Method for measuring local characteristics of tumor tissue by analyzing biological tissue image and computing system for performing same

Abstract

Provided are a method for measuring local characteristics of tumor tissue by analyzing a biological tissue image, and a computing system for performing same. One aspect of the present invention provides a method comprising the steps of: identifying each of tumor cells from a biological tissue image; determining the center point of each of the tumor cells identified from the biological tissue image; measuring characteristic values indicating predetermined characteristics of each of the tumor cells identified from the biological tissue image; constructing a two-dimensional W×H array A corresponding to the biological tissue image; and for each of a plurality of measurement target elements composing at least a portion of elements constituting the two-dimensional array A, calculating a local characteristic measurement value of tumor tissue corresponding to the measurement target element. The step of calculating the local characteristic measurement value of the tumor tissue corresponding to the measurement target element comprises the steps of: collecting surrounding measurement values of the measurement target element; and calculating the local characteristic measurement value of the tumor tissue corresponding to the measurement target element on the basis of the surrounding measurement values of the measurement target element.

Description

Method for measuring regional characteristics of tumor tissue by analyzing biological tissue images and computing system for performing the same

[0001] The present invention relates to a method for measuring regional characteristics of tumor tissue by analyzing a biological tissue image and a computing system for performing the same.

[0002]

[0003] Identifying the characteristics of tumor tissue, such as the distribution of morphological features such as nuclear size and eccentricity of tumor cells, the distribution of immune cells within tumor tissue, and the distribution of cells with specific genetic mutations, is an important task that accounts for a significant portion of pathological diagnosis.

[0004] A representative example of tumor tissue characteristics is the distribution of specific biomarkers within the tissue, and methods for confirming biomarker distribution based on immunohistochemistry (IHC) staining to identify the characteristics of tumor cells are currently being utilized in various ways in many types of cancer, as follows.

[0005] - Confirmation of hormone receptor (ER, PgR) distribution in breast cancer tissue, confirmation of HER2 distribution

[0006] - Confirmation of the distribution of PD-1 / PD-L1 immune checkpoint-related biomarkers in various cancer tissues.

[0007] The IHC staining method is a method that uses a dye that expresses a specific color, such as brown or red, after binding to the target biomarker to be confirmed through an antigen-antibody binding method, so that the distribution of the biomarker can be confirmed with the naked eye. It is a staining method that is mainly used to confirm the distribution of biomarkers along with the immunofluorescence (IF) staining method.

[0008] Current methods for assessing the intra-tissue distribution of biomarkers, a representative example of tumor tissue characteristics, involve measuring the proportion of tumor cells (or total) in which the biomarker is identified, and the intensity of IHC staining, which is determined by the amount of biomarker present. These measurements are then converted into a single number or grade representing the entire tissue, which is then used to predict patient prognosis or treatment response.

[0009] These existing methods have the problem of not considering the local characteristics of biomarker distribution. That is, even if the overall ratio and staining intensity are the same, cases where the biomarker is uniformly identified in a portion of the tumor and cases where the biomarker is unevenly distributed throughout the tumor may have completely different biological characteristics. This also applies to characteristics of tumor tissue other than biomarker distribution (e.g., morphological features such as nuclear size and eccentricity). Therefore, methods that measure and provide regional characteristics of tumor tissue are suitable for precision medicine, providing more detailed biological characteristics than existing methods.

[0010] However, since existing measurement methods are not easy for humans to perform with the naked eye and have low reproducibility problems due to inter-observer variability, it is expected to be very difficult for humans to visually confirm / measure and display regional characteristics of tumor tissue.

[0011] Meanwhile, with the recent advancement of computer performance and image processing and analysis technologies including deep learning, various automatic analysis technologies are being developed, ranging from technologies that analyze biological tissue images to automatically measure the ratio and staining intensity of biomarkers, to technologies that analyze biological tissue images to identify tumor areas, assign histological grades, and predict genetic mutations. Therefore, if these technologies are utilized, it will also be possible to implement a method for automatically measuring regional characteristics of tumor tissue.

[0012]

[0013] The technical problem to be achieved by the present invention is to provide a method for measuring regional characteristics of tumor tissue (e.g., regional characteristics of biomarker distribution, etc.) by analyzing a biological tissue image.

[0014]

[0015] According to one aspect of the present invention, there is provided a method for measuring regional characteristics of tumor tissue by analyzing a biological tissue image, the method comprising: a step in which the computing system identifies each tumor cell from the biological tissue image; a step in which the computing system measures a characteristic value representing a predetermined characteristic of each tumor cell identified from the biological tissue image; a step in which the computing system determines a center point of each tumor cell identified from the biological tissue image; and a step in which the computing system constructs a two-dimensional array A of size W×H corresponding to the biological tissue image, wherein W = [W (0) / Z] and H = [H (0) / Z] and W (0) is the number of pixels in the horizontal direction of the above biological tissue image, and H (0)is the number of pixels in the vertical direction of the biological tissue image, [x] is an integer value approximated by an arbitrary real number x, Z is a natural number greater than or equal to 2 such that when the biological tissue image is divided into Z×Z partial images, no more than 2 center points of tumor cells are included in a single partial image (preferably a power of 2), and A(i, j), which is the (i, j)th element of the two-dimensional array A, corresponds to the (i, j)th partial image when the biological tissue image is divided into Z×Z partial images (1<=i<=W, 1<=j<=H), and when the center position of the tumor cell S is included in the (i, j)th partial image, a characteristic value of the tumor cell S is assigned to A(i, j), and when the center position of the tumor cell is not included in the (i, j)th partial image, a predetermined value indicating that there is no tumor cell is assigned to A(i, j); And a method is provided, wherein the computing system comprises a step of calculating, for each of a plurality of measurement target elements that are at least some of the elements constituting the two-dimensional array A, a regional characteristic measurement value of tumor tissue corresponding to the measurement target element, wherein the step of calculating the regional characteristic measurement value of tumor tissue corresponding to the measurement target element comprises a step of collecting peripheral measurement values ​​of the measurement target element, wherein the peripheral measurement values ​​of the measurement target element are characteristic values ​​assigned to each of the elements of the two-dimensional array A that fall within a predetermined range from the measurement target element; and a step of calculating a regional characteristic measurement value of tumor tissue corresponding to the measurement target element based on the peripheral measurement values ​​of the measurement target element.

[0016] In one embodiment, the step of identifying each tumor cell from the biological tissue image may include the step of determining a tumor area included in the biological tissue image; and the step of identifying each cell from the biological tissue image.

[0017] In one embodiment, the characteristic value of each tumor cell identified from the biological tissue image may be any one of the size of the nucleus of each identified tumor cell, the roundness of the shape of the nucleus, and the eccentricity when the nucleus is approximated as an ellipse.

