Peripheral blood cell morphology assisted extraction method under microscopic imaging

Abstract

The present application relates to the technical field of image recognition, and in particular to a peripheral blood cell morphology assisted extraction method under microscopic imaging, obtaining a lymphocyte image, the lymphocyte image containing ordinary lymphocytes, large granular lymphocytes and atypical lymphocytes; based on the distance between the nucleus and cytoplasm of the lymphocyte to be identified, ordinary lymphocytes are screened; according to the density of the lymphocytes around the remaining lymphocytes to be identified and the extrusion of the cytoplasmic outer edge of the remaining lymphocytes to be identified, large granular lymphocytes and atypical lymphocytes are screened, which can improve the accuracy of lymphocyte classification.

Description

Technical Field

[0001] This invention relates to the field of image recognition technology, specifically to a method for assisted extraction of peripheral blood cell morphology under microscopic imaging. Background Technology

[0002] Microscopic imaging is a common method for detecting peripheral blood cells in routine blood tests. Lymphocytes, as common and important cells, are crucial for morphological analysis. Lymphocytes are classified into normal lymphocytes, large granular lymphocytes, and atypical lymphocytes. Lymphocyte differentiation has a wide range of applications, including medical teaching and other uses. However, the small differences between different types of lymphocytes, especially large granular lymphocytes and atypical lymphocytes, can affect the accuracy of lymphocyte classification. Summary of the Invention

[0003] To address the technical problem of low accuracy in lymphocyte classification using existing methods, this invention aims to provide a method for peripheral blood cell morphology-assisted extraction under microscopic imaging. The specific technical solution adopted is as follows:

[0004] In a first aspect of the present invention, a method for morphological-assisted extraction of peripheral blood cells under microscopic imaging is provided, comprising:

[0005] Acquire lymphocyte images, wherein the lymphocytes to be identified in the lymphocyte images include normal lymphocytes, large granular lymphocytes, and atypical lymphocytes;

[0006] Based on the distance between the nucleus and cytoplasm of the lymphocytes to be identified, ordinary lymphocytes are obtained through screening.

[0007] Large granular lymphocytes were selected based on the density of lymphocytes surrounding the remaining lymphocytes to be identified and the compression of the outer edge of the cytoplasm of the remaining lymphocytes to be identified.

[0008] Atypical lymphocytes were obtained from the large granular lymphocytes selected through screening.

[0009] In one exemplary embodiment, the process of screening for ordinary lymphocytes based on the distance between the nucleus and cytoplasm of the lymphocyte to be identified includes:

[0010] Obtain the distances from the outer edge of the nucleus to the outer edge of the cytoplasm in multiple directions, starting from the center of the nucleus of the lymphocyte to be identified, and obtain the mean and variance of the distances;

[0011] Based on the area of ​​the lymphocyte to be identified, as well as the mean distance and the variance of the distance, the first probability that the lymphocyte to be identified belongs to ordinary lymphocytes is obtained; the first probability is inversely proportional to the area, inversely proportional to the variance of the distance, and inversely proportional to the mean distance.

[0012] Ordinary lymphocytes were obtained based on the first probability screening.

[0013] In an exemplary embodiment, the step of selecting large granular lymphocytes based on the density of lymphocytes surrounding the remaining lymphocytes to be identified and the compression of the outer edge of the cytoplasm of the remaining lymphocytes to be identified includes:

[0014] Based on the density of lymphocytes surrounding the remaining lymphocytes to be identified in the first screening, and the compression of the outer edge of the cytoplasm of the remaining lymphocytes to be identified in the first screening, large granular lymphocytes were obtained in the first screening.

[0015] Based on the similarity of the nucleus shape between the remaining lymphocytes to be identified in the second screening and the large granular lymphocytes obtained in the first screening, as well as the cell distance, large granular lymphocytes were obtained in the second screening.

[0016] In an exemplary embodiment, the first screening to obtain large granular lymphocytes based on the density of lymphocytes surrounding the first remaining unidentified lymphocytes and the compression of the outer edge of the cytoplasm of the first remaining unidentified lymphocytes includes:

[0017] Obtain a first number of second lymphocytes to be identified within a preset range around the first lymphocyte to be identified; the first lymphocyte to be identified is any one of the first remaining lymphocytes to be identified, and the second lymphocytes to be identified are other lymphocytes to be identified that were not the first lymphocyte to be identified.

[0018] Obtain a second number of third lymphocytes that are in contact with the first lymphocyte to be identified; the third lymphocytes to be identified are the other lymphocytes remaining from the first lymphocyte to be identified, excluding the first lymphocyte to be identified.

[0019] Based on the first quantity, the second quantity, and the compression of the outer edge of the cytoplasm, a second possibility is obtained that the first lymphocyte to be identified belongs to a large granular lymphocyte.

[0020] Based on the second possibility, determine whether the first lymphocyte to be identified belongs to large granular lymphocytes.

[0021] In one exemplary embodiment, the process of obtaining the second quantity includes:

[0022] The outer edge of the cytoplasm of the first lymphocyte to be identified is segmented according to the gray value to obtain multiple line segments;

[0023] Obtain the second quantity, which is the number of the first line segments, where the first line segments are line segments with a gray value less than a preset threshold.

[0024] In an exemplary embodiment, the process of obtaining the compression status of the outer edge of the cytoplasm of the first remaining lymphocytes to be identified includes:

[0025] Obtain the length of each line segment in each of the remaining unidentified lymphocytes in the second quantity greater than 1, and obtain the average length.

[0026] The standard line segment length is obtained based on the average length value.

[0027] Obtain the length difference between each line segment in the first lymphocyte to be identified and the standard line segment length, and obtain the average length difference corresponding to the first lymphocyte to be identified.

[0028] In one exemplary embodiment, the formula for calculating the second possibility is as follows:

[0029]

[0030] Among them, D i For the i-th remaining lymphocyte to be identified in the first round, N represents the second probability that it belongs to a large granular lymphocyte. i M represents the second number of unidentified lymphocytes remaining after the first identification, i.e., M. i L represents the first number of unidentified lymphocytes remaining in the first instance. i,n Let Li be the length of the nth first segment in the i-th remaining unidentified lymphocyte, and L0 be the standard segment length. To The negative correlation normalization, where norm is the normalization function.

