CN121207795A

CN121207795A - Data acquisition and analysis method and system for immunohistochemical dyeing machine

Info

Publication number: CN121207795A
Application number: CN202511738485.9A
Authority: CN
Inventors: 卜令斌; 张瑞; 刘秀财; 陈环; 胡子超; 祝田
Original assignee: Koran Medical Laboratory Xuzhou Co ltd
Current assignee: Koran Medical Laboratory Xuzhou Co ltd
Priority date: 2025-11-25
Filing date: 2025-11-25
Publication date: 2025-12-26

Abstract

The application relates to the technical field of data processing, and discloses a data acquisition and analysis method and system of an immunohistochemical dyeing machine. Acquiring multicellular cooperative data through neighborhood cell staining intensity association analysis, constructing a signal transduction intensity matrix based on a spatial distribution mode, calculating a micro-environment influence coefficient by utilizing a cascade reaction prediction network, carrying out context correction on single cell staining by combining micro-environment influence, and finally carrying out micro-environment perception analysis on cell populations to obtain a staining analysis result based on cell interaction. According to the application, by constructing a calculation model based on intercellular signal transduction cascade reaction, the problem of lack of microenvironment context information in the existing immunohistochemical analysis is solved, and the biological accuracy of dyeing evaluation is improved.

Description

Data acquisition and analysis method and system for immunohistochemical dyeing machine

Technical Field

The application relates to the technical field of data processing, in particular to a data acquisition and analysis method and system of an immunohistochemical dyeing machine.

Background

The existing data acquisition and analysis method of the immunohistochemical staining machine is mainly based on a technical framework of single-cell independent analysis, acquires staining intensity data of cells through an optical detection system, and then adopts a fixed threshold classification method to classify the cells into different grades such as negative, weak positive, strong positive and the like. The technology is widely applied to biomarker detection in tumor pathological diagnosis, such as protein expression analysis of PD-L1, HER2, ki67 and the like, and provides an important basis for clinical treatment scheme selection. The traditional data processing flow comprises the steps of image acquisition, cell segmentation, feature extraction, classification evaluation and the like, wherein each step is relatively independent and mainly focuses on the staining features of single cells.

The prior art has obvious technical defects, and mainly shows that the space interaction relation among cells and micro-environment influence factors are ignored. Single cell independent analysis methods fail to capture biological interactions such as signal transduction, metabolic regulation, and functional synergy among cells, resulting in lack of microenvironment context information for the analysis results. The fixed threshold classification method does not consider the difference of microenvironments where different cells are located, so that cells under the same staining intensity but different microenvironment conditions are endowed with the same classification result, and the simplified processing mode remarkably reduces the accuracy and biological significance of staining evaluation. Existing data processing algorithms lack quantitative descriptive capabilities for cell population dynamics and tissue heterogeneity, failing to identify functional partitions and interaction modes within the tissue.

Disclosure of Invention

The application provides a data acquisition and analysis method and a system of an immunohistochemical staining machine, which are used for solving the problem of lack of micro-environment context information in the traditional immunohistochemical analysis by constructing a calculation model based on intercellular signal transduction cascade reaction, and improving the biological accuracy of staining evaluation.

In a first aspect, the present application provides a method for collecting and analyzing data of an immunohistochemical staining machine, the method comprising:

performing multicellular cooperative staining data acquisition processing on the immunohistochemical section through neighborhood cell staining intensity correlation analysis to obtain an intercellular staining correlation data set;

carrying out quantitative modeling treatment on the signal transduction intensity between adjacent cells according to the spatial distribution mode in the intercellular staining association data set to obtain a signal transduction intensity matrix;

inputting the signal transmission intensity matrix into a cascade reaction prediction network, and performing cascade calculation processing on an intercellular dyeing state interaction mechanism to obtain a microenvironment influence coefficient;

based on the correction effect of the micro-environment influence coefficient on the target cell staining state, carrying out context correction processing on the single cell staining result to obtain a corrected staining evaluation value;

And performing microenvironment sensing analysis treatment on the cell population staining mode according to the correction staining evaluation value to obtain a staining analysis result based on cell interaction.

