited here.

[0213] For example, in an embodiment of the present disclosure, a subtraction operation is performed on the primary cell region map and the second cell region expansion map to obtain a first intermediate cell region outline map. The second cell region expansion map can be used as the background, and the primary cell region map can be used as the foreground.

[0214] For example, in an embodiment of the present disclosure, the second intermediate cell region outline map and the target cell nucleus map can be subjected to an AND operation to obtain the intersection region of the second intermediate cell region outline map and the target cell nucleus map. The connected domain of the intersection region in the second intermediate cell region outline map is retained, and other regions except the intersection region are removed to obtain a third intermediate cell region outline map.

[0215] For example, in an embodiment of the present disclosure, the third intermediate cell region outline map and the cell edge map can be subtracted to obtain the target cell region outline map. For example, the third intermediate cell region outline map can be subtracted from the cell edge map to obtain the target cell region outline map.

[0216] For example, in an embodiment of the present disclosure, obtaining a second intermediate cell region outline map based on a first intermediate cell region outline map and a target cell image may include the following operations: performing watershed processing on the first intermediate cell region outline map based on the target cell image to obtain a fourth intermediate cell region outline map; setting the pixel values ​​of pixels in the background region of the fourth intermediate cell region outline map to a first predetermined pixel value, and setting the pixel values ​​of pixels in other regions of the fourth intermediate cell region outline map except the background region to a second predetermined pixel value, to obtain a second intermediate cell region outline map.

[0217] For example, in an embodiment of the present disclosure, the watershed method may include at least one of the following: a distance transform-based watershed segmentation method, a marker-based watershed segmentation method, and a gradient-based watershed segmentation method. The watershed method can be configured according to actual business needs and is not limited here.

[0218] For example, in an embodiment of the present disclosure, a watershed process can be performed on the first intermediate cell region outline image on the target cell image to obtain a fourth intermediate cell region outline image. The first predetermined pixel value and the second predetermined pixel value can be configured according to actual business needs and are not limited here. For example, the first predetermined pixel value can be 0. The second predetermined pixel value can be 255.

[0219] For example, in an embodiment of the present disclosure, obtaining a third intermediate cell region outline map based on the second intermediate cell region outline map and the target cell nucleus map may include the following operations: generating a third template image. The size of the third template image is equal to the size of the second intermediate cell region outline map. The pixel values ​​of the pixels of the third template image are first predetermined pixel values. The third intermediate cell region outline map is obtained based on the third template image, the second intermediate cell region outline map, and the target cell nucleus map.

[0220] For example, in an embodiment of the present disclosure, the third model image may be an image having a pixel value of a first predetermined pixel value and a size equal to that of the second intermediate cell region contour image.

[0221] For example, in an embodiment of the present disclosure, obtaining a third intermediate cell region contour map based on a third template image, a second intermediate cell region contour map, and a target cell nucleus map may include the following operations: determining a third contour set corresponding to a connected domain in the target cell nucleus map. Traversing each fifth predetermined pixel included in the third contour set, modifying the pixel value of the pixel with the same label as the fifth predetermined pixel in the third template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, thereby obtaining the third intermediate cell region contour map. The label of the fifth predetermined pixel in the third template image is determined based on the label of the fifth predetermined pixel in the second intermediate cell region contour map.

[0222] For example, in an embodiment of the present disclosure, a label can be used to characterize a connected domain. Each connected domain has a label corresponding to the connected domain. The connected domain can have a one-to-one correspondence with the label. The pixel of the third template image has a label corresponding to the pixel. The labels of the pixels belonging to the same connected domain in the third template image are the same. The pixel of the second intermediate cell region outline image has a label corresponding to the pixel. The labels of the pixels belonging to the same connected domain in the second intermediate cell region outline image are the same. The label can be associated with the position of the pixel.

[0223] For example, in an embodiment of the present disclosure, the contours corresponding to all connected domains included in the target cell nucleus map can be determined to obtain a third contour set. The third contour set may include at least one third contour. Each third contour may have a plurality of pixels corresponding to the third contour. Each third contour may have a fifth predetermined pixel corresponding to the third contour. The fifth predetermined pixel corresponding to the third contour may be determined from the plurality of pixels corresponding to the third contour. The fifth predetermined pixel may be selected according to actual business needs and is not limited here. For example, the fifth predetermined pixel corresponding to the third contour may be the first pixel among the plurality of pixels corresponding to the third contour. Alternatively, the fifth predetermined pixel corresponding to the third contour may be a pixel located in the center area among the plurality of pixels corresponding to the third contour.

[0224] For example, in an embodiment of the present disclosure, for each third contour in the third contour set, the fifth predetermined pixel corresponding to the third contour is determined. Based on the position of the fifth predetermined pixel and the association between the label and the position of the pixel, the label corresponding to the fifth predetermined pixel of the third contour is determined. Based on the label corresponding to the fifth predetermined pixel of the third contour and the association between the label and the position of the pixel, the positions of the plurality of pixels in the third template image that have the same label as the fifth predetermined pixel of the third contour are determined. Based on the positions of the plurality of pixels that have the same label as the fifth predetermined pixel of the third contour, the pixel values ​​of the plurality of pixels that have the same label as the fifth predetermined pixel of the third contour are modified from the first predetermined pixel value to the second predetermined pixel value. Based on the above method, the second contour set is traversed to obtain the third intermediate cell area contour map.

[0225] For example, in an embodiment of the present disclosure, since the target cell nucleus map is an image including the cell nucleus, by traversing each fifth predetermined pixel included in the third contour set corresponding to the connected domain in the target cell nucleus map, the pixel value of the pixel with the same label as the fifth predetermined pixel in the third template image is modified from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, thereby increasing the probability that the third intermediate cell area contour map includes the cell nucleus and reducing the noise interference on the third intermediate cell area contour map.

[0226] For example, in an embodiment of the present disclosure, obtaining a third intermediate cell area contour map based on a third template image, a second intermediate cell area contour map and a target cell nucleus map can include the following operations: traversing each sixth predetermined pixel in the third template image, and when determining that the pixel value of the pixel corresponding to the sixth predetermined pixel in the target cell nucleus map is the expected pixel value, modifying the pixel value of the pixel with the same label as the sixth predetermined pixel in the third template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed to obtain the third intermediate cell area contour map.

[0227] For example, in the embodiments of the present disclosure, the expected pixel value can be configured according to actual business needs and is not limited here. For example, the expected pixel value can be the second predetermined pixel value. The sixth predetermined pixel can be any pixel in the third template image.

[0228] For example, in an embodiment of the present disclosure, for each sixth predetermined pixel in the third template image, it is determined whether the pixel value of the pixel corresponding to the sixth predetermined pixel in the target cell nuclear image is the expected pixel value. In the case of determining that the pixel value of the pixel corresponding to the sixth predetermined pixel in the target cell nuclear image is the expected pixel value, the positions of the plurality of pixels having the same label as the sixth predetermined pixel in the third template image can be determined based on the label corresponding to the sixth predetermined pixel and the association between the label and the position of the pixel. Based on the positions of the plurality of pixels having the same label as the sixth predetermined pixel in the third template image, the pixel values ​​of the plurality of pixels having the same label as the sixth predetermined pixel in the third template image are modified from the first predetermined pixel value to the second predetermined pixel value.

[0229] For example, in an embodiment of the present disclosure, by traversing each sixth predetermined pixel in the third template image, when it is determined that the pixel value of the pixel corresponding to the sixth predetermined pixel in the target cell nucleus image is the expected pixel value, the pixel value of the pixel with the same label as the sixth predetermined pixel in the third template image is modified from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, thereby increasing the probability that the third intermediate cell area contour map includes the cell nucleus and reducing the noise interference on the third intermediate cell area contour map.

[0230] For example, in an embodiment of the present disclosure, obtaining a target cell outline map based on a target cell region outline map and a target cell nucleus map based on a morphological method may include the following operations: subtracting the target cell region outline map from the target cell nucleus map to obtain an intermediate cell outline map; and performing watershed processing on the intermediate cell outline map based on the target cell image to obtain the target cell outline map.

[0231] For example, in an embodiment of the present disclosure, the target cell region outline image can be used as the background, the target cell nucleus image can be used as the foreground, and the intermediate cell outline image can be subjected to watershed processing on the target cell image to obtain the target cell outline image.

[0232] For example, in an embodiment of the present disclosure, the cell image processing method may further include the following operations: determining cell statistical information based on the target image segmentation result; and determining a cell quality assessment result based on the cell statistical information.

[0233] For example, in an embodiment of the present disclosure, the cell statistical information includes at least one of the following: cell number, cell area, cell perimeter, total cell area, cell major axis length, and cell minor axis length.

[0234] For example, in an embodiment of the present disclosure, a cell statistical analysis routine can be called to process the target image segmentation results using the cell statistical analysis routine to obtain cell statistical information. For example, the cell statistical analysis routine can include at least one of the following: a perimeter statistics routine, an area statistics routine, a cell number statistics routine, and a cell major and minor axis statistics routine. For example, the perimeter statistics routine can be cv2.arcLength(). The area statistics routine can be cv2.contourArea().

[0235] For example, in embodiments of the present disclosure, a cell number counting routine can be used to determine the cell number based on a target cell nucleus image. A cell area counting routine can be used to obtain the cell area based on a target cell outline image. A cell perimeter counting routine can be used to obtain the cell perimeter based on a target cell outline image. A cell major axis and cell minor axis counting routine can be used to obtain the cell major axis and cell minor axis length based on a target cell outline image. The total cell area can be obtained based on the cell number and cell area.

[0236] For example, in an embodiment of the present disclosure, the quality assessment result of the cell may be determined based on the cell statistical information.

[0237] For example, in the disclosed embodiments, cell images are segmented to obtain target image segmentation results, which are then used to assess cell quality. This quality assessment can help mitigate errors caused by human factors, further standardize the cell culture process, and effectively reduce production costs. In actual industrial production, this method can rapidly analyze continuous target cell images with statistical accuracy exceeding 90%.

[0238] The cell image processing method provided by the embodiments of the present disclosure will be further described below with reference to FIG. 4A to FIG. 4B and FIG. 5A to FIG. 5K in combination with specific embodiments.

[0239] For example, FIG4A schematically illustrates a schematic diagram of the principles of a cell image processing process according to an embodiment of the present disclosure. As shown in FIG4A , in step 300A, an initial image segmentation result can be obtained based on a target cell image 301 using a threshold segmentation method. The initial image segmentation result can include a cell edge map 302, a cell region map 303, and a cell nucleus region map 304.

[0240] Cell region map 303 is processed using a dilation operation to obtain a first cell region dilation map 305. Cell region dilation map 305 is processed using a fill operation to obtain a cell region dilation fill map 306. Cell region dilation fill map 306 is processed using an erosion operation to obtain a cell region fill map 307. Based on cell region fill map 307 and cell edge map 302, an intermediate cell region map 308 is obtained.

[0241] Based on the intermediate cell region map 308 and the predetermined foreground region partitioning strategy, at least one connected domain cell region map 309 is obtained. For the connected domain cell region map 309 in the at least one connected domain cell region map 309, a cell nucleus map corresponding to the connected domain cell region map 309 is obtained based on the cell nucleus region map 304 and the connected domain cell region map 309. Based on the cell region maps corresponding to each of the at least one connected domain cell region maps 309, a target cell nucleus map 310 is obtained.