[0031] In an exemplary embodiment, the second screening to obtain large granular lymphocytes, based on the similarity of the nucleus shape of the remaining unidentified lymphocytes in the second screening to the large granular lymphocytes obtained in the first screening, and the cell distance, includes:

[0032] The difference in the roundness of the nucleus shape between the fourth lymphocyte to be identified and the reference large granular lymphocyte is obtained; the fourth lymphocyte to be identified is any one of the remaining lymphocytes to be identified in the second screening, and the reference large granular lymphocyte is any one of the large granular lymphocytes obtained in the first screening.

[0033] Obtain the cell distance between the fourth lymphocyte to be identified and the reference large granular lymphocyte;

[0034] Based on the difference in the roundness of the shape and the cell distance, a reference weight is obtained for the reference large granular lymphocyte to the fourth lymphocyte to be identified. The reference weight is inversely proportional to the difference in the roundness of the shape and inversely proportional to the cell distance.

[0035] The second probability of the reference large granular lymphocyte belonging to large granular lymphocyte is weighted according to the reference weight and the mean is calculated to obtain the third probability of the fourth lymphocyte to be identified belonging to large granular lymphocyte.

[0036] Based on the third possibility, determine whether the fourth lymphocyte to be identified belongs to large granular lymphocytes.

[0037] In an exemplary embodiment, the process of obtaining the difference in the roundness of the shape includes:

[0038] Obtain the circumcircle of the nucleus of the fourth lymphocyte to be identified, and obtain the ratio of the area of ​​the nucleus of the fourth lymphocyte to the area of ​​its corresponding circumcircle, as the degree of roundness of the shape of the fourth lymphocyte to be identified.

[0039] Obtain the circumcircle of the nucleus of a reference large granular lymphocyte, and obtain the ratio of the area of ​​the nucleus of the reference large granular lymphocyte to the area of ​​its corresponding circumcircle, as the roundness of the shape of the reference large granular lymphocyte.

[0040] The difference between the roundness of the shape of the fourth lymphocyte to be identified and the roundness of the shape of the reference large granular lymphocyte is obtained.

[0041] In one exemplary embodiment, the formula for calculating the third possibility is as follows:

[0042]

[0043] Among them, Y j For the j-th remaining unidentified lymphocyte in the second round, the third probability is that it belongs to a large granular lymphocyte. k Let d be the second probability that the k-th reference large granular lymphocyte belongs to the large granular lymphocyte category, where K is the number of reference large granular lymphocytes. j,k M represents the cell distance between the j-th remaining unidentified lymphocyte and the k-th reference large granular lymphocyte. j M represents the roundness of the shape of the j-th remaining unidentified lymphocyte in the second round. k f(d) represents the roundness of the k-th reference large granular lymphocyte. j,k *|Mj -M k |) indicates that d j,k *|M j -M k | negative correlation normalization.

[0044] This invention offers the following advantages: In the peripheral blood cell morphology-assisted extraction method provided by this invention under microscopic imaging, due to the significant shape differences between ordinary lymphocytes, large granular lymphocytes, and atypical lymphocytes, ordinary lymphocytes are first screened based on the distance between the cell nucleus and cytoplasm. Then, large granular lymphocytes and atypical lymphocytes are screened more selectively, reducing interference from ordinary lymphocytes during the screening process. Since large granular lymphocytes and atypical lymphocytes differ in both the density of surrounding lymphocytes and the degree of compression they experience from other cells, large granular lymphocytes are screened based on the density of surrounding lymphocytes and the degree of compression on the outer edge of the cytoplasm of the remaining lymphocytes to be identified, thus obtaining atypical lymphocytes. This method improves the accuracy of lymphocyte classification. Attached Figure Description

[0045] Figure 1 This is a flowchart of a method for assisted extraction of peripheral blood cell morphology under microscopic imaging, provided in one embodiment of the present invention;

[0046] Figure 2 This is a schematic diagram of a lymphocyte image provided in one embodiment of the present invention;

[0047] Figure 3 This is a schematic diagram of various lymphocytes provided in one embodiment of the present invention;

[0048] Figure 4 This is a flowchart of a common lymphocyte recognition process provided in one embodiment of the present invention;

[0049] Figure 5 This is a schematic diagram of the aggregation of large granular lymphocytes provided in one embodiment of the present invention;

[0050] Figure 6 This is a flowchart of the overall screening process for large granular lymphocytes provided in one embodiment of the present invention;

[0051] Figure 7 This is a flowchart of the first screening process for large granular lymphocytes provided in one embodiment of the present invention;

[0052] Figure 8 This is a schematic diagram of the grayscale color of the outer edge of the cytoplasm provided in one embodiment of the present invention;

[0053] Figure 9 This is a flowchart illustrating the process of obtaining the second quantity according to one embodiment of the present invention;

[0054] Figure 10 This is a schematic diagram of the structure of atypical lymphocytes provided in one embodiment of the present invention;

[0055] Figure 11 This is a flowchart illustrating the process of obtaining information about the compression of the outer edge of the cytoplasm, provided in one embodiment of the present invention.

[0056] Figure 12 This is a schematic diagram of the structure of a free large granular lymphocyte provided in one embodiment of the present invention;

[0057] Figure 13 This is a flowchart of the second screening process for large granular lymphocytes provided in one embodiment of the present invention;

[0058] Figure 14 This is a flowchart illustrating the process of obtaining differences in the roundness of shapes according to an embodiment of the present invention. Detailed Implementation

[0059] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the present invention are described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. All data and information collected in this application have been obtained with full consent, and the collection, use, and processing of such information must comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0061] This embodiment provides a method for morphological-assisted extraction of peripheral blood cells under microscopic imaging, such as... Figure 1 As shown, it includes:

[0062] Step 1: Obtain lymphocyte images. The lymphocytes to be identified in the lymphocyte images include normal lymphocytes, large granular lymphocytes, and atypical lymphocytes.