Optionally, the performing multi-cell collaborative staining data acquisition processing on the immunohistochemical section through neighborhood cell staining intensity correlation analysis to obtain an intercellular staining correlation data set comprises:

carrying out multi-spectrum staining intensity separation detection treatment on each cell in the immunohistochemical section to obtain a single-cell multi-channel staining intensity vector;

Performing self-adaptive neighborhood radius calculation processing on the cell centroid coordinates according to the DAB dyeing specificity intensity value in the single-cell multichannel dyeing intensity vector to obtain a cell dynamic neighborhood range;

Performing synergy quantification treatment on the correlation of the intercellular staining intensity in the neighborhood based on the dynamic neighborhood range of the cells to obtain an intercellular staining synergy index;

And carrying out correlation intensity standardization treatment on the intercellular staining cooperative index according to the spatial distance weight to obtain an intercellular staining correlation data set.

Optionally, the performing quantitative modeling processing on the signal transduction intensity between adjacent cells according to the spatial distribution pattern in the intercellular staining associated data set to obtain a signal transduction intensity matrix, including:

Performing distance attenuation function calculation processing on the cell space coordinates in the intercellular staining association data set to obtain intercellular distance attenuation weights;

performing directional conduction judgment processing on the intercellular dyeing intensity gradient according to the distance attenuation weight to obtain a signal conduction direction vector;

conducting coefficient quantification treatment is conducted on the neighbor cell-to-cell dyeing influence intensity based on the signal transduction direction vector, and a cell-to-cell signal transduction value is obtained;

And carrying out matrix arrangement processing on the intercellular signal transduction values according to the cell space position index to obtain a signal transduction intensity matrix.

Optionally, inputting the signal transduction intensity matrix into a cascade reaction prediction network, and performing cascade calculation processing on a mechanism of interaction of dyeing states among cells to obtain a microenvironment influence coefficient, where the method includes:

Inputting the signal transmission intensity matrix into a cascade reaction prediction network to perform multi-layer forward propagation calculation processing to obtain a primary influence response value among cells;

carrying out cascade reaction path tracking calculation processing on the primary influence response value to obtain a multi-hop cell influence propagation sequence;

Performing accumulated superposition calculation processing on the cascade reaction intensity based on the multi-hop cell influence propagation sequence to obtain accumulated cascade influence intensity;

And carrying out normalization weight distribution processing on the accumulated cascade influence intensity to obtain a micro-environment influence coefficient.

Optionally, the cascade reaction path tracking calculation processing is performed on the primary influence response value to obtain a multi-hop cell influence propagation sequence, which includes:

performing graph traversal path search processing on the cell node connection relation in the primary influence response value to obtain an intercellular propagation path set;

Hierarchical grading treatment is carried out on the multi-hop propagation distances according to the path length parameters in the intercellular propagation path set, so as to obtain a hierarchical propagation path structure;

performing decremental calculation processing on the influence attenuation coefficient of each propagation path based on the layered propagation path structure to obtain a path attenuation weight sequence;

And carrying out serialization arrangement treatment on the path attenuation weight sequence according to the propagation time sequence order to obtain a multi-hop cell influence propagation sequence.

Optionally, the performing a context correction process on the single cell staining result based on the correction effect of the micro-environmental influence coefficient on the staining state of the target cell to obtain a corrected staining evaluation value includes:

performing weighted fusion calculation processing on the micro-environment influence coefficient and the original staining intensity of the target cell to obtain a micro-environment correction factor;

performing threshold redefinition treatment on the cell staining classification boundary according to the microenvironment correction factors to obtain an individualized staining judgment threshold;

And applying the personalized staining judgment threshold value to the staining state of the target cell to carry out reevaluation treatment to obtain a corrected staining evaluation value.