[0018] In one embodiment, the biological tissue image is an IHC stained tissue image stained by an IHC staining method, and a characteristic value of each tumor cell identified from the IHC stained tissue image is a biomarker expression level of each identified tumor cell, and the step of measuring the characteristic value of each tumor cell identified from the IHC stained tissue image may include the step of extracting a diaminobenzidine staining color value of each pixel constituting the IHC stained tissue image from the IHC stained tissue image or an image obtained by preprocessing the IHC stained tissue image; and the step of calculating a representative value of the diaminobenzidine staining color value of each pixel belonging to an area corresponding to the tumor cell, for each tumor cell identified from the IHC stained tissue image.

[0019] In one embodiment, for each of a plurality of measurement target elements that are at least some of the elements constituting the two-dimensional array A, the step of calculating a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element may include a step of scanning the two-dimensional array A in a sliding window manner, and calculating a regional characteristic measurement value of a tumor tissue corresponding to the window based on a characteristic value assigned to each of the elements within the window when the number of tumor cells included in partial images corresponding to the elements included within the window is greater than a predetermined reference value.

[0020] In one embodiment, each of the plurality of measurement target elements may be elements that include the center position of a tumor cell in a corresponding partial image among the elements constituting the two-dimensional array A.

[0021] In one embodiment, the step of calculating a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element based on the peripheral measurement values ​​of the measurement target element may include a step of calculating an average, variance, standard deviation, entropy, quadratic entropy, or data heterogeneity value thereof of the peripheral measurement values ​​of the measurement target element as the regional characteristic measurement value of the tumor tissue.

[0022] In one embodiment, the method may further include a step of determining a representative value of a regional characteristic measurement value of a biomarker distribution corresponding to each of the plurality of measurement target elements as a representative value representing a regional characteristic of a tumor tissue of the biological tissue image.

[0023] According to another aspect of the present invention, a method is provided, comprising: dividing an entire slide image into a plurality of tiles; applying the method to each of the plurality of divided tiles to determine a representative value representing a regional characteristic of tumor tissue in each of the plurality of tiles; and determining a representative value representing a regional characteristic of biomarker distribution in the entire slide image based on the representative value representing the regional characteristic of tumor tissue in each of the plurality of tiles.

[0024] According to another aspect of the present invention, a computer program is provided, which is installed in a data processing device and recorded on a medium for performing the above-described method.

[0025] According to another aspect of the present invention, a computer-readable recording medium having recorded thereon a computer program for performing the above-described method is provided.

[0026] According to another aspect of the present invention, there is provided a computing system comprising: a processor; and a memory storing a computer program, wherein the computer program, when executed by the processor, causes the computing system to perform the above-described method.

[0027] According to another aspect of the present invention, a computing system for measuring regional characteristics of tumor tissue by analyzing a biological tissue image, comprising: an identification module for identifying each tumor cell from the biological tissue image; a measurement module for measuring a characteristic value representing a predetermined characteristic of each tumor cell identified from the biological tissue image; a judgment module for determining the center point of each tumor cell identified from the biological tissue image; and a construction module for constructing a two-dimensional array A of size W×H corresponding to the biological tissue image, wherein W = [W (0) / Z] and H = [H (0) / Z] and W (0) is the number of pixels in the horizontal direction of the above biological tissue image, and H (0)is the number of pixels in the vertical direction of the biological tissue image, [x] is an integer value approximated by an arbitrary real number x, Z is a natural number greater than or equal to 2 such that when the biological tissue image is divided into Z×Z partial images, no more than 2 center points of tumor cells are included in a single partial image (preferably a power of 2), and A(i, j), which is the (i, j)th element of the two-dimensional array A, corresponds to the (i, j)th partial image when the biological tissue image is divided into Z×Z partial images (1<=i<=W, 1<=j<=H), and when the center position of the tumor cell S is included in the (i, j)th partial image, a characteristic value of the tumor cell S is assigned to A(i, j), and when the center position of the tumor cell is not included in the (i, j)th partial image, a predetermined value indicating that there is no tumor cell is assigned to A(i, j); And, for each of a plurality of measurement target elements that are at least some of the elements constituting the two-dimensional array A, a calculation module (150) that calculates a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element, wherein the calculation module (150) collects peripheral measurement values ​​of the measurement target element, wherein the peripheral measurement values ​​of the measurement target element are characteristic values ​​assigned to each of the elements of the two-dimensional array A that fall within a predetermined range from the measurement target element, and may be a computing system that calculates a regional characteristic measurement value of the tumor tissue corresponding to the measurement target element based on the peripheral measurement values ​​of the measurement target element.

[0028] In one embodiment, the identification module (110) can determine a tumor area included in the biological tissue image and identify each cell from the biological tissue image.

[0029] In one embodiment, the characteristic value of each tumor cell identified from the biological tissue image may be any one of the size of the nucleus of each identified tumor cell, the roundness of the shape of the nucleus, and the eccentricity when the nucleus is approximated as an ellipse.

[0030] In one embodiment, the biological tissue image is an IHC stained tissue image stained by an IHC staining method, and a characteristic value of each tumor cell identified from the IHC stained tissue image is a biomarker expression level of each identified tumor cell, and the measurement module (120) extracts a diaminobenzidine staining color value of each pixel constituting the IHC stained tissue image from the IHC stained tissue image or an image obtained by preprocessing the IHC stained tissue image, and can calculate a representative value of the diaminobenzidine staining color value of each pixel belonging to an area corresponding to the tumor cell for each tumor cell identified from the IHC stained tissue image.

[0031] In one embodiment, the output module (150) scans the two-dimensional array A in a sliding window manner, and when the number of tumor cells included in partial images corresponding to elements included in the window is greater than a predetermined reference value, the output module can output a regional characteristic measurement value of the tumor tissue corresponding to the window based on the characteristic value assigned to each element within the window.

[0032] In one embodiment, each of the plurality of measurement target elements may be elements that include the center position of a tumor cell in a corresponding partial image among the elements constituting the two-dimensional array A.

[0033] In one embodiment, the output module (150) can output the mean, variance, standard deviation, entropy, quadratic entropy, or data heterogeneity values ​​of the surrounding measurement values ​​of the measurement target element as a local characteristic measurement value of the tumor tissue corresponding to the measurement target element.

[0034] The above decision module (160) can determine a representative value of a regional characteristic measurement value of a biomarker distribution corresponding to each of the measurement target elements as a representative value representing the regional characteristic of tumor tissue of the biological tissue image.

[0035] According to another aspect of the present invention, a computing system is provided, including a division module for dividing an entire slide image into a plurality of tiles, and an application module for applying the method to each of the plurality of divided tiles to determine a representative value representing a regional characteristic of biomarker distribution of each of the plurality of tiles, and determining a representative value representing a regional characteristic of tumor tissue of the entire slide image based on the representative value representing the regional characteristic of biomarker distribution of each of the plurality of tiles.

[0036]

[0037] According to the technical concept of the present invention, a method for measuring regional characteristics of tumor tissue by analyzing a biological tissue image can be provided. In particular, according to one embodiment of the present invention, a method for measuring regional characteristics of biomarker distribution can be provided by analyzing an IHC stained tissue image.