[0063] Step 2: Based on the distance between the nucleus and cytoplasm of the lymphocytes to be identified, ordinary lymphocytes are screened out.

[0064] Step 3: Based on the density of lymphocytes surrounding the remaining lymphocytes to be identified and the compression of the outer edge of the cytoplasm of the remaining lymphocytes to be identified, large granular lymphocytes are selected.

[0065] Step 4: Obtain atypical lymphocytes based on the large granular lymphocytes obtained from the screening.

[0066] The specific implementation process of each step is explained in detail below with reference to the accompanying drawings.

[0067] Step 1: Obtain lymphocyte images. The lymphocytes to be identified in the lymphocyte images include normal lymphocytes, large granular lymphocytes, and atypical lymphocytes.

[0068] In an exemplary embodiment, the patient's blood is analyzed using an automated microscopic blood analyzer, or the patient's blood on a glass slide is examined using an optical microscope to obtain an initial blood image. This image is then preliminarily analyzed to determine the location and number of various peripheral blood cells, and to obtain the location and size of each lymphocyte, i.e., a lymphocyte microscopic distribution map, or lymphocyte image. Figure 2 As shown in the image, the darker-colored large cell areas are lymphocyte areas.

[0069] It should be understood that the initial blood image contains information about various peripheral blood cells. However, due to the significant differences between the various major cell categories in peripheral blood, peripheral blood cells can be more accurately classified using an automated microscopic blood analyzer to obtain a lymphocyte image, which contains only lymphocytes.

[0070] Therefore, in this embodiment, the obtained lymphocyte images are those that have undergone preliminary lymphocyte identification, meaning that other cell types have been filtered out, and the lymphocyte images only contain lymphocytes, facilitating subsequent lymphocyte classification. The lymphocytes to be identified in the lymphocyte images include normal lymphocytes, large granular lymphocytes, and atypical lymphocytes.

[0071] Step 2: Based on the distance between the nucleus and cytoplasm of the lymphocytes to be identified, ordinary lymphocytes are screened out.

[0072] The obtained lymphocyte images show that lymphocytes include three types: normal lymphocytes, large granular lymphocytes, and atypical lymphocytes. Under normal circumstances, they mainly exist as inactive normal lymphocytes, providing long-term immune surveillance and specific immune responses. After viral invasion, normal lymphocytes are activated and differentiate into large granular lymphocytes or atypical lymphocytes, thereby enhancing the immune response, especially in the early and later stages of the immune response against specific pathogens, playing an important role. However, in lymphocyte images, there are significant morphological differences between normal lymphocytes and activated lymphocytes (such as large granular lymphocytes and atypical lymphocytes), and these morphological differences are particularly pronounced due to differences in cell function. Figure 3 The diagram illustrates various types of lymphocytes. Ordinary lymphocytes are typically smaller with thinner, more uniformly distributed cytoplasm, while large granular lymphocytes and atypical lymphocytes are generally larger with thicker, less uniformly distributed cytoplasm. Therefore, when screening lymphocytes in an image, these morphological differences can be used to first identify and distinguish ordinary lymphocytes, thereby effectively extracting large granular lymphocytes and atypical lymphocytes.

[0073] Since all lymphocytes contain cytoplasm and nucleus, the difference lies in the fact that the cytoplasm distribution in all directions of the nucleus of ordinary lymphocytes is relatively thin, while the cytoplasm distribution in all directions of the nucleus of large granular lymphocytes and atypical lymphocytes is relatively thick. Therefore, to distinguish between ordinary lymphocytes, large granular lymphocytes, and atypical lymphocytes, each lymphocyte needs to be divided into two regions: the nucleus and the cytoplasm.

[0074] In one exemplary embodiment, the gray values ​​of the cell nucleus and cytoplasm are different, with the gray value of the nucleus region being smaller and the gray value of the cytoplasm region being larger. Therefore, the lymphocyte is first divided into two regions by edge segmentation. Then, a gray value threshold is preset. By comparing the average gray value of the two regions with the gray value threshold, the region with a gray value less than the gray value threshold, i.e., the region with a smaller gray value, is identified as the nucleus region, and the region with a gray value greater than or equal to the gray value threshold, i.e., the region with a larger gray value, is identified as the cytoplasm region.

[0075] Therefore, based on the distance between the nucleus and cytoplasm of the lymphocytes to be identified, ordinary lymphocytes are screened out. In an exemplary embodiment, such as Figure 4 As shown below, a specific identification process for ordinary lymphocytes is given:

[0076] Step 2-1: Obtain the distances from the outer edge of the nucleus to the outer edge of the cytoplasm in multiple directions, starting from the center of the nucleus of the lymphocyte to be identified, and obtain the mean and variance of the distances.

[0077] A two-dimensional coordinate system is constructed based on the length and width directions of the lymphocyte image. The lymphocyte image is then mapped onto this two-dimensional coordinate system, so that each pixel in the lymphocyte image corresponds to a two-dimensional coordinate point.

[0078] For any lymphocyte to be identified, obtain the center position of the cell nucleus of the lymphocyte to be identified. For example, average the x-coordinate of each position of the cell nucleus of the lymphocyte to be identified and average the y-coordinate of each position of the cell nucleus. The average x-coordinate and the average y-coordinate are used as the coordinates of its center position.

[0079] Starting from the center of the nucleus of the lymphocyte to be identified, rays are emitted in multiple directions around it, which are multiple directions within a 360° radius. The number of directions and their specific angles are set according to the actual situation; for example, 30° is one direction, for a total of 12 directions.

[0080] In each direction, the ray intersects both the outer edge of the nucleus and the outer edge of the cytoplasm of the lymphocyte to be identified. The distance between these two intersection points is then obtained, representing the distance from the outer edge of the nucleus to the outer edge of the cytoplasm in that direction. A distance is obtained for each direction, resulting in multiple distances. The mean and variance of these distances are then calculated to obtain the mean and variance of the distances.

