A method for detecting the density of induced pluripotent stem cells and a computer device

Abstract

This application relates to the field of image processing technology, and specifically relates to a method for detecting the density of induced pluripotent stem cells and a computer device. The method includes: obtaining an image of induced pluripotent stem cells; presetting the local area of each pixel point; obtaining the local brightness performance value according to the gray-scale distribution of the pixel points in the local area; obtaining the local boundary blur degree according to the gradient distribution characteristics; obtaining the local morphological regularity according to the uniformity characteristics of the gray-scale distribution within the local area of the pixel point; obtaining the density comprehensive performance value of each pixel point; clustering the pixel points to obtain a preset number of clustering clusters; respectively enhancing each clustering cluster by using histogram equalization to obtain an enhanced image of induced pluripotent stem cells, and performing density detection on the enhanced image of induced pluripotent stem cells. This application aims to enhance the image of induced pluripotent stem cells and improve the accuracy of the cell density detection result.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and particularly to a method for detecting the density of induced pluripotent stem cells and a computer device. Background Art

[0002] The detection of the density of induced pluripotent stem cells (iPSCs) plays an important role in regenerative medicine and cell therapy. With the continuous progress of iPSC technology, accurate detection and analysis of cell density have become particularly crucial because it directly affects the assessment of cell proliferation rate, the optimization of culture conditions, and the prediction of cell yield. Traditional manual counting methods have been gradually replaced by automated and semi-automated image analysis techniques, which can quickly and accurately identify and count cells with the help of computer vision and machine learning algorithms, thus significantly improving the detection efficiency and accuracy.

[0003] Currently, image enhancement often improves the contrast through histogram equalization. However, traditional histogram equalization methods are usually based on the global processing of the entire image and do not consider the local features of the image. This global processing method may lead to over-enhancement or under-enhancement of the contrast in some regions, thereby affecting the detail performance of the image. During the process of detecting the density of induced pluripotent stem cells, this problem may result in unclear cell boundaries, affecting the discrimination between cells, and thus reducing the accuracy of the final density detection. Summary of the Invention

[0004] In view of the above, it is necessary to provide a method for detecting the density of induced pluripotent stem cells and a computer device to solve the above problems.

[0005] According to one aspect of this application, a method for detecting the density of induced pluripotent stem cells is provided. The method includes:

[0006] Obtain an image of induced pluripotent stem cells;

[0007] Taking each pixel point as the center, preset a first local region and a second local region for each pixel point, where the second local region contains the first local region; obtain the local brightness performance value of each pixel point according to the gray-scale distribution of the pixel points in the first local region and the second local region; obtain the local boundary blurriness of each pixel point according to the gradient distribution characteristics of each pixel point in the preset direction in the second local region; obtain the local morphological regularity of each pixel point according to the uniformity characteristics of the gray-scale distribution in each direction in the second local region of each pixel point.

[0008] Based on the local boundary blurriness and local morphological regularity of each pixel, combined with the local brightness performance value, obtain the density comprehensive performance value of each pixel; cluster the pixels based on the density comprehensive performance value to obtain a preset number of clustering clusters; use histogram equalization to enhance each clustering cluster respectively to obtain the enhanced induced pluripotent stem cell image, and perform density detection on the enhanced induced pluripotent stem cell image.

[0009] Among them, the obtaining of the local brightness performance value of each pixel is specifically as follows:

[0010] Denote the average gray value of all pixels in the first local area of each pixel as the first average value; denote the average gray value of all pixels in the second local area of each pixel as the second average value; calculate the absolute value of the difference between the first average value and the second average value, and normalize the ratio of the first average value to the absolute value of the difference to obtain the local brightness performance value of each pixel.