[0038] In addition, the method according to the technical idea of ​​the present invention has the effect of providing consistent objective measurement values ​​without the subjective intervention of a judge such as a pathologist.

[0039]

[0040] In order to more fully understand the drawings cited in the detailed description of the present invention, a brief description of each drawing is provided.

[0041] FIG. 1 is a diagram illustrating a schematic system configuration for implementing a method for measuring regional characteristics of tumor tissue according to an embodiment of the present invention.

[0042] FIG. 2 is a flowchart illustrating a method for measuring regional characteristics of tumor tissue according to one embodiment of the present invention.

[0043] Figure 3 is a flowchart illustrating an example of step S110 of Figure 2.

[0044] Figure 4 is a flowchart illustrating an example of step S130 of Figure 2.

[0045] FIGS. 5A to 5C are drawings for explaining an example of a two-dimensional array constructed by a method according to one embodiment of the present invention.

[0046] Figure 6 is a flowchart illustrating an example of step S160 of Figure 2.

[0047] FIG. 7 is a flowchart illustrating a method for measuring regional characteristics of tumor tissue according to another embodiment of the present invention.

[0048] FIG. 8 is a block diagram illustrating a logical configuration of a computing system according to one embodiment of the present invention.

[0049] FIG. 9 is a block diagram illustrating the physical configuration of a computing system according to one embodiment of the present invention.

[0050]

[0051] The present invention is susceptible to various modifications and embodiments. Specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, but rather to encompass all modifications, equivalents, and alternatives falling within the spirit and technical scope of the present invention. In describing the present invention, detailed descriptions of related known technologies will be omitted if they are deemed to obscure the gist of the present invention.

[0052] Terms such as "first," "second," etc. may be used to describe various components, but these components should not be limited by these terms. Terms such as "first," "second," etc. do not denote a particular order and are used only to distinguish one component from another.

[0053] The terminology used in this application is solely for the purpose of describing specific embodiments and is not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise.

[0054] In this specification, terms such as “include” or “have” are intended to specify the presence of a feature, number, step, operation, component, part or combination thereof described in the specification, but should be understood not to exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts or combinations thereof.

[0055] Additionally, in the present specification, when a component "transmits" data to another component, it means that the component may transmit the data directly to the other component, or may transmit the data to the other component via at least one other component. Conversely, when a component "directly transmits" data to another component, it means that the data is transmitted from the component to the other component without going through the other component.

[0056] Hereinafter, the present invention will be described in detail, focusing on embodiments thereof, with reference to the attached drawings. The same reference numerals in each drawing represent the same components.

[0057] FIG. 1 is a diagram illustrating a schematic system configuration for implementing a method for measuring regional characteristics of tumor tissue according to an embodiment of the present invention.

[0058] Referring to FIG. 1, a method for measuring regional characteristics of tumor tissue according to the technical concept of the present invention can be performed by a computing system (100). The computing system (100) can analyze a biological tissue image and measure regional characteristics of tumor tissue appearing in the image.

[0059] A biological tissue image may be a slide image of a pathology specimen stained using a predetermined staining method, or a portion thereof. The pathology specimen may be a biopsy taken from various organs of the human body or a surgically excised biological tissue. Meanwhile, the biological tissue image may be a digital slide image of a stained pathology specimen, or a portion thereof. The pathology specimen slide may be a sliced ​​portion of the pathology specimen. A digital slide image of a pathology specimen can be created by slicing a pathology specimen to create a glass slide, staining the slide with a predetermined staining agent, and digitizing the slide.

[0060] Meanwhile, in one embodiment, the regional characteristics of the tumor tissue measured by the computing system (100) may be, in particular, regional characteristics of the biomarker distribution. In this case, the biological tissue image may be an IHC stained tissue image. The IHC stained tissue image may be a tissue slide image obtained by staining a pathology specimen slide with an immunohistochemical staining reagent, or a tissue slide image obtained by staining with a dye that mixes an immunohistochemical staining reagent and a counterstaining reagent. In this case, the immunohistochemical staining reagent may be a dye in the form of a combination of an antibody targeting a specific biomarker and a chromogen such as diaminobenzidine (DAB). The counterstaining reagent may be a hematoxylin staining reagent.

[0061] That is, the above immunohistochemical staining image may be an image of a pathology slide in which a pathology specimen is made into a slide and stained with a predetermined staining agent (e.g., an immunohistochemical staining agent or a staining agent mixing an immunohistochemical staining agent and a contrast staining reagent), or an image in which the stained pathology slide image is divided into a predetermined size.

[0062] In one embodiment, a computing system (100) may be installed on a predetermined server (10) to implement the technical idea of ​​the present invention. The server (10) refers to a data processing device having a computing capability for implementing the technical idea of ​​the present invention. Generally, a person skilled in the art will be able to easily infer that not only a data processing device that can be accessed by a client through a network, but also any device capable of performing a specific service, such as a personal computer or a mobile terminal, can be defined as a computing device.

[0063] In one embodiment, the computing system (100) needs to determine a tumor area during the process of analyzing a biological tissue image, and at this time, the computing system may utilize a pre-trained deep learning model (200). The deep learning model may be an artificial neural network that receives an entire slide image or a patch obtained by dividing the entire slide image into a certain size as input, and outputs a result value for identifying an area corresponding to a tumor in the input image.

[0064] The above deep learning model (200) is a neural network artificially constructed based on the operating principles of human neurons, including a multilayer perceptron model, and may refer to a set of information expressing a series of design specifications defining an artificial neural network. The above deep learning model (200) may include, but is not limited to, a convolutional neural network (CNN) widely known in the fields of artificial intelligence, machine learning, or deep learning.

[0065] The server (10) including the deep learning model (200) and the computing system (100) may be a diagnostic system for diagnosing a disease or a diagnostic assistance system for providing information necessary for a doctor to diagnose a disease. The deep learning model (200) and the computing system (100) may be provided in a single physical device, i.e., the server (10), but may also be provided in different physical devices depending on the embodiment, and various modifications may be possible as needed, as will be readily apparent to an average expert in the technical field of the present invention.

[0066] In one embodiment, the server (10) and / or the computing system (100) may communicate with a user terminal (20). For example, the user terminal (20) may transmit a biological tissue image to the computing system (100) and may also receive measured / analyzed results from the computing system (100).

[0067] Meanwhile, in FIG. 1, an example is shown in which the computing system (100) is implemented in the form of a subsystem installed in a server (10), but it is obvious that, depending on the embodiment, the computing system (100) may be implemented as an independent single system. In this case, the computing system (100) may be a computing system that is a data processing device having a computing capability for implementing the technical idea of ​​the present invention, and may generally include not only a server that is a data processing device that can be accessed by a client through a network, but also a computing device such as a personal computer or a mobile terminal.

[0068] FIG. 2 is a flowchart illustrating a method for measuring regional characteristics of tumor tissue according to one embodiment of the present invention.