[0081] Step 2-2: Based on the area of ​​the lymphocyte to be identified, as well as the mean and variance of the distance, determine the first probability that the lymphocyte to be identified belongs to a normal lymphocyte.

[0082] Because ordinary lymphocytes are smaller than large granular lymphocytes and atypical lymphocytes, the distance from the outer edge of the nucleus to the outer edge of the cytoplasm is shorter, and the cytoplasm is more evenly distributed, the first probability that a lymphocyte to be identified belongs to the category of ordinary lymphocytes is determined based on the area of ​​the lymphocyte to be identified, as well as the mean and variance of the distance. The smaller the area of ​​the lymphocyte to be identified, the higher the first probability; the smaller the mean distance, the higher the first probability; and the smaller the variance of the distance, the higher the first probability. The area of ​​the lymphocyte to be identified refers to the number of pixels contained within its region.

[0083] In one exemplary embodiment, a specific calculation method for the first possibility is given below:

[0084] P x =1-norm(σ) x *S x *L x );

[0085] Among them, Px S represents the first probability that the xth lymphocyte to be identified belongs to a normal lymphocyte. x Let σ be the area of ​​the x-th lymphocyte to be identified. x Let L be the distance variance of the x-th lymphocyte to be identified. x denoted as the mean distance to the xth lymphocyte to be identified.

[0086] The normalization function is denoted by norm. The normalization method here can be as follows: obtain the maximum and minimum values ​​of the product of the area, distance variance, and distance mean of each lymphocyte to be identified, and then use the maximum and minimum value normalization method to normalize the product of the area, distance variance, and distance mean of each lymphocyte to be identified to 0-1.

[0087] Steps 2-3: Obtain ordinary lymphocytes based on the first probability screening.

[0088] The higher the probability, the more likely the lymphocyte to be identified is to belong to the category of ordinary lymphocytes. In an exemplary embodiment, a threshold value for the probability of ordinary lymphocytes is preset. The value range of the threshold value is 0-1, and it is set according to the actual situation. The higher the threshold value is set, the more stringent the screening of ordinary lymphocytes is. This embodiment takes 0.7 as an example.

[0089] Lymphocytes with a probability greater than or equal to the threshold for common lymphocytes are identified as common lymphocytes. The lymphocytes in the lymphocyte image other than common lymphocytes are defined as the first remaining lymphocytes to be identified. Therefore, the first remaining lymphocytes in the lymphocyte image are large granular lymphocytes and atypical lymphocytes. Large granular lymphocytes and atypical lymphocytes have a high degree of similarity; therefore, after excluding common lymphocytes, further detailed differentiation between large granular lymphocytes and atypical lymphocytes can be performed.

[0090] Step 3: Based on the density of lymphocytes surrounding the remaining lymphocytes to be identified and the compression of the outer edge of the cytoplasm of the remaining lymphocytes to be identified, large granular lymphocytes are selected.

[0091] This step is used to screen large granular lymphocytes and atypical lymphocytes. After obtaining large granular lymphocytes, the remaining lymphocytes are atypical lymphocytes.

[0092] Because large granular lymphocytes and atypical lymphocytes have different functions and durations of action in the immune system, they also differ in lymphocyte imaging. When a virus invades the body, the immune system is rapidly activated, and large granular lymphocytes quickly concentrate in the infected area to rapidly identify and kill infected cells or pathogens. Therefore, in lymphocyte imaging, large granular lymphocytes are mostly in an aggregated state, increasing their likelihood of contact with other cells. Compared to atypical lymphocytes, the cytoplasm of large granular lymphocytes is more transparent.

[0093] like Figure 5 As shown, the circled area is an aggregation region, where large granular lymphocytes exhibit aggregation. Due to the aggregation, the cell membranes of large granular lymphocytes are compressed by surrounding cells, resulting in some deformation. Therefore, based on this characteristic, large granular lymphocytes can be obtained through the first screening, then the large granular lymphocytes in the aggregated cells can be screened out, and finally the free large granular lymphocytes can be screened out.

[0094] In one exemplary embodiment, such as Figure 6 As shown, the screening process for large granular lymphocytes generally includes the following two steps:

[0095] Step 3-1: Based on the density of lymphocytes surrounding the remaining lymphocytes to be identified in the first screening, and the compression of the outer edge of the cytoplasm of the remaining lymphocytes to be identified in the first screening, large granular lymphocytes are obtained.

[0096] In areas where large granular lymphocytes are clustered, the cell density is higher, making them more susceptible to compression by other cells. Because the cytoplasm inside large granular lymphocytes is lighter in color, the edges of the contact areas between them become darker when they are compressed by other cells. Therefore, large granular lymphocytes can be screened based on the areas where they are clustered.

[0097] In one exemplary embodiment, such as Figure 7 As shown, the specific implementation process of the first screening is given below:

[0098] Step 3-1-1: Obtain the first number of second lymphocytes to be identified within a preset range around the first lymphocyte to be identified.

[0099] For ease of explanation, the first lymphocyte to be identified is defined as any one of the remaining lymphocytes from the first identification. The number of remaining lymphocytes from the first identification within a predetermined range surrounding the first lymphocyte to be identified (specifically, the center of the cell nucleus of the first lymphocyte to be identified) is obtained. The size of this predetermined range is set according to actual conditions. In an exemplary embodiment, the radius of the predetermined range is four times the radius of the circumcircle of the first lymphocyte to be identified, and a circle is drawn with the center of the cell nucleus of the first lymphocyte to be identified as the center of the circle; the resulting circular area is the predetermined range surrounding the first lymphocyte to be identified.

[0100] The first remaining lymphocytes within a predetermined range surrounding the first lymphocyte to be identified are designated as the second lymphocyte to be identified. Therefore, both the second and first lymphocytes to be identified are the first remaining lymphocytes to be identified, but the second lymphocyte to be identified is not the same lymphocyte as the first lymphocyte to be identified.

[0101] The number of the first remaining lymphocytes to be identified within a preset range around the first lymphocyte to be identified, i.e., the number of the second lymphocytes to be identified within a preset range around the first lymphocyte to be identified, is defined as the first number.