[0011] Among them, the obtaining of the local boundary blurriness of each pixel is specifically as follows:

[0012] Denote the local boundary blurriness of the i-th pixel as , and its formula form is:

[0013]

[0014] Among them, L represents the number of preset directions; represents the number of pixels in the second local area in the -th direction of the i-th pixel; represents the set of gradient values of all pixels in the second local area in the -th direction of the i-th pixel; represents the gradient value of the m-th pixel in the second local area in the -th direction of the i-th pixel; norm() represents the normalization function; min() represents the minimum value function.

[0015] Among them, the obtaining of the local morphological regularity of each pixel is specifically as follows:

[0016] According to the distribution of the gray value differences between adjacent pixels in each direction within the second local area of each pixel, obtain the gray uniformity of each direction within the second local area of each pixel;

[0017] According to the degree of dispersion of the gray uniformity in multiple directions within the second local area of each pixel, obtain the local morphological regularity of each pixel.

[0018] Among them, obtaining the gray-scale uniformity in each direction within the second local region of each pixel point specifically includes:

[0019] The standard deviation of the gray-scale differences between all adjacent pixel points in each direction within the second local region of each pixel point; using the negative correlation mapping result of the standard deviation as the gray-scale uniformity in each direction within the second local region of each pixel point.

[0020] Among them, the specific formula for obtaining the density comprehensive performance value of each pixel point is: ; where represents the density comprehensive performance value of the i-th pixel point, represents the local brightness performance value of the i-th pixel point, represents the local morphological regularity of the i-th pixel point, represents the local boundary blur degree of the i-th pixel point.

[0021] Among them, clustering the pixel points includes:

[0022] Using the density comprehensive performance values of all pixel points as the input of the clustering algorithm, using the absolute value of the difference between the density comprehensive performance values of pixel points as the metric between pixel points in the clustering algorithm, and outputting a preset number of clustering clusters.

[0023] Among them, performing density detection on the enhanced induced pluripotent stem cell image specifically includes:

[0024] Training a neural network based on induced pluripotent stem cell images with known cell densities, using the induced pluripotent stem cell image to be detected as the input, and obtaining the cell density using the trained neural network.

[0025] According to another aspect of the present invention, there is provided a computer device for detecting the density of induced pluripotent stem cells, including:

[0026] An induced pluripotent stem cell image acquisition module for acquiring induced pluripotent stem cell images;

[0027] An induced pluripotent stem cell image analysis module for taking each pixel point as the center, presetting the first local region and the second local region of each pixel point, where the second local region includes the first local region; obtaining the local brightness performance value of each pixel point according to the gray-scale distribution of pixel points within the first local region and the second local region of each pixel point; obtaining the local boundary blur degree of each pixel point according to the gradient distribution characteristics in the preset direction within the second local region of each pixel point; obtaining the local morphological regularity of each pixel point according to the uniformity characteristics of the gray-scale distribution in each direction within the second local region of each pixel point;

[0028] The induced pluripotent stem cell image density detection module is used to obtain a comprehensive density performance value of each pixel based on the local boundary fuzziness and local morphological regularity of each pixel, combined with the local brightness performance value; cluster the pixels based on the comprehensive density performance value to obtain a preset number of clusters; use histogram equalization to enhance each cluster respectively to obtain an enhanced induced pluripotent stem cell image, and perform density detection on the enhanced induced pluripotent stem cell image.

[0029] In the above scheme, an induced pluripotent stem cell image is obtained, and a local area is preset with each pixel as the center. The local brightness expression value is obtained by analyzing the grayscale distribution between different local areas to describe the density distribution of the area where the pixel is located. Since the density distribution of induced pluripotent stem cells is different, the boundary characteristics of the corresponding areas are different. Therefore, the local boundary fuzziness is obtained according to the gradient distribution characteristics of the pixel in the preset direction in the local area, which helps to further judge the density of the pixel location according to the fuzziness of the boundary. Since the cell morphology of induced pluripotent stem cells is related to density, the local morphological regularity is obtained according to the uniformity characteristics of the grayscale distribution in each direction in the second local area of each pixel, and the uniformity characteristics of the grayscale distribution are used to quantify whether the cell morphology is regular. The comprehensive density expression value of each pixel is obtained to measure the density characteristics of the local area of the pixel. Then, a clustering algorithm is used to obtain different density distribution areas, and each area is enhanced by histogram equalization to avoid the loss of local details and excessive enhancement caused by traditional histogram equalization, which helps to improve the accuracy of subsequent cell density detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A flowchart of the steps of a method for detecting the density of induced pluripotent stem cells provided in the present application;