[0242] A primary cell region map 311 can be obtained based on the target cell nucleus map 310 and the cell region map 303. The initial cell region map 311 is processed using a dilation operation to obtain a second cell region dilation map 312. A first intermediate cell region outline map 313 is obtained based on the second cell region dilation map 312 and the primary cell region map 311. A second intermediate cell region outline map 314 is obtained based on the first intermediate cell region outline map 313 and the target cell image 301. A third intermediate cell region outline map 315 is obtained based on the second intermediate cell region outline map 314 and the target cell nucleus map 310. A target cell region outline map 316 is obtained based on the third intermediate cell region outline map 315 and the cell edge map 302.

[0243] The target cell region outline image 316 is subtracted from the target cell nucleus image 310 to obtain an intermediate cell outline image 317. Based on the target cell image 301, a watershed process is performed on the intermediate cell outline image 317 to obtain a target cell outline image 318.

[0244] The following further describes the cell nucleus map corresponding to the connected domain cell region map 309 obtained based on the cell nucleus region map 304 and the connected domain cell region map 309 in FIG. 4A in conjunction with FIG. 4B .

[0245] For example, FIG4B schematically shows a flow chart of obtaining a cell nucleus map corresponding to a connected domain cell region map based on a cell nucleus region map and a connected domain cell region map according to an embodiment of the present disclosure.

[0246] For example, as shown in FIG4B , the method 300B includes operations S301 - S325 .

[0247] In operation S301 , a connected domain cell area map is obtained.

[0248] In operation S302 , is the area of ​​the connected domain cell region map greater than or equal to a first predetermined area threshold? If so, operations S302 to S314 are performed; if not, operations S315 to S325 are performed.

[0249] In operation S303 , the connected domain cell region map is processed based on a morphological method to obtain an intermediate connected domain cell region map.

[0250] In operation S304 , an AND operation is performed on the intermediate connected domain cell region map and the cell nucleus region map to obtain a first intersection cell region map.

[0251] In operation S305, a first template image is generated. The size of the first template image is equal to the size of the intermediate connected domain cell region map. The pixel values ​​of the pixels of the first template image are first predetermined pixel values.

[0252] In operation S306 , a first contour set corresponding to the connected domain in the first intersection cell region map is determined.

[0253] In operation S307 , a current first predetermined pixel in the first contour set is determined.

[0254] In operation S308 , the pixel value of the pixel in the first template image that has the same label as the current first predetermined pixel is modified from the first predetermined pixel value to a second predetermined pixel value.

[0255] In operation S309 , has the first predetermined pixel in the first contour set been completely traversed? If so, operations S310 and S312 to S314 are performed; if not, operation S311 is performed.

[0256] In operation S310 , a first candidate cell nucleus map corresponding to the connected domain cell region map is obtained.

[0257] In operation S311 , a new current first predetermined pixel in the first contour set is determined, and the process returns to operation S309 .

[0258] In operation S312, is the area of ​​the first candidate connected domain in the first candidate cell nucleus map greater than or equal to a second predetermined area threshold? If so, operation S313 is performed; if not, operation S314 is performed.

[0259] In operation S313 , the first candidate connected domain is determined as a connected domain in the cell nucleus map corresponding to the connected domain cell region map.

[0260] In operation S314 , the first candidate connected component is deleted.

[0261] In operation S315 , an AND operation is performed on the connected domain cell region map and the cell nucleus region map to obtain a second intersection cell region map.

[0262] In operation S316, a second template image is generated. The size of the second template image is equal to the size of the connected domain cell region map. The pixel values ​​of the pixels of the second template image are the first predetermined pixel values.

[0263] In operation S317 , a second contour set corresponding to the connected domain in the second intersection cell region map is determined.

[0264] In operation S318, a current third predetermined pixel in the second contour set is determined.

[0265] In operation S319 , the pixel value of the pixel in the second template image that has the same label as the current third predetermined pixel is modified from the first predetermined pixel value to the second predetermined pixel value.

[0266] In operation S320 , has the third predetermined pixel in the second contour set been completely traversed? If so, operation S321 and operations S323 to S325 are performed; if not, operation S322 is performed.

[0267] In operation S321 , a second candidate cell nucleus map corresponding to the connected domain cell region map is obtained.

[0268] In operation S322, a new current third predetermined pixel in the second contour set is determined.

[0269] In operation S323, is the area of ​​the second candidate connected domain in the second candidate cell nucleus map greater than or equal to the third predetermined area threshold? If so, operation S324 is performed; if not, operation S325 is performed.

[0270] In operation S324 , the second candidate connected domain is determined as a connected domain in the cell nucleus map corresponding to the connected domain cell region map.

[0271] In operation S325 , the second candidate connected component is deleted.

[0272] Taking stem cells as an example, the cell image processing method according to the present disclosure will be further described below with reference to FIG. 5A to FIG. 5K in combination with specific embodiments.

[0273] For example, FIG5A schematically illustrates an example of an initial image segmentation result according to an embodiment of the present disclosure. As shown in FIG5A , in step 400A, a target cell image 401 may be processed using a threshold segmentation method to obtain an initial image segmentation result. The initial image segmentation result may include a cell edge map 402, a cell region map 403, and a cell nucleus region map 404.

[0274] FIG5B schematically illustrates an example of obtaining an intermediate cell region map according to an embodiment of the present disclosure. As shown in FIG5B , in 400B, a dilation operation can be used to process cell region map 403 to obtain a first cell region dilation map 405. A fill operation can be used to process first cell region dilation map 405 to obtain a cell region dilation fill map 406. An erosion operation can be used to process cell region dilation fill map 406 to obtain cell region fill map 407. A subtraction operation is performed on cell region fill map 407 and cell edge map 402 to obtain an intermediate cell region map 408.

[0275] For example, FIG5C schematically shows an example schematic diagram of obtaining a connected domain cell area map according to an embodiment of the present disclosure. As shown in FIG5C , in 400C, the intermediate cell area map 408 can be divided according to a predetermined foreground area division strategy to obtain a connected domain cell area map set 409. The connected domain cell area map set 409 may include a connected domain cell area map 409_1, a connected domain cell area map 409_2, and a connected domain cell area map 409_3. The area of ​​the connected domain in the connected domain cell area map 409_1 belongs to a first predetermined area range. The area of ​​the connected domain in the connected domain cell area map 409_2 belongs to a second predetermined area range. The area of ​​the connected domain in the connected domain cell area map 409_3 belongs to a third predetermined area range. For example, the first predetermined area range may be greater than or equal to 600. The second predetermined area range may be greater than 200 and less than 600. The third predetermined area range may be less than or equal to 200.

[0276] FIG5D schematically illustrates an example of obtaining a cell nucleus map corresponding to a connected domain cell region map according to an embodiment of the present disclosure. As shown in FIG5D , in step 400D, the connected domain cell region map 409_1 can be processed using an erosion operation to obtain an intermediate connected domain cell region map 410. An AND operation is performed on the intermediate connected domain cell region map 410 and the cell nucleus map to obtain a first intersection cell region map 411. Based on the first intersection cell region map 411 and the intermediate connected domain cell region map 410, the connected domains in the intermediate connected domain cell region map 410 that are located in the first intersection cell region map 411 are retained, and the remaining regions are removed to obtain a first candidate cell nucleus map 412 corresponding to the connected domain cell region map. The first target connected domain can be deleted from the first candidate cell nucleus map 412 to obtain a first candidate cell nucleus map after deleting the first target connected domain. The first candidate cell nucleus map after deleting the first target connected domain is determined as the cell nucleus map 413 corresponding to the connected domain cell region map 409_1. The first target connected domain may refer to a connected domain whose area is less than a second predetermined area threshold. The center point of each connected domain in the cell nucleus map 413 is drawn in the target cell image to obtain the target cell image 414 .

[0277] FIG5E schematically illustrates an example schematic diagram of a cell nucleus map corresponding to a connected domain cell region map according to another embodiment of the present disclosure. As shown in FIG5E , in 400E, a cell nucleus map 415 corresponding to connected domain cell region map 409_2 can be obtained in a manner similar to that used to obtain connected domain cell region map 409_1. The center point of each connected domain in cell nucleus map 415 is plotted in target cell image 401 to obtain target cell image 416.

[0278] FIG5F schematically illustrates an example schematic diagram of obtaining a cell nucleus map corresponding to a connected domain cell region map according to another embodiment of the present disclosure. As shown in FIG5F , in step 400F, an AND operation is performed on the connected domain cell region map 409_3 and the cell nucleus map to obtain a second intersection cell region map 417. Based on the second intersection cell region map 417 and the connected domain cell region map 409_3, the connected domains in the connected domain cell region map 409_3 that are located in the second intersection cell region map 417 are retained, and the other regions are removed to obtain a first candidate cell nucleus map 418 corresponding to the connected domain cell region map. The second target connected domain can be deleted from the first candidate cell nucleus map 418 to obtain a first candidate cell nucleus map after deleting the second target connected domain. The first candidate cell nucleus map after deleting the second target connected domain is determined as the cell nucleus map 419 corresponding to the connected domain cell region map 409_3. The second target connected domain may refer to a connected domain whose area is less than a third predetermined area threshold. The center point of each connected domain in the cell nucleus map 419 is drawn in the target cell image to obtain the target cell image 420.

[0279] FIG5G schematically illustrates an example schematic diagram of obtaining a target cell nucleus map according to an embodiment of the present disclosure. As shown in FIG5G , in 400G, the cell nucleus map, the cell nucleus map, and the cell nucleus map can be merged to obtain a target cell nucleus map 421. In addition, the center point of each connected domain in the target cell nucleus map 421 can be drawn on the target cell image 401 to obtain a cell nucleus count map 422.

[0280] FIG5H schematically illustrates an example of obtaining a target cell region outline map according to an embodiment of the present disclosure. As shown in FIG5H , in step 400H, a primary cell region map 423 can be obtained based on the target cell nucleus map and the cell region map. The initial cell region map 423 is processed using an expansion operation to obtain a second cell region expansion map 424. Based on the second cell region expansion map 424 and the primary cell region map 423, a first intermediate cell region outline map 425 is obtained.

[0281] Based on the target cell image, a watershed process can be performed on the first intermediate cell region outline image 425 to obtain a fourth intermediate cell region outline image. The pixel values ​​of the pixels in the background region of the fourth intermediate cell region outline image are set to a first predetermined pixel value, and the pixel values ​​of the pixels in the other regions of the fourth intermediate cell region outline image, excluding the background region, are set to a second predetermined pixel value to obtain a second intermediate cell region outline image 426.

[0282] The second intermediate cell region outline map 426 and the target cell nucleus map can be ANDed together to obtain the intersection of the second intermediate cell region outline map and the target cell nucleus map. The connected domain of the intersection of the second intermediate cell region outline map is retained, and the remaining regions except the intersection are removed to obtain the third intermediate cell region outline map 427.

[0283] The cell edge map can be subtracted from the third intermediate cell region outline map 427 to obtain a target cell region outline map 428. In addition, the center point of each connected domain in the target cell region outline map 428 is drawn on the target cell image 401 to obtain a target cell image 429.

[0284] FIG5I schematically illustrates an example of obtaining a target cell outline map according to an embodiment of the present disclosure. As shown in FIG5I , in step 400I, a subtraction operation is performed between the target cell region outline map and the target cell nucleus map to obtain an intermediate cell outline map 430. Based on the target cell image, a watershed process is performed on the intermediate cell outline map 430 to obtain the target cell outline map. Furthermore, different cells can be stained on the target cell image to obtain a target cell image 431.

[0285] FIG5J schematically illustrates an example of a target image segmentation result at an early stage of cell culture according to an embodiment of the present disclosure. As shown in FIG5J , in step 400J, the target cell image 432 is processed using the cell image processing method according to an embodiment of the present disclosure to obtain a target cell nucleus image 433 and a target cell outline image 434. Based on the target cell nucleus image 433 and the target cell outline image 434, a target image segmentation result is obtained.