[0102] Step 3-1-2: Obtain the second number of third lymphocytes that are in contact with the first lymphocyte to be identified.

[0103] As mentioned above, the first lymphocyte to be identified is in contact with and compressed by other cells. Let's define the first remaining lymphocyte in contact with the first lymphocyte as the third lymphocyte to be identified. Therefore, both the third and first lymphocytes are remnants from the first set of lymphocytes to be identified, but they are not the same. Since the third lymphocyte is definitely within a predetermined area surrounding the first lymphocyte, some of the second lymphocytes to be identified may be the third lymphocyte.

[0104] The number of the first remaining lymphocytes that come into contact with the first lymphocyte to be identified, i.e. the number of the third lymphocytes that come into contact with the first lymphocyte to be identified, is defined as the second number.

[0105] In one exemplary embodiment, because the outer edge of the cytoplasm, i.e., the edge of the cell membrane, is an approximately circular line, the color of the edge is darker in the parts that come into contact with and are compressed by other cells, while the color of the edge is lighter in the parts that do not come into contact with and are not compressed by other cells. Figure 8 As shown. Therefore, the second number of remaining lymphocytes in contact with the first lymphocyte to be identified is obtained based on the color difference at the outer edge of the cytoplasm. Therefore, as... Figure 9 As shown, the process of obtaining the second quantity includes:

[0106] Step 3-1-2-1: Divide the outer edge of the cytoplasm of the first lymphocyte to be identified into segments according to the gray value to obtain multiple line segments.

[0107] For the outer edge of the cytoplasm of the first lymphocyte to be identified, starting from any edge pixel, the gray values ​​of each edge pixel are obtained clockwise to obtain a gray value sequence. Then, the difference sequence of the gray value sequence is obtained, where each element of the difference sequence is the absolute value of the difference between the next gray value and the previous gray value. The difference sequence is then curve-fitted to obtain the maxima (peaks) in the difference sequence curve. Using the maxima as the segmentation points of the outer edge of the cytoplasm, the outer edge of the cytoplasm of the first lymphocyte to be identified is segmented to obtain multiple line segments.

[0108] The color of the edges differs depending on whether they are in contact with or compressed by other cells. Therefore, by segmenting, each line segment is divided into two categories: one with a high gray value and one with a low gray value. The category with low gray values ​​represents the parts in contact with other cells, and the number of line segments in the category with low gray values ​​indicates the degree of aggregation of surrounding cells.

[0109] Therefore, the number of contacts between the outer edge of the cytoplasm of the first lymphocyte to be identified and the remaining lymphocytes to be identified in the first round is different, and the outer edge of the cytoplasm of the first lymphocyte to be identified will be divided into a number of line segments of varying depths.

[0110] Step 3-1-2-2: Obtain the second quantity, which is the number of the first line segments. The first line segments are line segments with gray values ​​less than a preset threshold.

[0111] The number of line segments with smaller gray values ​​corresponding to the outer edge of the cytoplasm of the first lymphocyte to be identified is obtained. In an exemplary embodiment, a threshold can be set, the average gray value of each line segment can be obtained, and the average gray value of each line segment can be compared with the preset threshold. The line segments with average gray values ​​less than the preset threshold are defined as the first line segments. Then, the number of the first line segments is used as the second number of third lymphocytes to be identified that are in contact with the first lymphocyte to be identified.

[0112] Using the above process, the second number of unidentified lymphocytes remaining after the first identification is obtained.

[0113] Because the nuclei of atypical lymphocytes are irregularly shaped and tend to be close to the cell membrane, the outer edge of their cytoplasm is darker than other edges. Therefore, in lymphocyte images, this appears as a single edge with a lower gray value, such as... Figure 10 As shown. Atypical lymphocytes are influenced by their nuclei, and the outer edge of their cytoplasm may also have variations in shade. Screening for large granular lymphocytes can easily cause interference. However, due to the uncertainty of the nucleus in atypical lymphocytes, the length of the lower gray-value edge on the outer edge of the cytoplasm is uncertain. Moreover, generally speaking, the outer edge of the cytoplasm of atypical lymphocytes does not have a lower gray-value, or there is only one lower gray-value line segment. Therefore, it is also necessary to obtain the compression condition of the outer edge of the cytoplasm of the first remaining lymphocytes to be identified. In an exemplary embodiment, such as... Figure 11 As shown, the following is a specific acquisition process:

[0114] Step 3-1-2-3: Obtain the length of each line segment in each of the remaining unidentified lymphocytes in the first round when the second number is greater than 1, and obtain the average length.

[0115] To reduce the interference caused by the difference in the depth of the outer edge of the cytoplasm of atypical lymphocytes on the screening of large granular lymphocytes, a second batch of more than one unidentified lymphocytes was obtained from each of the first batch of unidentified lymphocytes. Then, the lengths of each line segment in each of the second batch of more than one unidentified lymphocytes were obtained. Finally, the lengths of each line segment in all the second batch of more than one unidentified lymphocytes were averaged to obtain the mean length.

[0116] Step 3-1-2-4: Obtain the standard line segment length based on the average length.

[0117] The mean length reflects the overall level of the length of the segments in all the first remaining unidentified lymphocytes with a second quantity greater than 1. Therefore, the standard segment length is obtained based on the mean length. In an exemplary embodiment, the obtained mean length is used as the standard segment length.

[0118] Step 3-1-2-5: Obtain the length difference between each line segment in the first lymphocyte to be identified and the standard line segment length, and obtain the mean length difference corresponding to the first lymphocyte to be identified.

[0119] The length difference between each line segment in the first lymphocyte to be identified and the standard line segment length is obtained. In this embodiment, the length difference is the absolute value of the length difference. Then, the mean length difference corresponding to the first lymphocyte to be identified is calculated.

[0120] Step 3-1-3: Based on the first number, the second number, and the compression of the outer edge of the cytoplasm, the second possibility that the first lymphocyte to be identified belongs to a large granular lymphocyte is obtained.