[0031] Figure 2 A schematic diagram for obtaining the comprehensive density performance value provided for this application;

[0032] Figure 3 A block diagram of a computer device for detecting density of induced pluripotent stem cells provided in this application. DETAILED DESCRIPTION

[0033] In the description of the embodiments of the present application, words such as "exemplary", "or", "for example" and the like are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "or", "for example" and the like is intended to present related concepts in a concrete manner.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs. The terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0035] In addition, it should be noted that the terms "first" and "second" in this application and the accompanying drawings are used to distinguish similar objects and are not used to describe a specific order or sequence. For the methods disclosed in the embodiments of this application or shown in the flowcharts, including one or more steps for implementing the method, without departing from the scope of this application, the execution order of multiple steps can be interchanged with each other, and some steps can also be deleted.

[0036] Please refer to Figure 1 , which shows a step flowchart of a method for detecting the density of induced pluripotent stem cells provided by an embodiment of this application. The method includes the following steps:

[0037] Step 1: Obtain an image of induced pluripotent stem cells.

[0038] Collect a certain number of cells from the induced pluripotent stem cell culture dish and suspend them in an appropriate culture medium. To prevent the cells from moving during the detection process, fixation treatment is usually required. A common method is to add a certain concentration of formaldehyde or other fixatives to stabilize the cell structure. Then, use DAPI (4',6-diamidino-2-phenylindole) to stain the cells so as to observe the cell morphology and structure under a microscope. Finally, use a microscope to obtain the stained cell image and perform grayscale processing to obtain the induced pluripotent stem cell image.

[0039] Step 2: With each pixel point as the center, preset the first local region and the second local region of each pixel point, where the second local region contains the first local region; obtain the local brightness performance value of each pixel point according to the gray-scale distribution of the pixel points in the first local region and the second local region; obtain the local boundary blur degree of each pixel point according to the gradient distribution characteristics of each pixel point in the preset direction in the second local region; obtain the local morphological regularity of each pixel point according to the uniformity characteristics of the gray-scale distribution in each direction within the second local region of each pixel point.

[0040] During the culture process of induced pluripotent stem cells, the growth and development of cells and the changes in culture conditions will result in different distribution patterns of cell density in different regions.

[0041] In areas with higher cell density, cells are closely arranged and overlap more, resulting in increased absorption and scattering of light, which ultimately causes these areas to appear lower brightness or darker grayscale values in the image. Conversely, in areas with lower cell density, there are larger gaps between cells, and more light can pass through these areas to reach the image sensor, causing these areas to appear as brighter grayscale values.

[0042] Based on this, with each pixel as the center, a first local area and a second local area of each pixel are preset; in this embodiment, the size of the first local area of the pixel is ; The size of the second local area is ; The second local area includes the first local area; the implementer can adjust the size of the first local area and the second local area according to actual conditions, and this application does not impose any restrictions on this.

[0043] According to the grayscale distribution of pixels in two local areas of each pixel, the local brightness expression value of each pixel is obtained: the mean of the grayscale values of all pixels in the first local area of each pixel is recorded as the first mean; the mean of the grayscale values of all pixels in the second local area of each pixel is recorded as the second mean; the absolute value of the difference between the first mean and the second mean is calculated, and the ratio of the first mean to the absolute value of the difference is normalized to obtain the local brightness expression value of each pixel.