[0069] The computing system (100) can acquire an image of a biological tissue to be analyzed (S100). According to an embodiment, the computing system (100) can acquire the image of the biological tissue from the user terminal (20). Alternatively, the computing system (100) can acquire the image of the biological tissue from the storage device or database of the server (10) or the computing system (100).

[0070] The computing system (100) can identify each tumor cell from the biological tissue image (S110).

[0071] Figure 3 is a flowchart illustrating an example of step S110 of Figure 2.

[0072] Referring to FIG. 3, the computing system (100) can determine a tumor area included in the biological tissue image to identify each tumor cell from the biological tissue image (S111).

[0073] In step S111, the computing system (100) can determine the tumor area included in the biological tissue image by referring to the tumor area information of the biological tissue image annotated in advance by a pathologist, etc.

[0074] Alternatively, the computing system (100) may determine a tumor area using a deep learning model (200). More specifically, the computing system (100) may input the biological tissue image into an input layer of the deep learning model (200) and identify pixels belonging to a tumor area in the biological tissue image based on an output value output from an output layer of the deep learning model (200). At this time, the deep learning model (200) may be an artificial neural network for segmenting an area corresponding to a tumor from an input image, and may be an artificial neural network that has been trained in advance to receive an image through an input layer and output an output value for identifying a tumor area included in the input image on a pixel basis to an output layer.

[0075] Meanwhile, the computing system (100) can identify each cell from the biological tissue image (S112).

[0076] In one embodiment, the computing system (100) may perform a predetermined preprocessing step before proceeding with the subsequent steps for cell identification.

[0077] The computing system (100) can correct color and brightness distortion as one of the preprocessing steps. More specifically, the computing system (100) can selectively perform blank-field correction (Blank-field correction for achieving a uniform white background in brightfield digital photomicrographs, BioTechniques 2007; 42(6): 716, 718, 720.), shading correction (A BaSiC tool for background and shading correction of optical microscopy images, Nat. Comm. 2017; 8: 14836.), etc.

[0078] The computing system (100) may change the color space of the biological tissue image as one of the preprocessing steps. The computing system (100) may change the biological tissue image from RGB space to HSV space, Lab space, YCbCr space, HED (hematoxylin eosin diaminobenzidine) space, etc. In some cases, the computing system (100) may change the RGB space to two or more spaces among HSV space, Lab space, YCbCr space, and HED space, and then combine images belonging to each color channel of the changed spaces.

[0079] The computing system (100) described above may perform preprocessing by applying only one of the previously described color and brightness distortion correction or color space conversion, or in some cases, both. Furthermore, the preprocessing step is not mandatory, and in some embodiments, subsequent steps may be performed without preprocessing.

[0080] In one embodiment, the computing system (100) can identify the cell nucleus region of each cell and segment it from the biological tissue image by using an adaptive thresholding method, such as Otsu's method (A threshold selection method from gray-level histograms, IEEE Trans. Sys. Man. Cyber. 1979; 9(1):62-66), among image processing methods, or by using a deep learning-based region segmentation method, such as U-Net (U-Net: Convolutional Networks for Biomedical Image Segmentation, MICCAI 2015.), or a cell nucleus detection method, such as StarDist (Cell detection with star-convex polygons, MICCAI 2018.).

[0081] If the staining method used to generate the above-described biological tissue image is a method of staining cell membranes or if the biomarker in the above-described biological tissue image is expressed in the cell membrane, the computing system (100) can identify and segment the cell membrane by applying an adaptive thresholding method used to identify / segment the cell nucleus region as described above and then using an image processing technology or deep learning technology to find a closed loop made of pixels.

[0082] Meanwhile, steps S111 and S112 do not necessarily have to be performed sequentially, and may be performed in parallel, or step S112 may be performed before step S111.

[0083] Referring again to FIG. 2, the computing system (100) can determine the center point of each tumor cell identified from the biological tissue image (S120). In other words, the computing system (100) can determine the center points of the area occupied by each cell identified in step S110.

[0084] If the staining method used to generate the above biological tissue image is a method of staining cell nuclei or if the biomarker in the biological tissue image is expressed in the cell nuclei, the computing system (100) can determine the center point of the area corresponding to the cell nucleus as the center point of the cell.

[0085] If the staining method used to generate the above biological tissue image is a method of staining the cell membrane or if the biomarker in the above biological tissue image is expressed in the cell membrane, the center point of the area distinguished by the cell membrane may be determined as the center point of the cell.

[0086] In one embodiment, the computing system (100) can treat the area occupied by each cell as a polygon and determine the center coordinates within the polygon area. In addition, the computing system (100) can determine the center points of the area occupied by each cell through various well-known methods.

[0087] Meanwhile, the computing system (100) can measure the characteristics of each tumor cell identified from the biological tissue image (S130).

[0088] The above characteristic value may be a value indicating a certain characteristic of a tumor cell, and in particular, may be various numerical values ​​that are helpful in pathological judgment of a single tumor cell or a tumor tissue containing the same. The characteristic value may be, for example, the level of expression of a biomarker in each tumor cell. However, the technical idea of ​​the present invention is not limited thereto, and for example, various values ​​expressing the geometric characteristics of a cell (such as the size of the nucleus of each tumor cell, the roundness of the shape of the nucleus, and the eccentricity when the nucleus is approximated as an ellipse) may also be treated as a characteristic value.

[0089] If the above characteristic value is a value expressing the geometric characteristic of the cell (i.e., the size of the nucleus, the circularity of the nucleus shape, the eccentricity when the nucleus is approximated by an ellipse, etc.), the computing system (100) can calculate the characteristic value using a well-known algorithm that obtains the corresponding value (i.e., the size or circularity of the polygon or closed surface, or the eccentricity of the ellipse approximated thereto, etc.) from a polygon or a closed surface. For example, the computing system (100) can calculate the area of ​​a polygon by dividing the polygon into a plurality of basic shapes (e.g., triangles) and adding up the areas of the divided basic shapes. Alternatively, the computing system (100) can calculate the circularity using a technique such as the least squares center method, the minimum area center method, the minimum circumscribed circle / inscribed circle center method, etc. In addition to these, it goes without saying that various well-known techniques can be applied.

[0090] Figure 4 is a flowchart illustrating an example of step S130 of Figure 2. Figure 4 exemplifies a case where the characteristic value is, in particular, a biomarker expression level of a tumor cell. In this case, the biological tissue image may be an IHC stained tissue image stained by an immunohistochemical staining method.

[0091] Referring to FIG. 4, the computing system (100) can preprocess an IHC stained tissue image (S131). The S131 step is not necessarily performed and may be optionally performed depending on the embodiment.

[0092] The computing system (100) can extract the diaminobenzidine staining color value of each pixel constituting the IHC staining tissue image from the IHC staining tissue image or the image obtained by preprocessing the IHC staining tissue image (S132).