[0286] FIG5K schematically illustrates an example of a target image segmentation result at a later stage of cell culture according to an embodiment of the present disclosure. As shown in FIG5K , in 400K , the target cell image 435 is processed using the cell image processing method according to an embodiment of the present disclosure to obtain a target cell nucleus image 436 and a target cell outline image 437. Based on the target cell nucleus image 436 and the target cell outline image 437, a target image segmentation result is obtained.

[0287] For example, FIG6 is a block diagram of a cell image processing apparatus according to at least one embodiment of the present disclosure. As shown in FIG6 , the cell image processing apparatus 500 includes a first acquisition module 510 , a second acquisition module 520 , a third acquisition module 530 , and a fourth acquisition module 540 .

[0288] For example, the first obtaining module 510 is configured to obtain a target cell nucleus image based on an initial image segmentation result of a target cell image. The target cell image is an image corresponding to at least one cell.

[0289] For example, the second obtaining module 520 is configured to obtain a target cell region contour map according to the target cell nucleus map and the initial image segmentation result.

[0290] For example, the third obtaining module 530 is configured to obtain a target cell contour map according to the target cell region contour map and the target cell nucleus map.

[0291] For example, the fourth obtaining module 540 is configured to obtain a target image segmentation result according to the target cell outline image and the target cell nucleus image.

[0292] For example, in an embodiment of the present disclosure, the cell image processing apparatus 500 may further include a fifth obtaining module.

[0293] The fifth acquisition module is configured to obtain an initial image segmentation result based on the target cell image based on a threshold segmentation method.

[0294] For example, in an embodiment of the present disclosure, the threshold segmentation method includes a maximum inter-class variance method.

[0295] For example, in an embodiment of the present disclosure, the fifth obtaining module may include a first obtaining sub-module, a second obtaining sub-module, a third obtaining sub-module, and a fourth obtaining sub-module.

[0296] The first obtaining submodule is configured to obtain a first grayscale threshold, a second grayscale threshold, and a third grayscale threshold based on the target cell image based on the maximum inter-class variance method, wherein the second grayscale threshold is greater than the third grayscale threshold and less than the first grayscale threshold.

[0297] The second obtaining submodule is configured to perform binarization processing on the target cell image according to the first grayscale threshold to obtain a cell edge map.

[0298] The third obtaining submodule is configured to perform binarization processing on the target cell image according to the second grayscale threshold to obtain a cell area map.

[0299] The fourth obtaining submodule is configured to perform binarization processing on the target cell image according to the third grayscale threshold to obtain a cell nucleus region map.

[0300] For example, in an embodiment of the present disclosure, the initial image segmentation result includes a cell edge map, a cell region map, and a cell nucleus region map.

[0301] For example, in an embodiment of the present disclosure, the first obtaining module 510 may include a fifth obtaining sub-module, a sixth obtaining sub-module, and a seventh obtaining sub-module.

[0302] The fifth acquisition submodule is configured to obtain an intermediate cell area map according to the cell area map and the cell edge map based on a morphological method.

[0303] The sixth obtaining submodule is configured to obtain at least one connected domain cell region map according to the intermediate cell region map and the predetermined foreground region division strategy.

[0304] The seventh acquisition submodule is configured to obtain a target cell nucleus map based on a morphological method according to the cell nucleus region map and at least one connected domain cell region map.

[0305] For example, in an embodiment of the present disclosure, the fifth obtaining submodule may include a first obtaining unit, a second obtaining unit, a third obtaining unit, and a fourth obtaining unit.

[0306] The first obtaining unit is configured to process the cell region map using an expansion operation to obtain a first cell region expansion map.

[0307] The second obtaining unit is configured to process the first cell region expansion map using a filling operation to obtain a cell region expansion filling map.

[0308] The third obtaining unit is configured to process the cell region expansion filling map using an erosion operation to obtain a cell region filling map.

[0309] The fourth obtaining unit is configured to obtain an intermediate cell area map according to the cell area filling map and the cell edge map.

[0310] For example, in an embodiment of the present disclosure, the fourth obtaining unit may include the first obtaining sub-unit.

[0311] The first obtaining subunit is configured to remove the target cell edge in the cell area filling map according to the cell area filling map and the cell edge map, so as to obtain an intermediate cell area map.

[0312] For example, in an embodiment of the present disclosure, the seventh obtaining submodule may include a fifth obtaining unit and a sixth obtaining unit.

[0313] The fifth obtaining unit is configured to obtain, for the connected domain cell area map in at least one connected domain cell area map, a cell nucleus map corresponding to the connected domain cell area map based on a morphological method and according to the cell nucleus area map and the connected domain cell area map.

[0314] The sixth obtaining unit is configured to obtain a target cell nucleus map according to the cell nucleus map corresponding to each of the at least one connected domain cell region maps.

[0315] For example, in an embodiment of the present disclosure, the fifth obtaining unit may include a second obtaining sub-unit and a third obtaining sub-unit.

[0316] The second acquisition subunit is configured to process the connected domain cell region map based on a morphological method to obtain an intermediate connected domain cell region map when it is determined that the area of ​​the connected domain cell region map is greater than or equal to a first predetermined area threshold; and to obtain a cell nucleus map corresponding to the connected domain cell region map based on the intermediate connected domain cell region map and the cell nucleus region map.

[0317] The third obtaining subunit is configured to obtain a cell nucleus map corresponding to the connected domain cell region map based on the cell nucleus region map and the connected domain cell region map when it is determined that the area of ​​the connected domain cell region map is less than the first predetermined area threshold.

[0318] For example, in an embodiment of the present disclosure, the second obtaining subunit may be configured as follows:

[0319] The middle connected domain cell region map and the cell nucleus region map are ANDed to obtain a first intersection cell region map. According to the first intersection cell region map and the middle connected domain cell region map, a cell nucleus map corresponding to the connected domain cell region map is obtained.

[0320] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the first intersection cell region map and the intermediate connected domain cell region map may include the following operations.

[0321] A first template image is generated. The size of the first template image is equal to the size of the intermediate connected domain cell region map, and the pixel values ​​of the pixels of the first template image are first predetermined pixel values. Based on the first template image, the first intersection cell region map, and the intermediate connected domain cell region map, a first candidate cell nucleus map corresponding to the connected domain cell region map is obtained. Based on the first candidate cell nucleus map corresponding to the connected domain cell region map, a cell nucleus map corresponding to the connected domain cell region map is obtained.

[0322] For example, in an embodiment of the present disclosure, obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map based on the first template image, the first intersection cell region map, and the intermediate connected domain cell region map may include the following operations.

[0323] Determine a first contour set corresponding to the connected domain in the first intersection cell region map. Traverse each first predetermined pixel included in the first contour set, and modify the pixel value of pixels in the first template image that have the same label as the first predetermined pixel from the first predetermined pixel value to the second predetermined pixel value until the traversal is complete, thereby obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map. The label of the first predetermined pixel in the first template image is determined based on the label of the first predetermined pixel in the intermediate connected domain cell region map.

[0324] For example, in an embodiment of the present disclosure, obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map based on the first template image, the first intersection cell region map, and the intermediate connected domain cell region map may include the following operations.

[0325] Each second predetermined pixel in the first template image is traversed, and when it is determined that the pixel value of the pixel corresponding to the second predetermined pixel in the first intersection cell region map is the expected pixel value, the pixel value of the pixel with the same label as the second predetermined pixel in the first template image is modified from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, thereby obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map. The label of the second predetermined pixel in the first template image is determined based on the label of the second predetermined pixel in the intermediate connected domain cell region map.

[0326] For example, in an embodiment of the present disclosure, the first candidate cell nucleus map includes at least one first candidate connected domain. The cell nucleus map includes at least one connected domain.

[0327] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the first candidate cell nucleus map corresponding to the connected domain cell region map may include the following operations: for each first candidate connected domain in the first candidate cell nucleus map corresponding to the connected domain cell region map, when it is determined that the area of ​​the first candidate connected domain is greater than or equal to a second predetermined area threshold, determining the first candidate connected domain as a connected domain in the cell nucleus map corresponding to the connected domain cell region map.

[0328] For example, in an embodiment of the present disclosure, the third obtaining subunit may be configured to: perform an AND operation on the connected domain cell region map and the cell nucleus region map to obtain a second intersection cell region map, and obtain a cell nucleus map corresponding to the connected domain cell region map based on the second intersection cell region map and the connected domain cell region map.

[0329] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the second intersection cell region map and the connected domain cell region map may include the following operations: generating a second template image. The size of the second template image is equal to the size of the connected domain cell region map, and the pixel values ​​of the pixels of the second template image are first predetermined pixel values. Based on the second template image, the second intersection cell region map, and the connected domain cell region map, obtaining a second candidate cell nucleus map corresponding to the connected domain cell region map. Based on the second candidate cell nucleus map corresponding to the connected domain cell region map, obtaining a cell nucleus map corresponding to the connected domain cell region map.

[0330] For example, in an embodiment of the present disclosure, obtaining a second candidate cell nucleus map corresponding to the connected domain cell region map based on the second template image, the second intersection cell region map, and the connected domain cell region map may include the following operations: determining a second contour set corresponding to the connected domain in the second intersection cell region map. Traversing each third predetermined pixel included in the second contour set, modifying the pixel value of the pixel with the same label as the third predetermined pixel in the second template image from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, thereby obtaining a second candidate cell nucleus map corresponding to the connected domain cell region map. The label of the third predetermined pixel in the second template image is determined based on the label of the third predetermined pixel in the connected domain cell region map.

[0331] For example, in an embodiment of the present disclosure, obtaining a second candidate cell nucleus map corresponding to the connected domain cell region map based on the second template image, the second intersection cell region map, and the connected domain cell region map may include the following operations: traversing each fourth predetermined pixel in the second template image, and when determining that the pixel value of the pixel corresponding to the fourth predetermined pixel in the second intersection cell region map is the expected pixel value, modifying the pixel value of the pixel with the same label as the fourth predetermined pixel in the second template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, thereby obtaining the second candidate cell nucleus map corresponding to the connected domain cell region map. The label of the fourth predetermined pixel in the second template image is determined based on the label of the fourth predetermined pixel in the connected domain cell region map.

[0332] For example, in an embodiment of the present disclosure, the second candidate cell nucleus map includes at least one second candidate connected domain. The cell nucleus map includes at least one connected domain.

[0333] For example, in an embodiment of the present disclosure, obtaining a cell nucleus map corresponding to the connected domain cell region map based on the second candidate cell nucleus map corresponding to the connected domain cell region map may include the following operations: for each second candidate connected domain in the second candidate cell nucleus map corresponding to the connected domain cell region map, when it is determined that the area of ​​the second candidate connected domain is greater than or equal to a third predetermined area threshold, determining the second candidate connected domain as a connected domain in the cell nucleus map corresponding to the connected domain cell region map.

[0334] For example, in an embodiment of the present disclosure, the initial image segmentation result includes a cell edge map and a cell area map.

[0335] For example, in an embodiment of the present disclosure, the second acquisition module 520 may include an eighth acquisition submodule, a ninth acquisition submodule, a tenth acquisition submodule, an eleventh acquisition submodule, a twelfth acquisition submodule, and a thirteenth acquisition submodule. The eighth acquisition submodule is configured to obtain a primary cell area map based on the target cell nucleus map and the cell area map. The ninth acquisition submodule is configured to process the initial cell area map using an expansion operation to obtain a second cell area expansion map. The tenth acquisition submodule is configured to obtain a first intermediate cell area contour map based on the second cell area expansion map and the primary cell area map. The eleventh acquisition submodule is configured to obtain a second intermediate cell area contour map based on the first intermediate cell area contour map and the target cell image. The twelfth acquisition submodule is configured to obtain a third intermediate cell area contour map based on the second intermediate cell area contour map and the target cell nucleus map. The thirteenth acquisition submodule is configured to obtain a target cell area contour map based on the third intermediate cell area contour map and the cell edge map.