[0044] It should be noted that in order to avoid the situation where the denominator is zero, a preset value needs to be added to the denominator. The preset value in this embodiment is 0.1, and the implementer can adjust it according to the actual situation.

[0045] For each pixel, the closer the grayscale distribution of the pixel in the first local area is to that in the second local area, it means that the pixel is in the core area of a certain cell density area. Conversely, it means that the pixel is in an area where the cell density changes. The higher the overall grayscale value of the pixel in the first local area, the higher the pixel is in an area with lower cell density, and the greater the local brightness performance value.

[0046] For iPS cell regions with different densities, the clarity of cell boundaries will also vary. In high-density regions, cell boundaries may appear blurred due to the close arrangement of cells, while in low-density regions, cell boundaries are usually more obvious and easy to identify due to the larger spacing between cells.

[0047] In high-density regions, the boundaries between cells are not distinct, resulting in insufficient contrast between cell boundaries and the surrounding background. Specifically, this means that the gradient value between cells and the background is low, and in these regions, the gradient change at cell boundaries is relatively smooth, lacking an obvious transition. As a result, the boundaries between cells become blurred, making it difficult to clearly distinguish cells from their surrounding background.

[0048] The Sobel operator is used to obtain the gradient information of each pixel in the induced pluripotent stem cell image. Based on the gradient distribution characteristics of each pixel in the preset direction within the second local region, the local boundary blur degree of each pixel is obtained: Denote the local boundary blur degree of the i-th pixel as , and its formula form is:

[0049]

[0050] where L represents the number of preset directions. In this embodiment, the eight-neighborhood directions of the pixel are taken; represents the number of pixels in the second local region in the -th direction of the i-th pixel; represents the set of gradient values of all pixels in the second local region in the -th direction of the i-th pixel; represents the gradient value of the m-th pixel in the second local region in the -th direction of the i-th pixel; norm() represents the normalization function, specifically the sigmoid function; min() represents the minimum value function.

[0051] By finding the lowest gradient value in each direction of the pixel and using this as a benchmark to compare with the gradient values of other pixels in that direction, is used as the weight to obtain the degree of closeness between the gradient value of the pixel in the same direction and the lowest gradient value, that is, , thereby reflecting the clarity of the regional boundary in each direction. However, since the regional boundaries may not fully coincide in different directions, the weakest visibility among all directions will ultimately be selected as the reference, that is, , and finally normalization is performed; since visibility is negatively correlated with blur degree, is used to measure the boundary blur degree index of the local region of the i-th pixel, that is, the local boundary blur degree. In short, by comparing the gradient values in each direction, the most blurred boundary performance is found, and this value is used to measure the boundary blur degree of the local region of the i-th pixel.

[0052] Thus, the local boundary blurriness of each pixel can be determined, which is analyzed based on the characteristics of induced pluripotent stem cells in high-density regions. The higher the value, the more it conforms to the boundary performance of high-density regions, and vice versa, the more it conforms to the boundary performance of low-density regions.

[0053] The cell morphology in different density regions may vary. In high-density regions, due to the crowding between cells, their morphology often becomes flatter or more irregular; while in low-density regions, cells generally maintain a more natural round or oval shape.

[0054] Therefore, the local gray-scale characteristics can be analyzed according to the different cell densities. In high-density regions, the irregular cell morphology leads to a chaotic and uneven gray-scale distribution; while in low-density regions, the cell morphology is more regular and the gray-scale distribution tends to be more orderly and uniform.

[0055] Based on this, according to the distribution of the gray-scale differences between adjacent pixel points in each direction within the second local region of each pixel, the gray-scale uniformity in each direction within the second local region of each pixel is obtained. In this embodiment, the distribution of the gray-scale differences between adjacent pixel points is specifically the standard deviation of the gray-scale differences between all adjacent pixel points in each direction within the second local region of each pixel, and the reciprocal of the standard deviation is used as the gray-scale uniformity in each direction within the second local region of each pixel. Among them, in order to avoid the situation where the denominator is zero, a preset value needs to be added to the denominator, and the preset value is 0.1.