[0121] The larger the first and second numbers, the higher the probability that the first lymphocyte to be identified belongs to the large granular lymphocyte category; the smaller the mean length difference corresponding to the first lymphocyte to be identified, the higher the probability that the first lymphocyte to be identified belongs to the large granular lymphocyte category. In an exemplary embodiment, the formula for calculating the second probability is given below:

[0122]

[0123] Among them, D i For the i-th remaining lymphocyte to be identified in the first round, N represents the second probability that it belongs to a large granular lymphocyte. i M represents the second number of unidentified lymphocytes remaining after the first identification, i.e., M. i L represents the first number of unidentified lymphocytes remaining in the first instance. i,n L0 represents the length of the first segment of the nth remaining unidentified lymphocyte in the i-th first identification, and L0 represents the standard segment length.

[0124] Through N i and M i The number reflects the number of the first segment and the number of surrounding cells in the i-th first remaining unidentified lymphocyte. The greater the number, the greater the influence of the remaining cells on the i-th first remaining unidentified lymphocyte, and the higher the probability that the i-th first remaining unidentified lymphocyte belongs to a large granular lymphocyte.

[0125] By calculating the difference from the standard line segment length, the morphological changes of the edge can be further quantified. To Negative correlation normalization. Since the edges of atypical lymphocytes may be more irregular, leading to greater differences in line segment lengths, the greater the difference from the standard line segment length, the less likely it is to belong to large granular lymphocytes. The negative correlation normalization method in this embodiment can be as follows: obtain the maximum and minimum values ​​of the mean length differences corresponding to each of the first remaining unidentified lymphocytes, and then use the maximum and minimum value normalization method to normalize the length difference corresponding to the i-th first remaining unidentified lymphocyte. Normalize to 0-1, and finally calculate the value 1 minus the normalized result to achieve negative correlation normalization.

[0126] `norm` is the normalization function. The normalization method here can also be: obtain the maximum and minimum values ​​of the sum of the first and second remaining numbers of lymphocytes to be identified, and then use the maximum and minimum value normalization method to normalize the first and second numbers to 0-1.

[0127] Step 3-1-4: Determine whether the first lymphocyte to be identified belongs to large granular lymphocytes based on the second possibility.

[0128] The higher the probability, the greater the likelihood that it belongs to a large granular lymphocyte. Therefore, in an exemplary embodiment, a first large granular lymphocyte probability threshold is preset. The value range of the first large granular lymphocyte probability threshold is 0-1, and it is set according to the actual situation. The higher the first large granular lymphocyte probability threshold is set, the more stringent the screening for large granular lymphocytes that are clustered. In this embodiment, 0.7 is used as an example.

[0129] The first lymphocyte to be identified, corresponding to the second probability threshold greater than or equal to the first large granular lymphocyte probability threshold, is determined as a large granular lymphocyte, thus achieving the first screening of large granular lymphocytes. The lymphocytes remaining after the first screening are designated as the second remaining lymphocytes to be identified. Therefore, the second remaining lymphocytes to be identified in the lymphocyte image are large granular lymphocytes and atypical lymphocytes.

[0130] Step 3-2: Based on the similarity of the nucleus shape between the remaining lymphocytes to be identified in the second screening and the large granular lymphocytes obtained in the first screening, as well as the cell distance, large granular lymphocytes are obtained in the second screening.

[0131] The reason for performing a second screening for large granular lymphocytes is that, in addition to the aggregation of large granular lymphocytes in the lymphocyte images, there are also some free large granular lymphocytes, i.e., large granular lymphocytes that do not exhibit aggregation, such as... Figure 12 As shown, the number of surrounding lymphocytes in these free large granular lymphocytes is relatively small, therefore there is no aggregation in the lymphocyte image, and the first line segment mentioned above does not exist. Thus, the large granular lymphocytes cannot be screened using the screening method given in step 3-1. However, since step 3-1 yielded a second possibility that the first lymphocyte to be identified belongs to the large granular lymphocyte category, and because the nuclei of atypical lymphocytes are more irregularly shaped while the nuclei of large granular lymphocytes are more rounded and regular (see reference for details),... Figure 3The probability that each free lymphocyte belongs to a large granular lymphocyte can be determined by comparing the nucleus of each free lymphocyte with that of the nuclei of lymphocytes identified as large granular lymphocytes, and by considering the distance between the free lymphocyte and its surrounding lymphocytes. Therefore, based on the similarity of the nuclei of the remaining lymphocytes in the second screening to the large granular lymphocytes obtained in the first screening, as well as the cell distance, the second screening identifies large granular lymphocytes.

[0132] In one exemplary embodiment, such as Figure 13 As shown, the specific implementation process of the second screening is given below:

[0133] Step 3-2-1: Obtain the difference in the roundness of the nucleus shape between the fourth lymphocyte to be identified and the reference large granular lymphocyte.

[0134] For ease of explanation, the fourth lymphocyte to be identified is defined as any one of the remaining lymphocytes to be identified in the second screening. The reference large granular lymphocyte is defined as any one of the large granular lymphocytes obtained in the first screening.

[0135] Based on the shapes of large granular lymphocytes and atypical lymphocytes, it is known that the more rounded the nucleus, the more likely it is to be a large granular lymphocyte. Therefore, it is necessary to obtain the difference in the roundness of the nuclei between the fourth lymphocyte to be identified and the reference large granular lymphocytes for use in subsequent screening. In an exemplary embodiment, such as... Figure 14 As shown, the process of obtaining the difference in the roundness of the shape includes:

[0136] Step 3-2-1-1: Obtain the circumcircle of the nucleus of the fourth lymphocyte to be identified, and obtain the ratio of the area of ​​the nucleus of the fourth lymphocyte to the area of ​​its corresponding circumcircle, as the degree of roundness of the shape of the fourth lymphocyte to be identified.

[0137] Step 3-2-1-2: Obtain the circumcircle of the nucleus of the reference large granular lymphocyte, and obtain the ratio of the area of ​​the nucleus of the reference large granular lymphocyte to the area of ​​its corresponding circumcircle, as the roundness of the shape of the reference large granular lymphocyte.