[0336] For example, in an embodiment of the present disclosure, the eleventh acquisition submodule may include a seventh acquisition unit and an eighth acquisition unit. The seventh acquisition unit is configured to perform watershed processing on the first intermediate cell region outline map based on the target cell image to obtain a fourth intermediate cell region outline map. The eighth acquisition unit is configured to set the pixel values ​​of the pixels in the background region of the fourth intermediate cell region outline map to a first predetermined pixel value, and set the pixel values ​​of the pixels in other regions of the fourth intermediate cell region outline map except the background region to a second predetermined pixel value, to obtain a second intermediate cell region outline map.

[0337] For example, in an embodiment of the present disclosure, the twelfth acquisition submodule may include a generation unit and a ninth acquisition unit. The generation unit is configured to generate a third template image. The size of the third template image is equal to the size of the second intermediate cell region outline image, and the pixel values ​​of the pixels of the third template image are first predetermined pixel values. The ninth acquisition unit is configured to obtain the third intermediate cell region outline image based on the third template image, the second intermediate cell region outline image, and the target cell nucleus image.

[0338] For example, in an embodiment of the present disclosure, the ninth obtaining unit may include a determining subunit and a fourth obtaining subunit. The determining subunit is configured to determine a third contour set corresponding to a connected domain in the target cell nucleus map. The fourth obtaining subunit is configured to traverse each fifth predetermined pixel included in the third contour set, and modify the pixel value of the pixel with the same label as the fifth predetermined pixel in the third template image from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, thereby obtaining a third intermediate cell region contour map. The label of the fifth predetermined pixel in the third template image is determined based on the label of the fifth predetermined pixel in the second intermediate cell region contour map.

[0339] For example, in an embodiment of the present disclosure, the ninth obtaining unit may include a fifth obtaining subunit. The fifth obtaining subunit is configured to traverse each sixth predetermined pixel in the third template image, and when it is determined that the pixel value of the pixel corresponding to the sixth predetermined pixel in the target cell nucleus image is the expected pixel value, the pixel value of the pixel with the same label as the sixth predetermined pixel in the third template image is modified from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, thereby obtaining the third intermediate cell region contour map. The label of the sixth predetermined pixel in the third template image is determined based on the label of the sixth predetermined pixel in the second intermediate cell region contour map.

[0340] For example, in an embodiment of the present disclosure, the third acquisition module 530 may include a fourteenth acquisition submodule and a fifteenth acquisition submodule. The fourteenth acquisition submodule is configured to perform a subtraction operation on the target cell region outline map and the target cell nucleus map to obtain an intermediate cell outline map. The fifteenth acquisition submodule is configured to perform a watershed process on the intermediate cell outline map based on the target cell image to obtain the target cell outline map.

[0341] For example, in an embodiment of the present disclosure, the cell image processing apparatus 500 may further include a first determination module and a second determination module. The first determination module is configured to determine cell statistical information based on the target image segmentation result. The second determination module is configured to determine a cell quality assessment result based on the cell statistical information.

[0342] For example, any number of modules, submodules, units, and subunits in the embodiments of the present disclosure, or at least part of the functions of any number of them can be implemented in one module. According to any one or more of the modules, submodules, units, and subunits in the embodiments of the present disclosure, it can be split into multiple modules for implementation. According to any one or more of the modules, submodules, units, and subunits in the embodiments of the present disclosure, it can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware in any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation modes of software, hardware, and firmware or in an appropriate combination of any of them. Alternatively, according to one or more of the modules, submodules, units, and subunits in the embodiments of the present disclosure, it can be at least partially implemented as a computer program module, which can perform the corresponding function when the computer program module is run.

[0343] For example, any multiple of the first acquisition module 510, the second acquisition module 520, the third acquisition module 530, and the fourth acquisition module 540 can be combined into one module / unit / sub-unit for implementation, or any one of the modules / units / sub-units can be split into multiple modules / units / sub-units. Alternatively, at least part of the functions of one or more of these modules / units / sub-units can be combined with at least part of the functions of other modules / units / sub-units and implemented in one module / unit / sub-unit. For example, in an embodiment of the present disclosure, at least one of the first acquisition module 510, the second acquisition module 520, the third acquisition module 530, and the fourth acquisition module 540 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware in any other reasonable manner of integrating or packaging the circuit, or implemented in any one of the three implementation modes of software, hardware, and firmware, or in an appropriate combination of any of them. Alternatively, at least one of the first obtaining module 510 , the second obtaining module 520 , the third obtaining module 530 , and the fourth obtaining module 540 may be at least partially implemented as a computer program module, which may perform corresponding functions when executed.

[0344] It should be noted that the cell image processing apparatus in the embodiment of the present disclosure corresponds to the cell image processing method in the embodiment of the present disclosure. The description of the cell image processing apparatus refers to the cell image processing method, which will not be repeated here.

[0345] For example, Figure 7 schematically shows a block diagram of an electronic device suitable for implementing the cell image processing method according to an embodiment of the present disclosure. The electronic device shown in Figure 7 is only an example and should not bring any limitation to the functions and scope of use of the embodiment of the present disclosure.

[0346] For example, as shown in Figure 7, the electronic device 600 according to an embodiment of the present disclosure includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage portion 608 into a random access memory (RAM) 603. The processor 601 may, for example, include a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a dedicated microprocessor (e.g., an application-specific integrated circuit (ASIC)), and the like. The processor 601 may also include onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0347] Various programs and data required for the operation of the electronic device 600 are stored in the RAM 603. The processor 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. The processor 601 executes the various operations of the method flow according to the embodiment of the present disclosure by executing the programs in the ROM 602 and / or the RAM 603. It should be noted that the programs may also be stored in one or more memories other than the ROM 602 and the RAM 603. The processor 601 may also execute the various operations of the method flow according to the embodiment of the present disclosure by executing the programs stored in the one or more memories.

[0348] For example, in an embodiment of the present disclosure, electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to bus 604. System 600 may also include one or more of the following components connected to I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and speakers; a storage section 608 including a hard disk; and a communication section 609 including a network interface card such as a LAN card or a modem. Communication section 609 performs communication processing via a network such as the Internet. Drive 610 is also connected to I / O interface 605 as needed. Removable media 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed in drive 610 as needed, so that computer programs read therefrom can be installed into storage section 608 as needed.

[0349] For example, in an embodiment of the present disclosure, the method flow according to the embodiment of the present disclosure can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, the above-mentioned functions defined in the system of the embodiment of the present disclosure are executed. For example, in an embodiment of the present disclosure, the system, device, apparatus, module, unit, etc. described above can be implemented by a computer program module.

[0350] For example, the embodiments of the present disclosure utilize the technology of segmentation and computational analysis of cell morphology images to perform morphological analysis on cell images, establish the relationship between cell morphology and cell biological function, and determine characteristics such as the degree of cell aging. Mesenchymal stem cells can be extracted and separated from different tissues and organs, and cells from the same tissue but different donors have certain differences in proliferation, aging, and function. The embodiments of the present disclosure establish the relationship between cell morphology and function to non-destructively evaluate the quality of mesenchymal stem cells, shorten the time for subsequent screening of more suitable donors, establish suitable master cell banks and working cell banks, and achieve the purpose of preliminary screening of samples, thereby improving efficiency while reducing costs.

[0351] For example, FIG8 is a flow chart of the proliferation culture of mesenchymal stem cells provided by at least one embodiment of the present disclosure. As shown in FIG8 , the proliferation culture of mesenchymal stem cells includes the following steps.

[0352] Step S701: adding fetal bovine serum to the mesenchymal stem cell culture medium to form a first culture medium.

[0353] Step S702: adding mesenchymal stem cells to a first culture medium for culture, and when the confluence of the proliferated mesenchymal stem cells is in the range of 70% to 95%, performing a subculture operation on the mesenchymal stem cells.

[0354] For example, in one embodiment of the present disclosure, the mesenchymal stem cell culture medium is α-MEM (corning, 10-022-CV) culture medium, and the mass percentage of fetal bovine serum FBS (Gibco, 10099-141) in the first culture medium is 1% to 10%. For example, in one embodiment, the mass percentage of fetal bovine serum FBS (Gibco, 10099-141) in the first culture medium is 10%. For example, during the process of culturing mesenchymal stem cells in the first culture medium, the mesenchymal stem cell culture medium is replaced every two days to ensure that the mesenchymal stem cells have sufficient nutrition. Cell passage operation is performed when the cell proliferation confluence reaches 70% to 95%. For example, in one embodiment, cell passage operation is performed when the cell proliferation confluence reaches 80% to 90%. For example, in another embodiment, cell passage operation is performed when the cell proliferation confluence reaches 85%.

[0355] For example, in one embodiment of the present disclosure, the seeding density of mesenchymal stem cells in the first culture medium is 1000 to 30000 / cm 2 For example, in another embodiment of the present disclosure, the seeding density of mesenchymal stem cells in the first culture medium is 3000 to 20000 / cm 2For example, in another embodiment of the present disclosure, the seeding density of mesenchymal stem cells in the first culture medium is 10000 cm 2 .

[0356] For example, in an embodiment of the present disclosure, the subculture operation of mesenchymal stem cells includes: aspirating the mesenchymal stem cell culture medium in the first cell culture container, and cleaning the cell culture container with a buffer detergent having a pH of 7.2 to 7.4, and aspirating the buffer detergent. Adding 7ML to 10ML of pancreatic enzyme to the first cell culture container, digesting the mesenchymal stem cells with pancreatic enzyme for 1 to 2 minutes, so that the mesenchymal stem cells leave the bottom surface of the first cell culture container, adding mesenchymal stem cell culture medium to the first cell culture container to terminate the digestion of the mesenchymal stem cells, and forming a first cell suspension. Using a 10ml pipette, the cell suspension is collected into a 50ml centrifuge tube and centrifuged to obtain a cell pellet, adding 3ML to 5ML of the mesenchymal stem cell culture medium to the cell pellet, and forming a second suspension.

[0357] For example, in an embodiment of the present disclosure, the passaging operation of mesenchymal stem cells further includes: mixing the second suspension evenly, aspirating the second suspension into a sterile centrifuge tube, adding an amount of dye equal to the aspirated second suspension into the centrifuge tube, and mixing evenly to obtain a third suspension.

[0358] For example, in an embodiment of the present disclosure, the passaging operation of mesenchymal stem cells further includes: adding the third suspension to a cell counting plate to perform cell counting, then inoculating the second suspension into a second cell culture container according to a predetermined inoculation density, then adding mesenchymal stem cell culture medium to the second cell culture container, and culturing the next generation of mesenchymal stem cells.