[0056] Furthermore, according to the degree of dispersion of the gray-scale uniformity in multiple directions within the second local region of each pixel, the local morphological regularity of each pixel is obtained. In this embodiment, the coefficient of variation is used to measure the degree of dispersion between multiple variables. Implementers can also choose other ways to measure the degree of dispersion, such as variance, standard deviation, etc. for measurement. This application does not limit this.

[0057] The closer the gray-scale uniformity in each direction is, the more uniform the gray-scale distribution in the second local region of the corresponding pixel is, so the greater the possibility that the pixel is in the low-density region of induced pluripotent stem cells.

[0058] Step 3: According to the local boundary blurriness and local morphological regularity of each pixel, combined with the local brightness performance value, obtain the density comprehensive performance value of each pixel; cluster the pixels based on the density comprehensive performance value to obtain a preset number of clustering clusters; use histogram equalization to enhance each clustering cluster respectively to obtain an enhanced induced pluripotent stem cell image, and use a neural network to detect the density of the induced pluripotent stem cell image.

[0059] Due to the complex actual characteristics of high-density cell distribution, a single factor cannot accurately determine the cell position. To improve the accuracy of analysis, this application starts from two features, namely local boundary blurriness and local morphological regularity, and combines the local brightness performance value to comprehensively analyze each pixel point in order to more accurately identify and determine the distribution of induced pluripotent stem cells.

[0060] Based on the comprehensive brightness performance degree of the local area, when it is larger, more attention is paid to the local morphological regularity; when it is smaller, more attention is paid to the local edge blurriness; based on this, according to the local boundary blurriness and local morphological regularity of each pixel point, combined with the local brightness performance value, the density comprehensive performance value of each pixel point is obtained, and its formula form is: ; where represents the density comprehensive performance value of the i-th pixel point, represents the local brightness performance value of the i-th pixel point, represents the local morphological regularity of the i-th pixel point, represents the local boundary blurriness of the i-th pixel point. Among them, the larger the density comprehensive performance value, the greater the possibility that the pixel point is located in the high-density distribution area, and the smaller the value, the greater the possibility that the pixel point is located in the corresponding low-density distribution area.

[0061] Among them, the schematic diagram for obtaining the density comprehensive performance value is as Figure 2 shown.

[0062] Based on the density comprehensive performance value, all pixel points are clustered to obtain a preset number of clustering clusters, and each clustering cluster represents an area with similar density distribution; histogram equalization is used to enhance each clustering cluster respectively to obtain an enhanced image of induced pluripotent stem cells, and a neural network is used to detect the density of the induced pluripotent stem cell image.

[0063] In this embodiment, the preset number is set to 8, and the implementer can set it according to the actual situation; in this embodiment, the K-means clustering algorithm is used for clustering, and the absolute value of the difference in the density comprehensive performance value between pixel points is used as the metric between pixel points. It should be noted that the K-means algorithm and histogram equalization are both existing well-known technologies, and this application will not elaborate on them; through the clustering algorithm, the induced pluripotent stem cell image can be divided into multiple regions with similar density distributions. Histogram equalization is performed on different regions respectively, and the enhancement effects of each region will be different, making the overall gray distribution of the image more uniform, thereby enhancing some detail information and providing a basis for the accuracy of the subsequent density detection results of the induced pluripotent stem cell pictures using a neural network.

[0064] In this embodiment, a convolutional neural network (CNN) is trained based on the images of induced pluripotent stem cells with known cell density. The image of the induced pluripotent stem cells to be detected is used as the input, and the cell density is obtained by using the trained convolutional layer neural network CNN. The convolutional neural network CNN is a well-known existing technology, and this application will not elaborate on it.