[0093] There may be various methods for extracting the diaminobenzidine staining color value of each pixel constituting the IHC staining tissue image. For example, the computing system (100) may convert the signal intensity of each channel in the color space expressing the IHC staining tissue image into optical density and extract the diaminobenzidine staining color value of each pixel constituting the IHC staining tissue image using a color deconvolution-based extraction method (Quantification of histochemical staining by color deconvolution; Anal Quant Cytol Histol 23: 291-299, 2001.; Digital separation of diaminobenzidine-stained tissues via an automatic color-filtering for immunohistochemical quantification, Biomed Opt Express 2015; 6(2): 544-558, etc.).

[0094] Thereafter, the computing system (100) can perform steps S134 and S135 for each tumor cell identified from the IHC stained tissue image.

[0095] At step S134, the computing system (100) can calculate a representative value of the diaminobenzidine staining color value of each pixel belonging to the area corresponding to the tumor cell (S134). Examples of statistics that can be utilized as representative values ​​include the mean, median, and 3 / 4 quartiles.

[0096] In step S135, the computing system (100) can determine a characteristic value (i.e., a biomarker expression level measurement value) of the tumor cell based on a representative value of the diaminobenzidine staining color value of each pixel belonging to an area corresponding to the tumor cell (S135).

[0097] In one embodiment, the computing system (100) can determine the representative value itself calculated in step S134 as a biomarker expression level measurement value of the tumor cells.

[0098] In another embodiment, the computing system (100) may determine the result of converting the representative value produced in step S134 into a value in another unit (e.g., the expression level of the biomarker) as a biomarker expression level measurement value of the tumor cells. For example, the computing system (100) may perform the conversion by utilizing the relationship between the expression level and the representative color values ​​measured in the aforementioned manner from IHC staining result images of one or more cells in which the expression level of the biomarker has been artificially controlled (piecewise linear interpolation, etc.).

[0099] In another embodiment, the computing system (100) may quantize the representative value calculated in step S134 or the value converted into the expression level of the biomarker, etc., according to a predetermined classification standard, and may determine the quantized value as the biomarker expression level measurement value of the tumor cell.

[0100] Referring again to FIG. 2, the computing system (100) can construct a two-dimensional array A of size W×H corresponding to the biological tissue image (S140).

[0101] At this time, W = [W (0) / Z] and H = [H (0) / Z] can be, W (0) is the number of pixels in the horizontal direction of the above biological tissue image, and H (0) is the number of pixels in the vertical direction of the biological tissue image, and [x] is a value obtained by rounding up (up, down, or rounding down) an arbitrary real number x to an integer. Z may be a natural number greater than or equal to 2 such that when the biological tissue image is divided into partial images of size Z×Z, no more than 2 center points of tumor cells are included in a single partial image, and preferably may be a power of 2.

[0102] Meanwhile, A(i,j), which is the (i, j)th element of the two-dimensional array A, corresponds to the (i, j)th partial image when the biological tissue image is divided into partial images each having a size of Z pixels × Z pixels (1<=i<=W, 1<=j<=H), and when the center position of the tumor cell S is included in the (i, j)th partial image, a characteristic value of the tumor cell S is assigned to A(i,j), and when the center position of the tumor cell is not included in the (i, j)th partial image, a predetermined value (e.g., 0 or -1) indicating that there is no tumor cell is assigned to A(i,j).

[0103] Figures 5a to 5c are drawings for explaining an example of a two-dimensional array constructed in the above manner.

[0104] Fig. 5a is a diagram showing a case where steps S110 to S130 of Fig. 2 are performed to identify each tumor cell and its center point from a biological tissue image. In the example of Fig. 5a, the biological tissue image (1) has a horizontal pixel count of W (0), and the number of pixels in the horizontal direction is H (0) . Also, Figure 5a shows a total of 13 tumor cells (c1 to c 13 ) is shown in the identified case. In Fig. 5a, for convenience, each identified tumor cell is depicted as a circle, and the center point of each tumor cell is indicated as a separate dot.

[0105] Figure 5b is a drawing showing an example of equally dividing the biological tissue image (1) of Figure 5a into partial images of size Z×Z. In the example of Figure 5b, W=[W (0) / Z]=8, and H = [H (0) / Z]=4. As explained above, Z is a predetermined natural number greater than or equal to 2, preferably a power of 2, and may be a natural number that ensures that no more than 2 center points of tumor cells are included in one partial image (e.g., 21).

[0106] Fig. 5c is a diagram illustrating a two-dimensional array constructed from the example illustrated in Fig. 5b. The two-dimensional array (2) illustrated in Fig. 5c has a size of 8×4, and each element corresponds to a partial image. For example, the (7, 2)th element of the two-dimensional array (2) illustrated in Fig. 5c corresponds to the partial image (22) of Fig. 5b. Meanwhile, each element of the two-dimensional array (2) illustrated in Fig. 5c is assigned a characteristic value of the corresponding tumor cell if the corresponding partial image includes the center position of the tumor cell, and is assigned a predetermined value (0 in the example of Fig. 5c) indicating that there is no tumor cell if the center position of the tumor cell is not included. Therefore, M of Fig. 5c k is the tumor cell c of Figure 5b k is the characteristic value of (1<=k<=13).

[0107] Referring again to FIG. 2, the computing system (100) can calculate a regional characteristic measurement value of tumor tissue corresponding to each of a plurality of measurement target elements, which are at least some of the elements constituting the two-dimensional array A (S150, S160).

[0108] Figure 6 is a flowchart illustrating an example of step S160 of Figure 2.

[0109] Referring to FIG. 6, the computing system (100) can first collect peripheral measurement values ​​of the measurement target element in order to calculate a regional characteristic measurement value of the tumor tissue corresponding to the measurement target element (S161).

[0110] At this time, the peripheral measurement values ​​of the measurement target element are characteristic values ​​assigned to each element of the two-dimensional array within a predetermined measurement range from the measurement target element. The measurement range may have various shapes depending on the embodiment. For example, the measurement range may be a square with a size of L×L, or a circle with a radius R.

[0111] Thereafter, the computing system (100) can calculate a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element based on the peripheral measurement values ​​of the measurement target element (S162).

[0112] In one embodiment, the computing system (100) can calculate the mean, variance, standard deviation, entropy, quadratic entropy, or data heterogeneity values ​​of the surrounding measurement values ​​of the measurement target element as regional characteristic measurements of the tumor tissue.

[0113] Meanwhile, the method for determining the plurality of measurement target elements, which are at least some of the elements constituting the two-dimensional array, may also vary.

[0114] In one embodiment, the measurement target elements may be determined by a sliding window method. More specifically, the computing system (100) may scan the two-dimensional array by a sliding window method, and when the number of tumor cells included in partial images corresponding to elements included in the window is greater than a predetermined reference value, a regional characteristic measurement value of tumor tissue corresponding to the window may be calculated based on a characteristic value assigned to each element in the window. For example, the computing system (100) may scan the two-dimensional array by moving a window of size L×L by a sliding step of size K (1<=K<=L), and may calculate a regional characteristic measurement value of tumor tissue corresponding to the window.