[0359] For example, in one embodiment of the present disclosure, the cell passaging operation includes: aspirating and discarding the culture medium in the cell culture container, washing the cell culture container with fetal bovine serum (FBS) (Corning, 21-040-CV), aspirating and discarding the fetal bovine serum (FBS), adding 7 to 10 ml of recombinant synthetic trypsin TrypLE (Gibco, 12563-029) to the cell culture container, digesting the mesenchymal stem cells for about 1 to 2 minutes, so that the mesenchymal stem cells are separated from the bottom surface of the cell culture container, adding mesenchymal stem cell culture medium to the cell culture container to terminate the digestion of the mesenchymal stem cells, collecting the cell suspension in the cell culture container using a 10 ml pipette (Corning, 4488) into a 50 ml centrifuge tube (Corning, 430829), and centrifuging using a Thermo Scientific Sorvall centrifuge (Thermo Scientific Sorvall centrifuge). ST16R) were centrifuged (centrifugal force 300 × g, centrifugation time 5 minutes), the supernatant was discarded, and the cell pellet was resuspended in 3-5 ml of mesenchymal stem cell culture medium α-MEM (corning, 10-022-CV). The cell suspension was mixed using a 1 ml pipette tip (Kejin, KG1333), 20 μL of the cell suspension was drawn into a sterile EP tube (AXYGEN, MCT-150-C), 20 μL of AOPI (Countstar, RE010211) was added to the sterile EP tube, mixed evenly, 20 μL was taken and added to the Countstar cell counting plate, the instrument was read, and then the cells were counted according to the established seeding density (for example, 10,000 / cm 2 ) The mesenchymal stem cells were seeded into a new cell culture container, and a new mesenchymal stem cell culture medium, α-MEM (corning, 10-022-CV) culture medium containing 10% fetal bovine serum (FBS), was added to the cell culture container, and subsequent culture operations were continued.

[0360] For example, the cell culture method provided in the embodiments of the present disclosure further includes: performing cell activity detection, aging-related β-galactosidase assay, colony formation ability assay, cell cycle assay, in vitro angiogenesis ability assay, peripheral blood mononuclear cell immunosuppression ability assay, mesenchymal stem cell induced differentiation staining test, and mesenchymal stem cell surface marker test on the screened mesenchymal stem cells. For example, experiments can be performed based on umbilical cord mesenchymal stem cells from three different sample sources, using 5th generation MSC (P5) and 10th generation MSC (P10) of each mesenchymal stem cell.

[0361] The method for collecting mesenchymal stem cell images is as follows: mesenchymal stem cells of passage 5 P5 and passage 10 P10 were seeded into 6-well plates, respectively, and the medium was changed after culturing for 16 hours. The 6-well plates were observed under a microscope, and a picture was taken every 4 hours.

[0362] For example, Figure 9 shows images of passage 5 and passage 10 mesenchymal stem cells, according to at least one embodiment of the present disclosure. As shown in Figure 9, when adherent and proliferating in a culture flask, the mesenchymal stem cells exhibit a spindle-shaped morphology. In the same sample, passage 10 mesenchymal stem cells are slightly larger than passage 5 mesenchymal stem cells, with pseudopodia extending in more directions.

[0363] The analysis and measurement of mesenchymal stem cells includes: processing the obtained mesenchymal stem cell images to measure cell area, cell perimeter, cell number, cell confluence, aspect ratio of single cells, and cell density, where cell density is the ratio of the square of perimeter to area, i.e., cell density = cell perimeter. 2 / cell area. A curve of the data obtained through MSC image analysis versus time was then plotted. This curve, with time as the horizontal axis, plotted the trend of MSC characteristic parameters changing over time.

[0364] For example, Figures 10A to 10F are graphs illustrating computational analysis of morphological recognition of mesenchymal stem cells according to at least one embodiment of the present disclosure. For example, Figure 10A is a graph comparing the cell counts of passage 5 and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure. Figure 10B is a graph comparing the cell confluence of passage 5 and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure. Figure 10C is a graph comparing the cell area of ​​passage 5 and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure. Figure 10D is a graph comparing the cell perimeter of passage 5 and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure. Figure 10E is a graph comparing the cell aspect ratio of passage 5 and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure. Figure 10F is a graph comparing the cell density of passage 5 and passage 10 mesenchymal stem cells according to at least one embodiment of the present disclosure.

[0365] For example, combining Figures 10A to 10F, analysis and calculation using image segmentation and recognition methods revealed that, at the same time point, the three samples corresponding to passage 5 mesenchymal stem cells had greater cell numbers and higher cell confluence than the three samples corresponding to passage 10 mesenchymal stem cells. During the first 40 hours of culture, the three samples corresponding to passage 5 mesenchymal stem cells exhibited higher cell proliferation rates than the three samples corresponding to passage 10 mesenchymal stem cells. After 40 hours of culture, contact inhibition gradually occurred in the five-passage mesenchymal stem cells, leading to a gradual decrease in proliferation rate. The three samples corresponding to passage 10 mesenchymal stem cells had larger cell areas and larger cell perimeters than the three samples corresponding to passage 5 mesenchymal stem cells. Furthermore, the cell area and cell perimeters decreased with prolonged culture time. This is due to the gradual contraction of the pseudopodia of the mesenchymal stem cells due to contact inhibition as the culture time of the mesenchymal stem cells increases. The aspect ratios of mesenchymal stem cells all showed a trend of first decreasing and then increasing. This is because mesenchymal stem cells proliferate rapidly and take on a small spindle shape, but after contact inhibition, they become elongated. The aspect ratio curve of the fifth generation mesenchymal stem cells dropped to its lowest point more quickly than that of the tenth generation, indicating that the fifth generation mesenchymal stem cells proliferate more rapidly than the tenth generation.

[0366] For example, Figures 10A to 10F show similar trends in the cell density curve and aspect ratio curve, indicating that mesenchymal stem cells are dense during rapid proliferation and elongated after contact inhibition. The lowest values ​​of the aspect ratio curve and cell density curve for passage 5 mesenchymal stem cells are both lower than those for passage 10 mesenchymal stem cells from the same sample, indicating that passage 5 mesenchymal stem cells are more compact and have a better roundness.

[0367] For example, the cell activity test of the screened mesenchymal stem cells includes: adding CCK-8 to the mesenchymal stem cells to secrete an orange-red formazan product, and detecting the specific absorbance of light with a wavelength of 450nm. The higher the absorbance, the better the activity of the mesenchymal stem cells. The specific detection steps are: resuspending the 5th generation P5 and 10th generation P10 mesenchymal stem cells in a mesenchymal stem cell culture medium at a concentration of 6000 cells / well and seeding them in a 96-well plate. For 7 consecutive days, the CCK-8 detection reagent is used every day to evaluate the cell activity and draw a curve. For example, Figure 11 is a curve diagram of the cell activity of the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells provided in at least one embodiment of the present disclosure. As shown in Figure 11, the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells both show an "S"-shaped growth curve, and the proliferation rate of the 5th generation mesenchymal stem cells is higher than the proliferation rate of the 10th generation mesenchymal stem cells. The CCK-8 detection results shown in FIG11 are consistent with the results calculated by the morphological recognition calculation analysis diagrams of mesenchymal stem cells shown in FIG10A to FIG10F .

[0368] For example, the aging-related β-galactosidase assay for the screened mesenchymal stem cells includes: using an in situ β-galactosidase staining kit to detect cell senescence, and the more blue staining, the higher the proportion of senescent cells. The specific detection steps are: using an in situ β-galactosidase staining kit (Biyuntian, Beijing) to characterize cell senescence. P5 and P10 mesenchymal stem cells of the three samples were resuspended in mesenchymal stem cell culture medium at a concentration of 6000 cells / well and seeded in a 24-well plate. When the confluence reached 80%, the mesenchymal stem cells were fixed and then stained with β-galactosidase for cell senescence assessment. Lysosomes in senescent cells accumulate β-galactosidase, so that senescent cells can be stained blue by the kit, which has good detectability. The more blue staining, the higher the proportion of senescent cells. For example, FIG12 is an image of the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells provided in at least one embodiment of the present disclosure after being stained with β-galactosidase. As shown in FIG12 , in the three samples corresponding to the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells, respectively, the 10th generation mesenchymal stem cells have more obvious blue staining than the 5th generation mesenchymal stem cells, indicating that the number of senescent cells in the 10th generation mesenchymal stem cells is relatively large.

[0369] For example, the colony-forming ability of selected mesenchymal stem cells can be determined by seeding the cells at a concentration of one mesenchymal stem cell per well. After 14 days of culture, the number of wells containing 50 or more mesenchymal stem cells is determined using a microscope. This represents the colony-forming ability of the mesenchymal stem cells. Specifically, the assay involves seeding passage 10 mesenchymal stem cells and passage 5 mesenchymal stem cells in a 96-well plate at a concentration of one cell per well. After 14 days of culture, the number of wells containing 50 or more cells is determined using a microscope. This represents the number of CFUs formed. Mesenchymal stem cells have colony-forming ability, and the ability of a single mesenchymal stem cell to form CFUs can, to a certain extent, reflect surface cell viability. For example, Figure 13 is a comparative graph of the colony-forming ability of passage 5 and passage 10 mesenchymal stem cells, provided in at least one embodiment of the present disclosure. As shown in Figure 13, in all three pairs of samples, the colony-forming ability of passage 5 mesenchymal stem cells was higher than that of passage 10 mesenchymal stem cells.

[0370] For example, cell cycle detection of the screened mesenchymal stem cells includes: obtaining mesenchymal stem cells when the confluence of the mesenchymal stem cells reaches 60% to 70%, fixing the mesenchymal stem cells with 70% ethanol by volume, staining the mesenchymal stem cells with a cell cycle staining kit, and detecting the cell cycle using flow cytometry. The specific detection steps are: inoculating the 10th generation mesenchymal stem cells and the 5th generation mesenchymal stem cells in a T75 culture flask, harvesting the mesenchymal stem cells when the confluence of the mesenchymal stem cells reaches 60% to 70%, fixing the mesenchymal stem cells with 70% ethanol, then staining the cells with a cell cycle staining kit, and detecting the cell cycle using flow cytometry. For example, FIG14 is a comparison diagram of the cell cycles of the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells provided in at least one embodiment of the present disclosure. As shown in FIG14 , the three samples respectively included in the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells all have a typical diploid cell cycle, but the number of G2 / M phase cells in the 5th generation mesenchymal stem cells is significantly more than that in the 10th generation mesenchymal stem cells. As can be seen from FIG12 , the 5th generation mesenchymal stem cells have better proliferation ability.

[0371] For example, it can be concluded from Figures 12 to 14 above that, compared with the 10th generation mesenchymal stem cells, the proportion of senescent cells in the 5th generation mesenchymal stem cells is smaller, the colony-forming ability of the mesenchymal stem cells is stronger, and there are more cells in the division stage.