[0065] Based on the same concept as the method embodiment of this application, a computer device for detecting the density of induced pluripotent stem cells is proposed, including:

[0066] An induced pluripotent stem cell image acquisition module for acquiring images of induced pluripotent stem cells;

[0067] An induced pluripotent stem cell image analysis module for presetting a first local area and a second local area for each pixel point with the pixel point as the center, where the second local area includes the first local area; obtaining the local brightness performance value of each pixel point according to the gray-scale distribution of the pixel points in the first local area and the second local area; obtaining the local boundary blur degree of each pixel point according to the gradient distribution characteristics of each pixel point in the preset direction in the second local area; obtaining the local morphological regularity of each pixel point according to the uniformity characteristics of the gray-scale distribution in each direction in the second local area of each pixel point;

[0068] An induced pluripotent stem cell image density detection module for obtaining the density comprehensive performance value of each pixel point according to the local boundary blur degree and the local morphological regularity of each pixel point, combined with the local brightness performance value; clustering the pixel points based on the density comprehensive performance value to obtain a preset number of clustering clusters; respectively enhancing each clustering cluster by using histogram equalization to obtain an enhanced image of the induced pluripotent stem cells, and performing density detection on the enhanced image of the induced pluripotent stem cells.

[0069] Among them, a block diagram of a computer device for detecting the density of induced pluripotent stem cells is as Figure 3 shown.

[0070] In summary, induced pluripotent stem cell images are obtained, and local regions are preset with each pixel point as the center. By analyzing the gray-scale distribution among different local regions, local brightness performance values are obtained to characterize the density distribution of the region where the pixel point is located. Since the density distributions of induced pluripotent stem cells are different, the boundary features of their corresponding regions are also different. Therefore, according to the gradient distribution features of the pixel point in the preset direction within the local region, local boundary blurriness is obtained, which helps to further determine the density magnitude of the position where the pixel point is located based on the blurriness of the boundary. Since the cell morphology of induced pluripotent stem cells is related to density, according to the uniformity features of the gray-scale distribution in each direction within the second local region of each pixel point, local morphological regularity is obtained, and the uniformity features of the gray-scale distribution are used to quantify whether the cell morphology is regular. The density comprehensive performance value of each pixel point is obtained to measure the density characteristics of the local region of the pixel point. Furthermore, a clustering algorithm is used to obtain regions with different density distributions, and histogram equalization is performed on each region for enhancement processing, avoiding situations such as loss of local details and over-enhancement caused by traditional histogram equalization during enhancement, which helps to improve the accuracy of subsequent cell density detection results.

[0071] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions marked in the block may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the descriptions. Sometimes, there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. Each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.

[0072] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for detecting the density of induced pluripotent stem cells, characterized in that, The method includes the following steps: Obtain induced pluripotent stem cell images; Taking each pixel point as the center, preset the first local area and the second local area of each pixel point, where the second local area includes the first local area; according to the gray-scale distribution of the pixel points in the first local area and the second local area of each pixel point, obtain the local brightness performance value of each pixel point; according to the gradient distribution characteristics of each pixel point in the preset direction within the second local area, obtain the local boundary blurriness of each pixel point; according to the uniformity characteristics of the gray-scale distribution in each direction within the second local area of each pixel point, obtain the local morphological regularity of each pixel point; According to the local boundary blurriness and local morphological regularity of each pixel point, combined with the local brightness performance value, obtain the density comprehensive performance value of each pixel point; cluster the pixel points based on the density comprehensive performance value to obtain a preset number of clustering clusters; use histogram equalization to enhance each clustering cluster respectively to obtain an enhanced induced pluripotent stem cell image, and perform density detection on the enhanced induced pluripotent stem cell image; The density comprehensive performance value of each pixel point is obtained, and the specific formula is: ; where represents the density comprehensive performance value of the i-th pixel point, represents the local brightness performance value of the i-th pixel point, represents the local morphological regularity of the i-th pixel point, represents the local boundary blurriness of the i-th pixel point.