[0138] Step 3-2-1-3: Obtain the difference in the roundness of the fourth lymphocyte to be identified between the roundness of the reference large granular lymphocyte.

[0139] Specifically, circumcircles are constructed for the nuclei of the fourth lymphocyte to be identified and the reference large granular lymphocyte, and the areas of the nuclei of the fourth lymphocyte to be identified, the circumcircles of the nuclei of the fourth lymphocyte to be identified, the nuclei of the reference large granular lymphocyte, and the circumcircles of the nuclei of the reference large granular lymphocyte are obtained. Then, the ratios of the area of ​​the nucleus of the fourth lymphocyte to the area of ​​its circumcircle, and the ratio of the area of ​​the nucleus of the reference large granular lymphocyte to its circumcircle are calculated.

[0140] It should be understood that the larger the area ratio, the smaller the difference between the area of ​​the cell nucleus and its circumcircle, indicating that the two areas are closer, and thus the rounder the cell nucleus and the greater its roundness. Therefore, the ratio of the area of ​​the nucleus of the fourth lymphocyte to be identified to the area of ​​its corresponding circumcircle is used as the roundness of the fourth lymphocyte's shape; the ratio of the area of ​​the nucleus of the reference large granular lymphocyte to the area of ​​its corresponding circumcircle is used as the roundness of the reference large granular lymphocyte's shape.

[0141] The difference between the roundness of the fourth lymphocyte to be identified and the roundness of the reference large granular lymphocyte is specifically the absolute value of the difference between the two roundness values.

[0142] Step 3-2-2: Obtain the cell distance between the fourth lymphocyte to be identified and the reference large granular lymphocyte.

[0143] In an exemplary embodiment, the distance between each edge pixel of the outer edge of the cytoplasm of the fourth lymphocyte to be identified and each edge pixel of the outer edge of the cytoplasm of the reference large granular lymphocyte is obtained, and then the minimum distance among the obtained multiple distances is taken as the cell distance between the fourth lymphocyte to be identified and the reference large granular lymphocyte.

[0144] Step 3-2-3: Based on the differences in the roundness of the shape and the cell distance, obtain the reference weight of the reference large granular lymphocyte to the fourth lymphocyte to be identified.

[0145] The smaller the difference in roundness, the more similar the fourth lymphocyte to be identified is to the reference large granular lymphocyte in terms of roundness. In other words, the more similar their shapes, the more likely they are of the same type of lymphocyte, meaning the fourth lymphocyte is more likely to be a large granular lymphocyte. Since the reference weight of the reference large granular lymphocyte to the fourth lymphocyte represents the probability that the fourth lymphocyte is a large granular lymphocyte, the greater the reference weight of the reference large granular lymphocyte to the fourth lymphocyte. Conversely, the smaller the cell distance between the fourth lymphocyte to be identified and the reference large granular lymphocyte, the greater the potential aggregation of other lymphocytes around the fourth lymphocyte, further increasing the likelihood that the fourth lymphocyte is a large granular lymphocyte. Therefore, the reference weight is inversely proportional to both the difference in roundness and the cell distance.

[0146] By referencing the weights, the smaller the difference in the roundness of the shape, the greater the influence on classifying the fourth lymphocyte to be identified as a large granular cell. The smaller the cell distance, the closer the reference large granular lymphocyte is to the fourth lymphocyte to be identified in space, and the greater the influence on classifying the fourth lymphocyte to be identified as a large granular cell, thus enhancing the influence of spatial aggregation.

[0147] Step 3-2-4: Weight the second probability of the reference large granular lymphocyte belonging to large granular lymphocyte according to the reference weight and calculate the mean to obtain the third probability of the fourth lymphocyte to be identified belonging to large granular lymphocyte.

[0148] The higher the second probability that the reference large granular lymphocyte belongs to the large granular lymphocyte category, the greater its weight is given when determining the probability of whether the fourth lymphocyte to be identified belongs to the large granular lymphocyte category. Therefore, the second probability of the reference large granular lymphocyte belonging to the large granular lymphocyte category is weighted according to the reference weight and the mean is calculated to obtain the third probability that the fourth lymphocyte to be identified belongs to the large granular lymphocyte category.

[0149] In one exemplary embodiment, a specific calculation formula for the third possibility is given below:

[0150]

[0151] Among them, Y j For the j-th remaining unidentified lymphocyte in the second round, the third probability is that it belongs to a large granular lymphocyte. k Let d be the second probability that the k-th reference large granular lymphocyte belongs to the large granular lymphocyte category, where K is the number of reference large granular lymphocytes. j,kM represents the cell distance between the j-th remaining unidentified lymphocyte and the k-th reference large granular lymphocyte. j M represents the roundness of the shape of the j-th remaining unidentified lymphocyte in the second round. k The roundness of the shape of the k-th reference large granular lymphocyte.

[0152] f(d j,k *|M j -M k |) indicates that d j,k *|M j -M k |Negative correlation normalization of f(d) j,k *|M j -M k |) represents the reference weight of the k-th reference large granular lymphocyte relative to the j-th second remaining unidentified lymphocyte. The negative correlation normalization method here can be: obtain the maximum and minimum values ​​among all reference weights, and then, based on the maximum and minimum values, use the maximum-minimum-maximum normalization method to normalize d. j,k *|M j -M k Normalize the value, then subtract the normalized d from the value 1. j,k *|M j -M k |, to achieve d j,k *|M j -M k | negative correlation normalization.

[0153] Step 3-2-5: Determine whether the fourth lymphocyte to be identified belongs to large granular lymphocytes based on the third possibility.

[0154] The higher the probability, the greater the likelihood that it belongs to a large granular lymphocyte. Therefore, in an exemplary embodiment, a second large granular lymphocyte probability threshold is preset. The value range of the second large granular lymphocyte probability threshold is 0-1, and it is set according to the actual situation. The higher the second large granular lymphocyte probability threshold is set, the more stringent the screening for large granular lymphocytes that are clustered. In this embodiment, 0.8 is used as an example.