[0372] For example, the in vitro angiogenesis ability test of the screened mesenchymal stem cells includes: inoculating the mesenchymal stem cells in a culture flask, replacing the mesenchymal stem cell culture medium when the confluence of the mesenchymal stem cells reaches 60% to 70%, and continuing to culture the mesenchymal stem cells for 24 hours, collecting the supernatant of the mesenchymal stem cell culture medium, centrifuging the mesenchymal stem cell supernatant, and filtering the mesenchymal stem cell supernatant using a filter membrane with a pore size of 0.22 μm; incubating a well plate coated with matrix gel in an incubator for 30 minutes, inoculating human umbilical vein endothelial cells (HUVECs) on the matrix gel, adding 100 microliters of culture medium supernatant to the wells of the well plate, and then collecting the formed blood vessels to obtain blood vessel images, observing the connection points and total length of the blood vessels. The more connection points and the longer the total length of the blood vessels, the stronger the in vitro angiogenesis ability of the mesenchymal stem cells. The specific detection steps are as follows: 5th generation mesenchymal stem cells and 10th generation mesenchymal stem cells were seeded into T75 culture flasks, and the culture medium was replaced when the confluence reached 70%. After continuing to culture for 24 hours, the culture supernatant was collected and centrifuged at a centrifugal force of 1000×g for 5 minutes. The mesenchymal stem cell supernatant was filtered using a filter membrane with a pore size of 0.22μm. Matrigel was added to a 48-well plate at 100μl / well and incubated in an incubator for 30 minutes. 100μl human umbilical vein endothelial cells (HUVECs) were plated at a rate of 1×10 4 The cells were inoculated on the matrix gel at a seeding density of 100 μl / well, and 100 μl MSC culture supernatant was added to the wells. 100 μl of fresh mesenchymal stem cell culture medium was used as a negative control, and each sample had 3 parallel wells. The 48-well plate was placed in an incubator and cultured for 12 hours, and an image was taken of each well using a microscope. Image analysis was performed using the "Angiogenesis analyzer" plug-in in ImageJ software. "Number of connections" and "total length" were used as evaluation indicators. For example, Figure 15 is a comparison of the in vitro angiogenesis ability of the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells provided in at least one embodiment of the present disclosure. As shown in Figure 15, the mesenchymal stem cell culture supernatants of the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells can promote the formation of microtubules. The MSC supernatant in the three samples of the 5th generation mesenchymal stem cells has a stronger ability to promote angiogenesis than the MSC supernatant in the three samples of the 10th generation mesenchymal stem cells.

[0373] For example, testing the immunosuppressive capacity of peripheral blood mononuclear cells (PBMCs) on selected mesenchymal stem cells includes treating the mesenchymal stem cells with colchicine to inhibit their proliferation. The PBMCs are then cultured in a 6-well plate using 1640 medium containing 10% fetal bovine serum (FBS) at 37°C and 95% humidity. The PBMCs are then stained using a BrdU staining kit and analyzed using flow cytometry. PBMCs that stain positively for BrdU are considered newly proliferated. The specific testing steps are as follows: three samples of mesenchymal stem cells at passage 5 and passage 10, respectively, are seeded into wells of a 6-well plate and treated with colchicine to inhibit their proliferation. Peripheral blood mononuclear cells (PBMCs) are then cultured in a 6-well plate using 1640 medium containing 10% (v / v) FBS at 37°C, 5% (v / v) CO2, and 95% humidity. For the lymphocyte proliferation assay, PBMCs are activated with 10 μg / mL PHA-M and BrdU is added. PBMCs are collected, stained using a BrdU staining kit, and analyzed using flow cytometry. BrdU is a thymidine analog that can also be incorporated into replicating DNA molecules during cell proliferation, replacing thymine (T). PBMCs that stain positively for BrdU are newly proliferated. This test can characterize the inhibitory effect of mesenchymal stem cells on PBMC proliferation when co-cultured with mesenchymal stem cells. For example, Figure 16 is a graph showing the test results of the immunosuppressive ability of peripheral blood mononuclear cells of the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells provided in at least one embodiment of the present disclosure. As shown in Figure 16, both mesenchymal stem cells can inhibit the proliferation of PBMCs. The proliferation rate of active PBMCs (BRDU-positive) when PBMCs are co-cultured with 5th generation mesenchymal stem cells is lower than the proliferation rate of active PBMCs (BRDU-positive) when PBMCs are co-cultured with 10th generation mesenchymal stem cells, indicating that 5th generation mesenchymal stem cells have stronger immunosuppressive ability than 10th generation mesenchymal stem cells.

[0374] For example, combining Figures 15 and 16, compared with the 10th generation mesenchymal stem cells, the 5th generation mesenchymal stem cells have a stronger ability to promote angiogenesis and a stronger inhibitory effect on PBMC. The 10th generation mesenchymal stem cells are relatively aged compared to the 5th generation mesenchymal stem cells, which is consistent with the analysis and calculation results of the image segmentation and recognition methods.

[0375] For example, the mesenchymal stem cells screened out are subjected to a mesenchymal stem cell induced differentiation staining test, including: inducing differentiation of the mesenchymal stem cells using an adipogenic and osteogenic differentiation kit, and then detecting the differentiation results of adipogenesis, chondrogenesis and osteogenesis by Oil Red O staining, Alcian blue staining and Alizarin red staining. The specific detection steps are: the P5 and P10 MSCs of the three samples are respectively inoculated in a 24-well plate, and MSCs are induced to differentiate using an adipogenic and osteogenic differentiation kit (Sartorius, Beijing, China). The P5 and P10 MSCs of the three samples are collected to the bottom of a centrifuge tube using a centrifugation method, and chondrogenic differentiation is induced using a chondrogenic differentiation kit (Sartorius, Beijing, China). Finally, the adipogenic, chondrogenic and osteogenic differentiation results are detected by Oil Red O staining, Alcian blue staining and Alizarin red staining. Figure 17 is an electron micrograph of three samples provided by at least one embodiment of the present disclosure, each of which includes the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells, and each of which is induced to differentiate into adipocytes, chondroblasts, and osteoblasts in vitro. As can be seen from Figure 17, the three samples provided by the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells, respectively, are induced to differentiate into adipocytes, chondroblasts, and osteoblasts in vitro. In the osteogenic induction differentiation, the differentiation ability of the 5th generation mesenchymal stem cells is significantly better than that of the 10th generation mesenchymal stem cells. The differentiation abilities of the three samples corresponding to the same generation of mesenchymal stem cells are also significantly different. The osteogenic induction differentiation abilities of sample 1 and sample 3 of the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells are significantly better than that of sample 2.

[0376] For example, the screened mesenchymal stem cells are subjected to a mesenchymal stem cell surface marker test, which includes: washing the mesenchymal stem cells twice with a buffer detergent having a pH of 7.2 to 7.4, staining with antibodies CD11B-PE, CD19-FITC, CD34-PE, CD45-APC-CY7, CD73-PE, CD90-FITC, CD105-APC, HLA-DR-APC, and corresponding isotype control antibodies at room temperature in the dark for 30 minutes, then washing twice with a buffer detergent having a pH of 7.2 to 7.4, and detecting the markers on the surface of the mesenchymal stem cells by flow cytometry. Table 1 shows the results of surface marker testing of three samples, including the 5th generation mesenchymal stem cells and the 10th generation mesenchymal stem cells.

[0377] Table 1: Results of surface marker testing of samples of P5 and P10 mesenchymal stem cells

[0378] As can be seen in Table 1, the three samples of passage 5 and passage 10 mesenchymal stem cells each expressed positive CD73, CD90, and CD105, and negative CD11b, CD19, CD34, CD45, and HLA-DR, meeting the standards established by the International Society for Cellular Therapy in 2006. Among the three samples of passage 5 and passage 10 mesenchymal stem cells, passage 10 mesenchymal stem cells exhibited higher levels of negative expression of surface markers than passage 5 mesenchymal stem cells, and the expression levels of these negative markers remained within the standard requirements for mesenchymal stem cells.

[0379] As mesenchymal stem cells are cultured in vitro for extended periods and their passages increase, they gradually age. High-passage mesenchymal stem cells experience a decreased proliferation rate, poor in vitro differentiation ability, and reduced cell function. High-passage mesenchymal stem cells also grow in area and perimeter, while their proliferation rate decreases. By comparing data from passage 5 and passage 10 mesenchymal stem cells, a link between cell morphology and function is established. Image segmentation and recognition methods can be used to preliminarily analyze cell morphology, roughly determine the degree of cell aging, and thus infer cell function, providing guidance for operators.

[0380] At least one embodiment of the present disclosure further provides a mesenchymal stem cell, which is formed using the cell culture method provided by any of the above embodiments. The embodiments of the present disclosure establish the relationship between cell morphology and function to non-destructively assess the quality of mesenchymal stem cells, thereby shortening the time for subsequent screening of more suitable donors, establishing a suitable master cell bank and working cell bank, and achieving the purpose of preliminary screening of samples, thereby improving efficiency while reducing costs. That is, the process of forming mesenchymal stem cells using this cell culture method is simple and low-cost.

[0381] At least one embodiment of the present disclosure also provides a cell screening method, which includes: real-time acquisition of mesenchymal stem cells during proliferation culture to obtain a target cell image; obtaining a target cell nuclear image based on an initial image segmentation result of the target cell image, wherein the target cell image is an image corresponding to at least one cell; obtaining a target cell area contour image based on the target cell area contour image and the target cell nuclear image; obtaining a target cell contour image based on the target cell area contour image and the target cell nuclear image; obtaining a target image segmentation result based on the target cell contour image and the target cell nuclear image; determining cell statistical information based on the target image segmentation result; and when the cell statistical information is obtained, the target cell area contour image and the target cell nuclear image are obtained. When the information meets the threshold requirements of the evaluation criteria, qualified mesenchymal stem cells are screened out. The mesenchymal stem cells are then proliferated and cultured using the cell screening method. The mesenchymal stem cells during the proliferation and culture process are captured in real time to obtain a target cell image, and a final target image segmentation result is obtained. Cell statistical information is determined based on the final target image segmentation result. When the cell statistical information meets the threshold requirements of the evaluation criteria, qualified mesenchymal stem cells are screened out and cultured for the next generation. This avoids the need to culture poor-quality mesenchymal stem cells, thus avoiding ineffective work and improving cell culture efficiency. The disclosed embodiments establish a relationship between the morphology and function of mesenchymal stem cells and provide a cell culture method based on cell image segmentation, which can rapidly and non-destructively evaluate cell quality, thereby saving time and cost. Furthermore, the above operation can obtain image segmentation results without the need for manual data annotation, and the above operation is independent of the environment in which the cell images are captured, thus having strong robustness.

[0382] For example, FIG18 is a flow chart of a cell screening method provided in at least one embodiment of the present disclosure. As shown in FIG18 , the cell screening method includes the following steps.

[0383] Step S801: Real-time acquisition of target cell images is performed on mesenchymal stem cells during proliferation culture.

[0384] Step S802: obtaining a target cell nucleus image according to the initial image segmentation result of the target cell image, wherein the target cell image is an image corresponding to at least one cell.

[0385] Step S803: Obtaining a target cell region contour map according to the target cell nucleus map and the initial image segmentation result.

[0386] Step S804: Obtain a target cell outline map according to the target cell region outline map and the target cell nucleus map.

[0387] Step S805: Obtaining a target image segmentation result according to the target cell outline image and the target cell nucleus image.

[0388] Step S806: Determine cell statistical information based on the target image segmentation result.

[0389] Step S807: When the cell statistical information meets the threshold requirement of the evaluation standard, the qualified mesenchymal stem cells are screened out.

[0390] The specific operations in the above steps can be found in the relevant description above and will not be repeated here.

[0391] For example, assessing cell quality based on image segmentation results can help mitigate errors caused by human factors, standardize cell culture processes, and effectively reduce production costs. In actual production processes, this technology can rapidly analyze continuous target cell images with statistical accuracy exceeding 90%.

[0392] For example, in the embodiments of the present disclosure, the threshold requirement for the number of cells is greater than or equal to 50,000, the threshold requirement for the area of ​​a single cell is less than or equal to 800 square microns, the threshold requirement for the perimeter of a single cell is less than or equal to 160 microns, the threshold requirement for the ratio of the major axis length of a single cell to the minor axis length of a single cell is less than or equal to 2.8, the threshold requirement for the cell density is less than or equal to 34, and the threshold requirement for the cell confluence is greater than or equal to 80%.