2. The method for detecting the density of induced pluripotent stem cells according to claim 1, characterized in that, The obtaining of the local brightness performance value of each pixel point is specifically as follows: Denote the average gray-scale value of all pixel points in the first local area of each pixel point as the first average value; denote the average gray-scale value of all pixel points in the second local area of each pixel point as the second average value; calculate the absolute value of the difference between the first average value and the second average value, and normalize the ratio of the first average value to the absolute value of the difference to obtain the local brightness performance value of each pixel point.

3. The method for detecting the density of induced pluripotent stem cells according to claim 1, characterized in that The obtaining of the local boundary blurriness of each pixel point is specifically as follows: Denote the local boundary blur degree of the \(i\)-th pixel as , and its formula form is: where L represents the number of preset directions; represents the number of pixel points in the second local area in the th direction of the i-th pixel point; represents the set of gradient values of all pixel points in the second local area in the th direction of the i-th pixel point; represents the gradient value of the m-th pixel point in the second local area in the th direction of the i-th pixel point; norm() represents the normalization function; min() represents the minimum value function.

4. The method for detecting the density of induced pluripotent stem cells according to claim 1, characterized in that, The obtaining of the local morphological regularity of each pixel point is specifically as follows: According to the distribution of the gray-scale differences between adjacent pixel points in each direction within the second local area of each pixel point, obtain the gray-scale uniformity in each direction within the second local area of each pixel point; According to the degree of dispersion of the gray-scale uniformity in multiple directions within the second local area of each pixel point, obtain the local morphological regularity of each pixel point.

5. The method for detecting the density of induced pluripotent stem cells according to claim 4, wherein The obtaining of the gray-scale uniformity in each direction within the second local area of each pixel point is specifically as follows: The standard deviation of the gray-scale differences between all adjacent pixel points in each direction within the second local area of each pixel point; use the negative correlation mapping result of the standard deviation as the gray-scale uniformity in each direction within the second local area of each pixel point.

6. The method for detecting the density of induced pluripotent stem cells according to claim 1, wherein The clustering of the pixel points includes: Take the density comprehensive performance values of all pixel points as the input of the clustering algorithm, take the absolute value of the difference between the density comprehensive performance values between pixel points as the metric between pixel points in the clustering algorithm, and output a preset number of clustering clusters.

7. The method for detecting the density of induced pluripotent stem cells according to claim 1, wherein The density detection of the enhanced induced pluripotent stem cell image is specifically as follows: Train a neural network according to the induced pluripotent stem cell image with known cell density, take the induced pluripotent stem cell image to be detected as the input, and use the trained neural network to obtain the cell density.

8. A computer device for detecting the density of induced pluripotent stem cells, which implements the method according to claim 1, characterized in that, It includes: An induced pluripotent stem cell image acquisition module for obtaining induced pluripotent stem cell images; An induced pluripotent stem cell image analysis module is used to preset a first local area and a second local area for each pixel point with the pixel point as the center, where the second local area contains the first local area; obtain the local brightness performance value of each pixel point according to the gray-scale distribution of the pixel points in the first local area and the second local area; obtain the local boundary blur degree of each pixel point according to the gradient distribution characteristics of each pixel point in the preset direction in the second local area; obtain the local morphological regularity of each pixel point according to the uniformity characteristics of the gray-scale distribution in each direction in the second local area of each pixel point. An induced pluripotent stem cell image density detection module is used to obtain the density comprehensive performance value of each pixel point according to the local boundary blur degree and the local morphological regularity of each pixel point, combined with the local brightness performance value; cluster the pixel points based on the density comprehensive performance value to obtain a preset number of clustering clusters; perform histogram equalization on each clustering cluster respectively to obtain an enhanced induced pluripotent stem cell image, and perform density detection on the enhanced induced pluripotent stem cell image.

Images