[0115] In another embodiment, each of the plurality of measurement target elements may be elements among the elements constituting the two-dimensional array, the corresponding partial images of which include the center positions of tumor cells. In the example of FIG. 5b, the plurality of measurement target elements are (1,1), (1,2), (2,1), (2,2), (2, 3), (4,1), (4, 1), (5, 2), (6, 2), (5, 3), (5, 4), (8, 3), (7, 4), the corresponding partial images of which include the center positions of tumor cells. The computing system (100) may collect peripheral measurement values ​​in a certain area (e.g., a square with a size of L×L or a circle with a radius R) from the center position of each tumor cell, and may calculate a regional characteristic measurement value of a tumor tissue corresponding to the center position of each tumor cell based on these.

[0116] Referring back to FIG. 2, the computing system (100) can determine a representative value of a regional characteristic measurement value of tumor tissue corresponding to each of a plurality of measurement target elements as a representative value representing the regional characteristic of tumor tissue in an IHC stained tissue image (S170). At this time, a statistical value such as an average or mode can be used as the representative value. Alternatively, a value obtained by quantizing a statistical value such as an average or mode according to a predetermined classification standard can be used as the representative value.

[0117] Meanwhile, the method described with reference to FIG. 2 so far may be a method performed on a tile that is part of an entire slide image. In this case, the computing system (100) may perform the method described with reference to FIG. 2 on each tile that divides the entire slide image into grids of a certain size, thereby determining a representative value representing the regional characteristics of tumor tissue for each tile, and may measure the regional characteristics of tumor tissue in the entire slide image based on the determined value for each tile. The flowchart thereof is illustrated in FIG. 7.

[0118] Referring to FIG. 7, the computing system (100) can divide an entire slide image stained by a predetermined staining method into a plurality of tiles (S210). In one embodiment, the staining method may be an immunohistochemical staining method.

[0119] Thereafter, the computing system (100) can determine a representative value representing the regional characteristics of the tumor tissue of each of the plurality of tiles by applying the method described with reference to FIG. 2 to each of the plurality of divided tiles (S220, S230).

[0120] Thereafter, the computing system (100) can determine a representative value representing the regional characteristics of the tumor tissue of the entire slide image based on the representative value representing the regional characteristics of the tumor tissue of each of the plurality of tiles (S240). For example, the computing system (100) can determine the most frequently appearing mode value among the representative values ​​representing the regional characteristics of the tumor tissue of each of the plurality of tiles as the representative value representing the regional characteristics of the tumor tissue of the entire slide image.

[0121] FIG. 8 is a block diagram illustrating a logical configuration of a computing system (100) according to one embodiment of the present invention.

[0122] The computing system (100) may refer to a logical configuration equipped with hardware resources and / or software necessary to implement the technical idea of ​​the present invention, and does not necessarily refer to a single physical component or device. That is, the computing system (100) may refer to a logical combination of hardware and / or software provided to implement the technical idea of ​​the present invention, and, if necessary, may be implemented as a set of logical configurations for implementing the technical idea of ​​the present invention by being installed in devices spaced apart from each other and performing their respective functions. In addition, the computing system (100) may refer to a set of configurations that are separately implemented for each function or role for implementing the technical idea of ​​the present invention. Each configuration of the computing system (100) may be located on different physical devices or may be located on the same physical device. In addition, depending on the implementation example, the combination of software and / or hardware constituting each component of the computing system (100) may also be located on different physical devices, and the configurations located on different physical devices may be organically combined with each other to implement each of the modules.

[0123] Furthermore, the term "module" in this specification may refer to a functional and structural combination of hardware for implementing the technical concepts of the present invention and software for operating the hardware. For example, the module may refer to a logical unit of a given code and hardware resources for executing the given code, and it is readily apparent to an average expert in the technical field of the present invention that it does not necessarily refer to physically connected code or a single type of hardware.

[0124] Referring to FIG. 8, the computing system (100) may include an identification module (110), a measurement module (120), a judgment module (130), a construction module (140), an output module (150), and a decision module (160). Depending on the embodiment, the computing system (100) may further include a division module (170) and an application module (180), and in some cases, may further include an output module (190). Depending on the embodiment of the present invention, some of the above-described components may not necessarily correspond to components essential to the implementation of the present invention, and further, depending on the embodiment, the computing system (100) may include more components than this. For example, the computing system (100) may further include a communication module (not shown) for communicating with an external device, a storage module (not shown) for storing data, a control module (not shown) for controlling components and resources of the computing system (100), etc.

[0125] The above identification module (110) can identify each tumor cell from a biological tissue image. To this end, the identification module (110) can determine a tumor area included in the biological tissue image and identify each cell from the biological tissue image.

[0126] The above measurement module (120) can measure the characteristic value of each tumor cell identified from the biological tissue image. In particular, when the biological tissue image is an IHC stained tissue image stained by an immunohistochemical staining method, and the characteristic value is the expression level of a biomarker, the measurement module (120) extracts the diaminobenzidine staining color value of each pixel constituting the IHC stained tissue image from the IHC stained tissue image or an image obtained by preprocessing the IHC stained tissue image, and, for each tumor cell identified from the IHC stained tissue image, can calculate a representative value of the diaminobenzidine staining color value of each pixel belonging to an area corresponding to the tumor cell, and can determine the characteristic value of the tumor cell based on the representative value of the diaminobenzidine staining color value of each pixel belonging to an area corresponding to the tumor cell.

[0127] The above judgment module (130) can determine the center point of each tumor cell identified from the biological tissue image.

[0128] The above construction module (140) can construct a two-dimensional array A of size W×H corresponding to the above biological tissue image. At this time, W = [W (0) / Z] and H = [H (0) / Z] and W (0) is the number of pixels in the horizontal direction of the above biological tissue image, and H (0)is the number of pixels in the vertical direction of the biological tissue image, [x] is a value obtained by rounding up (up, down, or rounding down) an arbitrary real number x to an integer, Z is a predetermined value (preferably a power of 2) such that when the biological tissue image is divided into Z×Z partial images, no more than two center points of tumor cells are included in a single partial image, and A(i,j), which is the (i, j)th element of the two-dimensional array A, corresponds to the (i, j)th partial image when the biological tissue image is divided into Z×Z partial images (1<=i<=W, 1<=j<=H), and when the center position of the tumor cell S is included in the (i, j)th partial image, a characteristic value of the tumor cell S is assigned to A(i,j), and when the center position of the tumor cell is not included in the (i, j)th partial image, a predetermined value indicating that there is no tumor cell is assigned to A(i,j).

[0129] The above-mentioned output module (150) can collect peripheral measurement values ​​of the measurement target element and, based on the peripheral measurement values ​​of the measurement target element, calculate a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element. At this time, the peripheral measurement values ​​of the measurement target element are characteristic values ​​assigned to each element of the two-dimensional array A that falls within a predetermined range from the measurement target element.