[0393] For example, when evaluating cell quality based on image segmentation results, one criterion for determining whether the cell quality meets the requirements may be that the seven evaluation indicators, namely, the number of cells, the area of ​​a single cell, the perimeter of a single cell, the ratio of the major axis length to the minor axis length of a single cell, the cell density, and the cell confluence, are all within the corresponding ranges. Another criterion for determining whether the cell quality meets the requirements may be that six of the seven evaluation indicators, namely, the number of cells, the area of ​​a single cell, the perimeter of a single cell, the ratio of the major axis length to the minor axis length of a single cell, the cell density, and the cell confluence, are within the corresponding ranges. Yet another criterion for determining whether the cell quality meets the requirements may be that five of the seven evaluation indicators, namely, the number of cells, the area of ​​a single cell, the perimeter of a single cell, the ratio of the major axis length to the minor axis length of a single cell, the cell density, and the cell confluence, are within the corresponding ranges. Another criterion for judging whether the quality of cells meets the requirements may be that four of the seven evaluation indicators, namely, the number of cells, the area of ​​a single cell, the perimeter of a single cell, the ratio of the major axis length of a single cell to the minor axis length of a single cell, the cell density and the cell confluence, are within the corresponding ranges.

[0394] At least one embodiment of the present disclosure provides a cell culture method, a cell screening method, and a mesenchymal stem cell, each having at least the following beneficial technical effects: the cell culture method and the cell screening method acquire target cell images in real time from mesenchymal stem cells during proliferation and culture, and obtain a final target image segmentation result. Cell statistical information is determined based on the final target image segmentation result. When the cell statistical information meets the threshold requirements of the evaluation criteria, qualified mesenchymal stem cells are screened and cultured for the next generation. This avoids the need to culture poor-quality mesenchymal stem cells, thus avoiding ineffective work and improving cell culture efficiency. Specifically, the embodiments of the present disclosure establish a relationship between the morphology and function of mesenchymal stem cells and provide a cell culture method based on cell image segmentation, thereby enabling rapid and non-destructive evaluation of cell quality, thereby saving time and cost. Furthermore, the above operation can obtain image segmentation results without the need for manual data annotation, and the above operation is independent of the environment in which the cell images were captured, thus exhibiting strong robustness.

[0395] There are a few points to note:

[0396] (1) The drawings of the embodiments of the present disclosure only relate to the structures related to the embodiments of the present disclosure. Other structures may refer to conventional designs.

[0397] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present disclosure, the thickness of layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale.

[0398] (3) In the absence of conflict, the embodiments of the present disclosure and the features therein may be combined with each other to form new embodiments.

[0399] The above description is only a specific embodiment of the present disclosure, but the protection scope of the present disclosure is not limited thereto. The protection scope of the present disclosure shall be based on the protection scope of the claims.

Claims

1. A cell culture method comprising: Proliferating and culturing mesenchymal stem cells, and acquiring target cell images of the mesenchymal stem cells in real time; Analyzing the target cell image to determine cell statistical information; When the cell statistical information meets the threshold requirement of the evaluation standard, qualified mesenchymal stem cells are screened out.

2. The cell culture method according to claim 1, wherein Analyzing the target cell image includes: Obtaining a target cell nuclear image according to an initial image segmentation result of the target cell image, wherein the target cell image is an image corresponding to at least one cell; Obtaining a target cell region contour map according to the target cell nucleus map and the initial image segmentation result; Obtaining the target cell contour map according to the target cell region contour map and the target cell nucleus map; A target image segmentation result is obtained according to the target cell outline map and the target cell nucleus map.

3. The cell culture method according to claim 1 or 2, wherein: The cell statistical information includes at least one of the following: cell number, area of ​​a single cell, perimeter of a single cell, ratio of the major axis length of a single cell to the minor axis length of a single cell, cell density and cell confluence.

4. The cell culture method according to claim 3, wherein: The threshold requirement for the number of cells is greater than or equal to 50,000, the threshold requirement for the area of ​​a single cell is less than or equal to 800 square microns, the threshold requirement for the perimeter of a single cell is less than or equal to 160 microns, the threshold requirement for the ratio of the major axis length of a single cell to the minor axis length of a single cell is less than or equal to 2.8, the threshold requirement for the cell density is less than or equal to 34, and the threshold requirement for the cell confluence is greater than or equal to 80%.

5. The cell culture method according to any one of claims 1 to 4, further comprising: A quality assessment result of the cells is determined according to the cell statistical information.

6. The cell culture method according to any one of claims 2 to 5, wherein Based on the threshold segmentation method, an initial image segmentation result of the target cell image is obtained.

7. The cell culture method according to claim 6, wherein: The threshold segmentation method includes a maximum inter-class variance method, the initial image segmentation result includes a cell edge map, a cell region map and a cell nucleus region map, and based on the threshold segmentation method, the initial image segmentation result of the target cell image is obtained, including: Based on the maximum inter-class variance method, a first grayscale threshold, a second grayscale threshold and a third grayscale threshold are obtained according to the target cell image, wherein the second grayscale threshold is greater than the third grayscale threshold and less than the first grayscale threshold; Performing binarization processing on the target cell image according to the first grayscale threshold to obtain the cell edge map; Binarizing the target cell image according to the second grayscale threshold to obtain the cell region map; and The target cell image is binarized according to the third grayscale threshold to obtain the cell nucleus region map.

8. The cell culture method according to claim 7, wherein: According to the initial image segmentation result of the target cell image, a target cell nucleus map is obtained, including: Obtaining an intermediate cell region map according to the cell region map and the cell edge map; Obtaining at least one connected domain cell region map according to the intermediate cell region map and the predetermined foreground region division strategy; and The target cell nucleus map is obtained according to the cell nucleus region map and the at least one connected domain cell region map.

9. The cell culture method according to claim 8, wherein: According to the cell area map and the cell edge map, an intermediate cell area map is obtained, including: Processing the cell region map by using an expansion operation to obtain a first cell region expansion map; Processing the first cell region expansion map by a filling operation to obtain a cell region expansion filling map; Processing the cell region expansion filling map by using a corrosion operation to obtain a cell region filling map; and According to the cell region filling map and the cell edge map, the target cell edge in the cell region filling map is removed to obtain the intermediate cell region map.

10. The cell culture method according to claim 8, wherein: Obtaining the target cell nucleus map according to the cell nucleus region map and the at least one connected domain cell region map, comprising: For a connected domain cell region map in the at least one connected domain cell region map, obtaining a cell nucleus map corresponding to the connected domain cell region map according to the cell nucleus region map and the connected domain cell region map; and The target cell nucleus map is obtained according to the cell nucleus maps corresponding to the at least one connected domain cell region map.

11. The cell culture method according to claim 10, wherein: According to the cell nucleus region map and the connected domain cell region map, a cell nucleus map corresponding to the connected domain cell region map is obtained, including: In the case where it is determined that the area of ​​the connected domain cell region map is greater than or equal to a first predetermined area threshold, the connected domain cell region map is processed to obtain an intermediate connected domain cell region map; and, based on the intermediate connected domain cell region map and the cell nucleus region map, a cell nucleus map corresponding to the connected domain cell region map is obtained; and When it is determined that the area of ​​the connected domain cell region map is smaller than the first predetermined area threshold, a cell nucleus map corresponding to the connected domain cell region map is obtained according to the cell nucleus region map and the connected domain cell region map.

12. The cell culture method according to claim 11, wherein The step of obtaining a cell nucleus map corresponding to the connected domain cell region map according to the intermediate connected domain cell region map and the cell nucleus region map comprises: Performing an AND operation on the intermediate connected domain cell region map and the cell nucleus region map to obtain a first intersection cell region map; and According to the first intersection cell region map and the intermediate connected domain cell region map, a cell nucleus map corresponding to the connected domain cell region map is obtained.

13. The cell culture method according to claim 12, wherein: The step of obtaining a cell nucleus map corresponding to the connected domain cell region map according to the first intersection cell region map and the intermediate connected domain cell region map comprises: Generate a first template image, wherein the size of the first template image is equal to the size of the intermediate connected domain cell area map, and the pixel value of the pixel of the first template image is a first predetermined pixel value; Obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map according to the first template image, the first intersection cell region map and the intermediate connected domain cell region map; and According to the first candidate cell nucleus map corresponding to the connected domain cell region map, a cell nucleus map corresponding to the connected domain cell region map is obtained.

14. The cell culture method according to claim 13, wherein: The step of obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map according to the first template image, the first intersection cell region map and the intermediate connected domain cell region map comprises: Determine a first contour set corresponding to the connected domain in the first intersection cell region map; and Traverse each first predetermined pixel included in the first contour set, and modify the pixel value of the pixel with the same label as the first predetermined pixel in the first template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, and obtain the first candidate cell nucleus map corresponding to the connected domain cell area map, wherein the label of the first predetermined pixel in the first template image is determined according to the label of the first predetermined pixel in the intermediate connected domain cell area map.

15. The cell culture method according to claim 13, wherein: The step of obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map according to the first template image, the first intersection cell region map and the intermediate connected domain cell region map comprises: Traversing each second predetermined pixel in the first template image, and in the case of determining that the pixel value of the pixel corresponding to the second predetermined pixel in the first intersection cell region map is the expected pixel value, modifying the pixel value of the pixel having the same label as the second predetermined pixel in the first template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, and obtaining a first candidate cell nucleus map corresponding to the connected domain cell region map, wherein the label of the second predetermined pixel in the first template image is based on the second predetermined pixel in the intermediate connected domain cell region map. The label is determined.

16. The cell culture method according to claim 11, wherein The step of obtaining a cell nucleus map corresponding to the connected domain cell region map according to the cell nucleus region map and the connected domain cell region map comprises: Performing an AND operation on the connected domain cell region map and the cell nucleus region map to obtain a second intersection cell region map; and According to the second intersection cell region map and the connected domain cell region map, a cell nucleus map corresponding to the connected domain cell region map is obtained; Generate a second template image, wherein the size of the second template image is equal to the size of the connected domain cell area map, and the pixel value of the pixel of the second template image is the first predetermined pixel value; According to the second template image, the second intersection cell region map and the connected domain cell region map, a second candidate cell nucleus map corresponding to the connected domain cell region map is obtained; and According to the second candidate cell nucleus map corresponding to the connected domain cell region map, a cell nucleus map corresponding to the connected domain cell region map is obtained.

17. The cell culture method according to claim 16, wherein: The step of obtaining a second candidate cell nucleus map corresponding to the connected domain cell region map according to the second template image, the second intersection cell region map and the connected domain cell region map comprises: Determine a second contour set corresponding to the connected domain in the second intersection cell region map; and Traverse each third predetermined pixel included in the second contour set, and modify the pixel value of the pixel with the same label as the third predetermined pixel in the second template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, and obtain the second candidate cell nucleus map corresponding to the connected domain cell area map, wherein the label of the third predetermined pixel in the second template image is determined according to the label of the third predetermined pixel in the connected domain cell area map.

18. The cell culture method according to claim 16, wherein: The step of obtaining a second candidate cell nucleus map corresponding to the connected domain cell region map according to the second template image, the second intersection cell region map and the connected domain cell region map comprises: Traverse each fourth predetermined pixel in the second template image, and determine the second intersection detail. When the pixel value of the pixel corresponding to the fourth predetermined pixel in the cell area map is the expected pixel value, the pixel value of the pixel with the same label as the fourth predetermined pixel in the second template image is modified from the first predetermined pixel value to the second predetermined pixel value until the traversal is completed, and a second candidate cell nucleus map corresponding to the connected domain cell area map is obtained, wherein the label of the fourth predetermined pixel in the second template image is determined according to the label of the fourth predetermined pixel in the connected domain cell area map.