[0155] The fourth lymphocyte to be identified, which corresponds to the third probability threshold that is greater than or equal to the second large granular lymphocyte probability threshold, is identified as a large granular lymphocyte, thus achieving a second screening of large granular lymphocytes.

[0156] Step 4: Obtain atypical lymphocytes based on the large granular lymphocytes obtained from the screening.

[0157] Step 2 filters to obtain normal lymphocytes, and step 3 filters to obtain large granular lymphocytes. The remaining lymphocytes in the lymphocyte image are then classified as atypical lymphocytes. This process achieves the classification of lymphocytes in the lymphocyte image.

[0158] In subsequent applications, the number of normal lymphocytes, large granular lymphocytes, and atypical lymphocytes in lymphocyte images can be counted, and this data can be integrated with other data analyzed by the automated microscope blood analyzer and uploaded to the backend system for later use.

[0159] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0160] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for assisted extraction of peripheral blood cell morphology under microscopic imaging, characterized by, The method comprises the following steps: acquiring lymphocyte images, wherein lymphocytes to be identified in the lymphocyte images include common lymphocytes, large granular lymphocytes, and atypical lymphocytes; screening common lymphocytes based on the distance between the nucleus and the cytoplasm of the lymphocytes to be identified; screening large granular lymphocytes according to the density of lymphocytes around the remaining lymphocytes to be identified and the compression of the outer edge of the cytoplasm of the remaining lymphocytes to be identified; obtaining atypical lymphocytes according to the screened large granular lymphocytes; the screening of the large granular lymphocytes according to the density of lymphocytes around the remaining lymphocytes to be identified and the compression of the outer edge of the cytoplasm of the remaining lymphocytes to be identified comprises: screening large granular lymphocytes for the first time according to the density of lymphocytes around the remaining lymphocytes to be identified for the first time and the compression of the outer edge of the cytoplasm of the remaining lymphocytes to be identified for the first time; screening large granular lymphocytes for the second time based on the similarity of the nucleus shape of the remaining lymphocytes to be identified for the second time and the large granular lymphocytes screened for the first time and the cell distance; the screening of the large granular lymphocytes for the first time according to the density of lymphocytes around the remaining lymphocytes to be identified for the first time and the compression of the outer edge of the cytoplasm of the remaining lymphocytes to be identified for the first time comprises: acquiring a first number of second lymphocytes to be identified within a preset range around a first lymphocyte to be identified; the first lymphocyte to be identified is any remaining lymphocyte to be identified for the first time, and the second lymphocyte to be identified is any remaining lymphocyte to be identified for the first time except the first lymphocyte to be identified; acquiring a second number of third lymphocytes to be identified in contact with the first lymphocyte to be identified; the third lymphocyte to be identified is any remaining lymphocyte to be identified for the first time except the first lymphocyte to be identified; obtaining a second possibility that the first lymphocyte to be identified belongs to large granular lymphocytes according to the first number, the second number, and the compression of the outer edge of the cytoplasm; judging whether the first lymphocyte to be identified belongs to large granular lymphocytes according to the second possibility; the acquisition process of the second number comprises: segmenting the outer edge of the cytoplasm of the first lymphocyte to be identified according to the gray value to obtain a plurality of line segments; acquiring the second number, which is the number of first line segments, wherein the first line segment is a line segment with a gray value less than a preset threshold.

2. A method of assisted extraction of peripheral blood cell morphology under microscopy as claimed in claim 1, wherein, the screening of the common lymphocytes based on the distance between the nucleus and the cytoplasm of the lymphocytes to be identified comprises: acquiring the distance from the outer edge of the nucleus to the outer edge of the cytoplasm in multiple directions with the center of the nucleus of the lymphocyte to be identified as the starting point, and obtaining the mean distance and the distance variance; obtaining a first possibility that the lymphocyte to be identified belongs to common lymphocytes according to the area of the lymphocyte to be identified and the mean distance and the distance variance; the first possibility is inversely proportional to the area, inversely proportional to the distance variance, and inversely proportional to the mean distance. According to the first possibility, the common lymphocytes are screened.

3. A method of assisted extraction of peripheral blood cell morphology under microscopy as claimed in claim 1, wherein, The process for obtaining the extracellular edge compression condition of the first remaining lymphocytes to be identified comprises: The length of each line segment in each of the second number of first remaining lymphocytes to be identified is obtained, and a length average is obtained; A standard line segment length is obtained according to the length average; The length difference between the length of each line segment in the first lymphocyte to be identified and the standard line segment length is obtained, and a length difference average corresponding to the first lymphocyte to be identified is obtained.

4. A method of assisted extraction of peripheral blood cell morphology under microscopy as claimed in claim 3, wherein, The second possibility is calculated according to the following formula: ； wherein, is a second possibility that the i-th first remaining lymphocyte to be identified belongs to a large granular lymphocyte, is a second number corresponding to the i-th first remaining lymphocyte to be identified, is a first number corresponding to the i-th first remaining lymphocyte to be identified, is a length of the n-th first line segment in the i-th first remaining lymphocyte to be identified, is a standard line segment length, is a negative correlation normalization of is a normalization function.

5. A method of assisted extraction of peripheral blood cell morphology under microscopy as claimed in claim 1 wherein, The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula:

6. A method of assisted extraction of peripheral blood cell morphology under microscopy as claimed in claim 5 wherein, The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula:

7. A method of assisted extraction of peripheral blood cell morphology under microscopy as claimed in claim 5 wherein, The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second possibility is calculated according to the following formula: The second ； wherein, is a third possibility that the jth secondarily remaining lymphocyte to be recognized belongs to large granular lymphocytes, is a second possibility that the kth reference large granular lymphocyte belongs to large granular lymphocytes, and K is the number of reference large granular lymphocytes, is a cell distance between the jth secondarily remaining lymphocyte to be recognized and the kth reference large granular lymphocyte, is a degree of roundness of shape of the jth secondarily remaining lymphocyte to be recognized, is a degree of roundness of shape of the kth reference large granular lymphocyte, denotes a negative correlation normalization of .

Images