[0130] In one embodiment, the output module (150) scans the two-dimensional array A in a sliding window manner, and when the number of tumor cells included in partial images corresponding to elements included in the window is greater than a predetermined reference value, the output module (150) can output a regional characteristic measurement value of tumor tissue corresponding to the window based on the characteristic value assigned to each element within the window.

[0131] Alternatively, according to an embodiment, each of the plurality of measurement target elements may be elements that include the center position of a tumor cell in a corresponding partial image among the elements constituting the two-dimensional array A.

[0132] Meanwhile, in one embodiment, the output module (150) can output the mean, variance, standard deviation, entropy, quadratic entropy, or data heterogeneity values ​​of the surrounding measurement values ​​of the measurement target element as regional characteristic measurement values ​​of tumor tissue corresponding to the measurement target element.

[0133] The above decision module (160) can determine a representative value of a regional characteristic measurement value of tumor tissue corresponding to each of the measurement target elements as a representative value representing the regional characteristic of tumor tissue in the bio-tissue image.

[0134] Meanwhile, the segmentation module (170) can segment the entire slide image stained by a predetermined staining method into a plurality of tiles, and the application module (180) can apply the regional characteristic measurement method of tumor tissue described above to each of the plurality of segmented tiles to determine a representative value representing the regional characteristic of tumor tissue in each of the plurality of tiles, and can determine a representative value representing the regional characteristic of tumor tissue in the entire slide image based on the representative value representing the regional characteristic of tumor tissue in each of the plurality of tiles.

[0135] The above output module (190) can output the measurement / analysis results to a predetermined output device (e.g., a display device, etc.) connected to the computing system (100) or to a predetermined external device (e.g., the user terminal (20)) connected via a network.

[0136] Meanwhile, depending on the implementation example, the computing system (100) may include a processor and a memory that stores a program executed by the processor. In this case, the computing system (100) may have a configuration as illustrated in FIG. 9. FIG. 8 is a diagram illustrating a schematic configuration of a computing system (100) according to an embodiment of the present invention.

[0137] The above computing system (100) may be equipped with a memory (102) in which a program for implementing the technical idea of ​​the present invention is stored, and a processor (101) for executing the program stored in the memory (102).

[0138] An average expert in the technical field of the present invention will be able to easily infer that the processor (101) may be named by various names such as CPU, APU, mobile processor, etc., depending on the implementation example of the computing system (100). The processor may include a single-core CPU or a multi-core CPU. In addition, the computing system (100) may be implemented by organically combining a plurality of physical devices, and in this case, an average expert in the technical field of the present invention will be able to easily infer that the computing system (100) of the present invention may be implemented by providing at least one processor (101) for each physical device.

[0139] The above memory (102) stores the program and may be implemented as any type of storage device that the processor can access to run the program. The memory (102) may include a high-speed random access memory and may also include non-volatile memory such as one or more magnetic disk storage devices, flash memory devices, or other non-volatile solid-state memory devices. Access to the memory by the processor and other components may be controlled by a memory controller. In addition, depending on the hardware implementation example, the memory (102) may be implemented as a plurality of storage devices rather than a single storage device. In addition, the memory (102) may include a temporary storage device as well as a main memory device. In addition, it may be implemented as a volatile memory or a non-volatile memory, and may be defined to mean all types of information storage means that are implemented so that the program can be stored and run by the processor.

[0140] The above program, when executed by the processor (101), can cause the computing system (100) to perform the method for measuring regional characteristics of tumor tissue described above.

[0141] Meanwhile, the method according to an embodiment of the present invention may be implemented in the form of computer-readable program commands and stored on a computer-readable recording medium, and the control program and target program according to an embodiment of the present invention may also be stored on a computer-readable recording medium. A computer-readable recording medium includes all types of recording devices that store data that can be read by a computer system.

[0142] The program commands recorded on the recording medium may be specially designed and configured for the present invention or may be known and available to those skilled in the software field.

[0143] Examples of computer-readable storage media include magnetic media such as hard disks, floppy disks, and magnetic tape; optical media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Additionally, the computer-readable storage media can be distributed over network-connected computer systems so that the computer-readable code can be stored and executed in a distributed manner.

[0144] Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a device that processes information electronically, such as a computer, using an interpreter.

[0145] The hardware device described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

[0146] The foregoing description of the present invention is for illustrative purposes only, and those skilled in the art will readily appreciate that the present invention can be readily modified into other specific forms without altering the technical spirit or essential characteristics of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single entity may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined manner.

[0147] The scope of the present invention is indicated by the claims described below rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention.

[0148]

[0149] The present invention can be used in a method for measuring regional characteristics of tumor tissue by analyzing a biological tissue image and a computing system for performing the same.

Claims

1. A method for measuring regional characteristics of tumor tissue by analyzing a biological tissue image, the method comprising: A step in which the computing system identifies each tumor cell from the biological tissue image; A step in which the computing system determines the center point of each tumor cell identified from the biological tissue image; A step in which the computing system measures a characteristic value representing a predetermined characteristic of each tumor cell identified from the biological tissue image; The step of the computing system constructing a two-dimensional array A of size W×H corresponding to the biological tissue image, where W = [W (0) / Z] and H = [H (0) / Z] and W (0) is the number of pixels in the horizontal direction of the above biological tissue image, and H (0) is the number of pixels in the vertical direction of the biological tissue image, [x] is an integer value approximated by an arbitrary real number x, Z is a natural number greater than or equal to 2 such that when the biological tissue image is divided into Z×Z partial images, no more than 2 center points of tumor cells are included in a single partial image, and the (i, j)th element A(i,j) of the two-dimensional array A corresponds to the (i, j)th partial image when the biological tissue image is divided into Z×Z partial images (1<=i<=W, 1<=j<=H), and when the (i, j)th partial image includes the center position of the tumor cell S, a characteristic value of the tumor cell S is assigned to A(i,j), and when the (i, j)th partial image does not include the center position of the tumor cell, a predetermined value indicating that there is no tumor cell is assigned to A(i,j); and The computing system comprises a step of calculating, for each of a plurality of measurement target elements that are at least some of the elements constituting the two-dimensional array A, a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element, The step of calculating the regional characteristic measurement value of the tumor tissue corresponding to the above measurement target element is: A step of collecting peripheral measurement values ​​of the above measurement target element, wherein the peripheral measurement values ​​of the above measurement target element are characteristic values ​​assigned to each of the elements of the two-dimensional array A within a predetermined range from the measurement target element; and A method comprising the step of calculating a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element based on peripheral measurement values ​​of the measurement target element.

2. In paragraph 1, The step of identifying each tumor cell from the above biological tissue image is: A step of determining a tumor area included in the above biological tissue image; and A method comprising the step of identifying each cell from the above biological tissue image.

3. In paragraph 1, A method in which the characteristic value of each tumor cell identified from the above-mentioned biological tissue image is one of the following: the size of the cell nucleus of each identified tumor cell, the roundness of the cell nucleus shape, and the eccentricity when the cell nucleus is approximated as an ellipse.