19. The cell culture method according to claim 6, wherein: The initial image segmentation result includes a cell edge map and a cell region map; and a target cell region contour map is obtained according to the target cell nucleus map and the initial image segmentation result, including: Obtaining a primary cell region map according to the target cell nucleus map and the cell region map; Processing the initial cell region map by using an expansion operation to obtain a second cell region expansion map; Obtaining a first intermediate cell region contour map according to the second cell region expansion map and the primary cell region map; Obtaining a second intermediate cell region contour map according to the first intermediate cell region contour map and the target cell image; Obtaining a third intermediate cell region outline map according to the second intermediate cell region outline map and the target cell nucleus map; and The target cell area contour map is obtained according to the third intermediate cell area contour map and the cell edge map.

20. The cell culture method according to claim 19, wherein: The step of obtaining a second intermediate cell region contour map according to the first intermediate cell region contour map and the target cell image comprises: Based on the target cell image, performing watershed processing on the first intermediate cell region contour map to obtain a fourth intermediate cell region contour map; and The pixel values ​​of the pixels in the background area of ​​the fourth intermediate cell area contour map are set to the first predetermined pixel values, and the pixel values ​​of the pixels in other areas of the fourth intermediate cell area contour map except the background area are set to the second predetermined pixel values, to obtain the second intermediate cell area contour map.

21. The cell culture method according to claim 19 or 20, wherein: The step of obtaining the third intermediate cell region contour map according to the second intermediate cell region contour map and the target cell nucleus map comprises: generating a third template image, wherein the size of the third template image is equal to the size of the second intermediate cell region contour image, and the pixel value of the pixel of the third template image is the first predetermined pixel value; and The third intermediate cell region contour map is obtained according to the third template image, the second intermediate cell region contour map and the target cell nucleus map.

22. The cell culture method according to claim 21, wherein The step of obtaining the third intermediate cell region contour map according to the third template image, the second intermediate cell region contour map and the target cell nucleus map comprises: Determine a third contour set corresponding to the connected domain in the target cell nucleus map; and Traverse each fifth predetermined pixel included in the third contour set, and modify the pixel value of the pixel with the same label as the fifth predetermined pixel in the third template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, to obtain the third intermediate cell area contour map, wherein the label of the fifth predetermined pixel in the third template image is determined according to the label of the fifth predetermined pixel in the second intermediate cell area contour map.

23. The cell culture method according to claim 21, wherein: The step of obtaining the third intermediate cell region contour map according to the third template image, the second intermediate cell region contour map and the target cell nucleus map comprises: Traverse each sixth predetermined pixel in the third template image, and when it is determined that the pixel value of the pixel corresponding to the sixth predetermined pixel in the target cell nuclear image is the expected pixel value, modify the pixel value of the pixel with the same label as the sixth predetermined pixel in the third template image from the first predetermined pixel value to the second predetermined pixel value, until the traversal is completed, and obtain the third intermediate cell area contour map, wherein the label of the sixth predetermined pixel in the third template image is determined according to the label of the sixth predetermined pixel in the second intermediate cell area contour map.

24. The cell culture method according to any one of claims 2 to 23, wherein The step of obtaining the target cell contour map according to the target cell region contour map and the target cell nucleus map comprises: Subtracting the target cell region contour map from the target cell nucleus map to obtain the intermediate cell contour map; and Based on the target cell image, the intermediate cell outline map is subjected to watershed processing to obtain the target cell outline map.

25. The cell culture method according to any one of claims 2 to 24, wherein The proliferation culture of mesenchymal stem cells comprises: adding fetal bovine serum to the mesenchymal stem cell culture medium to form a first culture medium; The mesenchymal stem cells are added to the first culture medium for culture, and when the confluence of the proliferated mesenchymal stem cells is in the range of 70% to 95%, the mesenchymal stem cells are passaged.

26. The cell culture method according to claim 25, wherein: The mass percentage of the fetal bovine serum in the first culture medium is 1% to 10%.

27. The cell culture method according to claim 25, wherein: The seeding density of the mesenchymal stem cells in the first culture medium is 1000-30000 / cm 2 .

28. The cell culture method according to any one of claims 25 to 27, wherein During the process of culturing the mesenchymal stem cells in the first culture medium, the mesenchymal stem cell culture medium is replaced every two days.

29. The cell culture method according to claim 28, wherein: The subculturing operation of the mesenchymal stem cells comprises: Aspirating the mesenchymal stem cell culture medium in the first cell culture container, washing the cell culture container with a buffer detergent having a pH of 7.2 to 7.4, and aspirating the buffer detergent; Adding 7-10 ml of pancreatic enzyme into the first cell culture container, and digesting the mesenchymal stem cells with the pancreatic enzyme for 1-2 minutes; The mesenchymal stem cells are separated from the bottom surface of the first cell culture container, and the mesenchymal stem cell culture medium is added to the first cell culture container to terminate the consumption of the mesenchymal stem cells. Activate and form a first cell suspension; The cell suspension was collected into a 50 ml centrifuge tube using a 10 ml pipette, and centrifuged to obtain a cell pellet; 3 ML to 5 ML of the mesenchymal stem cell culture medium is added to the cell pellet to form a second suspension.

30. The cell culture method according to claim 29, further comprising: The second suspension is mixed evenly, and the second suspension is pipetted into a sterile centrifuge tube. An equal amount of dye to the pipetted second suspension is added into the centrifuge tube, and the mixture is mixed evenly to obtain a third suspension.

31. The cell culture method according to claim 30, further comprising: The third suspension is added to a cell counting plate to perform cell counting, and then the second suspension is inoculated into a second cell culture container according to a predetermined inoculation density, and then the mesenchymal stem cell culture medium is added to the second cell culture container to culture the next generation of mesenchymal stem cells.

32. The cell culture method according to any one of claims 2 to 31, further comprising: The screened mesenchymal stem cells were subjected to cell activity detection, aging-related β-galactosidase detection, colony formation ability detection, cell cycle detection, in vitro angiogenesis ability detection, peripheral blood mononuclear cell immunosuppression ability detection, mesenchymal stem cell induced differentiation staining test and mesenchymal stem cell surface marker test.

33. The cell culture method according to claim 32, wherein: The cell activity detection of the selected mesenchymal stem cells includes: adding CCK-8 to the mesenchymal stem cells to secrete orange-red formazan products, and detecting the specific absorbance of light with a wavelength of 450nm. The higher the absorbance, the better the activity of the mesenchymal stem cells.

34. The cell culture method according to claim 32, wherein: The senescence-related β-galactosidase assay of the selected mesenchymal stem cells includes: using an in situ β-galactosidase staining kit to detect cell senescence, and the more blue staining there is, the higher the proportion of senescent cells.

35. The cell culture method according to claim 32, wherein: The colony-forming ability of the screened mesenchymal stem cells is determined by inoculating the cells at a concentration of one mesenchymal stem cell in one well, and after culturing for 14 days, observing the number of wells containing 50 or more mesenchymal stem cells under a microscope, which is the colony-forming ability of the mesenchymal stem cells.

36. The cell culture method according to claim 32, wherein: The cell cycle detection of the screened mesenchymal stem cells includes: when the confluence of the mesenchymal stem cells reaches 60% to 70%, the mesenchymal stem cells are obtained, the mesenchymal stem cells are fixed with 70% ethanol by volume, the mesenchymal stem cells are stained with a cell cycle staining kit, and the cell cycle is detected by flow cytometry.

37. The cell culture method according to claim 32, wherein: The in vitro angiogenesis ability detection of the selected mesenchymal stem cells includes: inoculating the mesenchymal stem cells in a culture bottle, when the confluence of the mesenchymal stem cells reaches 60% to 70%, replacing the mesenchymal stem cell culture medium, and continuing to culture the mesenchymal stem cells for 24 hours, collecting the supernatant of the mesenchymal stem cell culture medium, centrifuging the mesenchymal stem cell supernatant, and filtering the mesenchymal stem cell supernatant using a filter membrane with a pore size of 0.22 μm; The well plate coated with matrix gel was incubated in an incubator for 30 minutes, human umbilical vein endothelial cells were seeded on the matrix gel, and 100 microliters of the culture medium supernatant was added to the wells of the well plate. Then, the formed blood vessels were imaged to obtain vascular images, and the connection points and total length of the blood vessels were observed. The more connection points and the longer the total length of the blood vessels, the stronger the ability of the mesenchymal stem cells to perform in vitro angiogenesis.

38. The cell culture method according to claim 32, wherein: The immunosuppressive ability test of peripheral blood mononuclear cells on the selected mesenchymal stem cells includes: The mesenchymal stem cells are treated with colchicine to inhibit the proliferation of the mesenchymal stem cells, peripheral blood mononuclear cells are added to a 6-well plate, a 1640 culture medium containing 10% by volume of fetal bovine serum is used, and co-culture is carried out at a temperature of 37° C. and a humidity of 95%, and then the peripheral blood mononuclear cells are stained with a BrdU staining kit, and then the peripheral blood mononuclear cells are detected by a flow cytometer, and the peripheral blood mononuclear cells that are positive for BrdU staining during the detection are newly proliferated.

39. The cell culture method according to claim 32, wherein: The mesenchymal stem cells selected are subjected to a mesenchymal stem cell induced differentiation staining test, including: inducing differentiation of the mesenchymal stem cells using an adipogenic and osteogenic differentiation kit, and then detecting the adipogenic, chondrogenic and osteogenic differentiation results by Oil Red O staining, Alcian Blue staining and Alizarin Red staining, respectively.

40. The cell culture method according to claim 32, wherein: The screened mesenchymal stem cells are tested for surface markers of mesenchymal stem cells, including: washing the mesenchymal stem cells twice with a buffer detergent having a pH of 7.2 to 7.4, staining with antibodies CD11B-PE, CD19-FITC, CD34-PE, CD45-APC-CY7, CD73-PE, CD90-FITC, CD105-APC, HLA-DR-APC and corresponding isotype control antibodies at room temperature in the dark for 30 minutes, then washing twice with a buffer detergent having a pH of 7.2 to 7.4, and detecting the markers on the surface of the mesenchymal stem cells by flow cytometry analysis.

41. A mesenchymal stem cell formed by the cell culture method according to any one of claims 1 to 40.

42. A cell screening method comprising: Real-time acquisition of the mesenchymal stem cells during proliferation culture to obtain target cell images; Obtaining a target cell nuclear image according to an initial image segmentation result of the target cell image, wherein the target cell image is an image corresponding to at least one cell; Obtaining a target cell region contour map according to the target cell nucleus map and the initial image segmentation result; Obtaining the target cell contour map according to the target cell region contour map and the target cell nucleus map; Obtaining a target image segmentation result according to the target cell outline map and the target cell nucleus map; Determining cell statistical information according to the target image segmentation result; When the cell statistical information meets the threshold requirement of the evaluation standard, qualified mesenchymal stem cells are screened out.

43. The cell screening method according to claim 42, wherein: The cell statistical information includes at least one of the following: cell number, area of ​​a single cell, perimeter of a single cell, ratio of the major axis length of a single cell to the minor axis length of a single cell, cell density and cell confluence.

44. The cell screening method according to claim 42 or 43, wherein: The threshold requirement for the number of cells is greater than or equal to 50,000, the threshold requirement for the area of ​​a single cell is less than or equal to 800 square microns, the threshold requirement for the perimeter of a single cell is less than or equal to 160 microns, the threshold requirement for the ratio of the major axis length of a single cell to the minor axis length of a single cell is less than or equal to 2.8, the threshold requirement for the cell density is less than or equal to 34, and the threshold requirement for the cell confluence is greater than or equal to 80%.

45. The cell screening method according to any one of claims 42 to 44, further comprising: A quality assessment result of the cells is determined according to the cell statistical information.

Images