4. In paragraph 1, The above biological tissue image is an IHC stained tissue image stained by the IHC (immunohistochemistry) staining method. The characteristic value of each tumor cell identified from the above IHC stained tissue image is the biomarker expression level of each identified tumor cell. The step of measuring the characteristic value of each tumor cell identified from the above IHC stained tissue image is as follows. A step of extracting a diaminobenzidine staining color value of each pixel constituting the IHC staining tissue image from the IHC staining tissue image or an image that has been preprocessed from the IHC staining tissue image; and For each tumor cell identified from the above IHC stained tissue image, A method comprising the step of calculating a representative value of the diaminobenzidine staining color value of each pixel belonging to an area corresponding to the tumor cells.

5. In paragraph 1, For each of a plurality of measurement target elements that are at least some of the elements constituting the above two-dimensional array A, the step of calculating a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element is as follows: A method comprising the step of scanning the two-dimensional array A in a sliding window manner, and calculating a regional characteristic measurement value of a tumor tissue corresponding to the window based on a characteristic value assigned to each element within the window when the number of tumor cells included in partial images corresponding to elements included within the window is greater than a predetermined reference value.

6. In paragraph 1, A method in which each of the above plurality of measurement target elements is an element among the elements constituting the two-dimensional array A, the corresponding partial image of which includes the center position of the tumor cell.

7. In paragraph 1, The step of calculating a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element based on the peripheral measurement values ​​of the measurement target element is as follows. A step of calculating the mean, variance, standard deviation, entropy, quadratic entropy, or data heterogeneity of the surrounding measurement values ​​of the above measurement target element as a regional characteristic measurement value of tumor tissue.

8. In paragraph 1, The above method, A method further comprising the step of determining a representative value of a regional characteristic measurement value of a tumor tissue corresponding to each of the plurality of measurement target elements as a representative value representing the regional characteristic of the tumor tissue of the biological tissue image.

9. Step of dividing the entire slide image into multiple tiles; A step of applying the method of claim 8 to each of the divided plurality of tiles to determine a representative value representing the regional characteristics of the tumor tissue of each of the plurality of tiles; and A method comprising the step of determining a representative value representing a regional characteristic of tumor tissue of the entire slide image based on a representative value representing a regional characteristic of tumor tissue of each of the plurality of tiles.

10. A computer program recorded on a medium for installing in a data processing device and performing the method described in clause 1 or clause 9.

11. A computer-readable recording medium having recorded thereon a computer program for performing the method described in paragraph 1 or paragraph 9.

12. As a computing system, a processor; and a memory storing a computer program, The above computer program, A computing system which, when executed by said processor, causes said computing system to perform the method described in claim 1 or claim 9.

13. A computing system for measuring regional characteristics of tumor tissue by analyzing a biological tissue image, An identification module for identifying each tumor cell from the above biological tissue image; A measurement module for measuring characteristic values ​​representing predetermined characteristics of each tumor cell identified from the above-mentioned biological tissue image; A judgment module for determining the center point of each tumor cell identified from the above biological tissue image; A construction module that constructs a two-dimensional array A of size W×H corresponding to the above biological tissue image, where W = [W (0) / Z] and H = [H (0) / Z] and W (0) is the number of pixels in the horizontal direction of the above biological tissue image, and H (0) is the number of pixels in the vertical direction of the biological tissue image, [x] is an integer value approximated by an arbitrary real number x, Z is a natural number greater than or equal to 2 such that when the biological tissue image is divided into Z×Z partial images, no more than 2 center points of tumor cells are included in a single partial image, and the (i, j)th element A(i,j) of the two-dimensional array A corresponds to the (i, j)th partial image when the biological tissue image is divided into Z×Z partial images (1<=i<=W, 1<=j<=H), and when the (i, j)th partial image includes the center position of the tumor cell S, a characteristic value of the tumor cell S is assigned to A(i,j), and when the (i, j)th partial image does not include the center position of the tumor cell, a predetermined value indicating that there is no tumor cell is assigned to A(i,j); and For each of a plurality of measurement target elements that are at least some of the elements constituting the above two-dimensional array A, a calculation module is included that calculates a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element. The above output module is, Collecting peripheral measurement values ​​of the above measurement target element, wherein the peripheral measurement values ​​of the above measurement target element are characteristic values ​​assigned to each of the elements of the two-dimensional array A within a predetermined range from the measurement target element, A computing system that calculates a regional characteristic measurement value of a tumor tissue corresponding to the measurement target element based on peripheral measurement values ​​of the measurement target element.

14. In paragraph 13, The above identification module, Determine the tumor area included in the above biological tissue image, A computing system for identifying each cell from the above biological tissue image.

15. In paragraph 13, A computing system in which each characteristic value of a tumor cell identified from the above-mentioned biological tissue image is one of the following: the size of the cell nucleus of each identified tumor cell, the roundness of the cell nucleus shape, and the eccentricity when the cell nucleus is approximated as an ellipse.

16. In paragraph 13, The above biological tissue image is an IHC stained tissue image stained by the IHC (immunohistochemistry) staining method. The characteristic value of each tumor cell identified from the above IHC stained tissue image is the biomarker expression level of each identified tumor cell. The above measurement module, Extracting the diaminobenzidine staining color value of each pixel constituting the IHC staining tissue image from the IHC staining tissue image or the image preprocessed from the IHC staining tissue image, For each tumor cell identified from the above IHC stained tissue image, A computing system for calculating a representative value of the diaminobenzidine staining color value of each pixel belonging to an area corresponding to the tumor cell.

17. In paragraph 13, The above output module is, A computing system comprising a step of calculating a regional characteristic measurement value of a biomarker distribution corresponding to a window based on a characteristic value assigned to each element within the window when the number of tumor cells included in partial images corresponding to elements included within the window is greater than a predetermined reference value while scanning the two-dimensional array A in a sliding window manner.

18. In paragraph 13, A computing system in which each of the plurality of measurement target elements is an element that includes a center position of a tumor cell in a corresponding partial image among the elements constituting the two-dimensional array A.

19. In paragraph 13, The above output module is, A computing system that calculates the mean, variance, standard deviation, entropy, quadratic entropy, or data heterogeneity values ​​of the surrounding measurement values ​​of the above measurement target element as a regional characteristic measurement value of tumor tissue corresponding to the above measurement target element.

20. In paragraph 13, The above computing system, A computing system further comprising a decision module that determines a representative value of a regional characteristic measurement value of tumor tissue corresponding to each of the above measurement target elements as a representative value representing the regional characteristic of tumor tissue in the biological tissue image.

21. A splitting module that splits the entire slide image into multiple tiles; and Applying the method of clause 8 to each of the divided plurality of tiles, a representative value representing the regional characteristics of the biomarker distribution of each of the plurality of tiles is determined, A computing system including an application module that determines a representative value representing a regional characteristic of tumor tissue of the entire slide image based on a representative value representing a regional characteristic of biomarker distribution of each of the plurality of tiles.

Images