Optionally, performing microenvironment sensing analysis processing on the cell population staining mode according to the corrected staining evaluation value to obtain a staining analysis result based on cell interaction, including:

Clustering and grouping the correction dyeing evaluation values according to the cell population distribution areas to obtain cell population dyeing mode partitions;

According to the dyeing intensity distribution characteristics in the cell population dyeing mode partitions, carrying out quantitative analysis treatment on the interaction intensity among the partitions to obtain an interaction intensity matrix among the populations;

Comprehensively evaluating the dyeing heterogeneity of the cell population based on the interaction intensity matrix between the populations to obtain a microenvironment heterogeneity index;

And carrying out result integration treatment on the microenvironment heterogeneity index and the cell population staining pattern partition to obtain a staining analysis result based on cell interaction.

In a second aspect, the present application provides an immunohistochemical staining machine data acquisition and analysis system, the immunohistochemical staining machine data acquisition and analysis system comprising:

The acquisition module is used for carrying out multicellular cooperative dyeing data acquisition processing on the immunohistochemical section through neighborhood cell dyeing intensity correlation analysis to obtain an intercellular dyeing correlation data set;

The modeling module is used for carrying out quantitative modeling processing on the signal transmission intensity between adjacent cells according to the spatial distribution mode in the intercellular staining correlation data set to obtain a signal transmission intensity matrix;

The calculation module is used for inputting the signal transmission intensity matrix into a cascade reaction prediction network, and carrying out cascade calculation processing on the intercellular dyeing state interaction mechanism to obtain a micro-environment influence coefficient;

the correction module is used for carrying out context correction processing on the single-cell dyeing result based on the correction effect of the micro-environment influence coefficient on the dyeing state of the target cell to obtain a corrected dyeing evaluation value;

and the analysis module is used for carrying out microenvironment perception analysis treatment on the cell population staining mode according to the correction staining evaluation value to obtain a staining analysis result based on cell interaction.

In a third aspect, an immunohistochemical staining machine data acquisition and analysis device is provided, and the device comprises a memory and at least one processor, wherein instructions are stored in the memory, and the at least one processor calls the instructions in the memory so that the immunohistochemical staining machine data acquisition and analysis device can execute the above-mentioned immunohistochemical staining machine data acquisition and analysis method.

In a fourth aspect, a computer readable storage medium having instructions stored therein, which when run on a computer, cause the computer to perform the above-described immunohistochemical staining machine data collection analysis method is provided.

According to the technical scheme provided by the application, through the technical characteristics of correlation analysis of the dyeing intensity of the neighborhood cells, a cooperative dyeing data acquisition mechanism among cells is established, the technical limitation of traditional single-cell independent analysis is broken through, and immunohistochemical data acquisition is changed from isolated punctiform detection to correlation network detection, so that a multidimensional data set containing intercellular interaction information is obtained. Based on the quantitative modeling processing technical characteristics of the spatial distribution mode in the intercellular staining-associated data set, a signal transmission intensity matrix is constructed, the technical problem that the degree of intercellular interaction cannot be quantitatively described in the prior art is solved, and the biological interaction between cells is converted from qualitative description to accurate numerical expression. The technical characteristics of the cascade reaction prediction network realize the deep modeling of a complex interaction mechanism among cells through multi-layer forward propagation calculation processing, and compared with the traditional linear analysis method, the technical characteristics can capture nonlinear interaction among cells and cascade amplification effect, and the biological authenticity of immunohistochemical analysis is obviously enhanced. The technical characteristics of the context correction processing of the micro-environment influence coefficient realize personalized dyeing evaluation through the correction effect analysis of the target cell dyeing state, overcome the rigidity limitation of a fixed threshold classification method, and enable the dyeing evaluation of each cell to accurately reflect the real expression state of each cell in a specific micro-environment. The technical characteristics of microenvironment perception analysis and processing construct a tissue-level heterogeneity description framework through systematic analysis of cell population staining patterns, fill the blank of the traditional technology in the aspect of tissue integral characteristic analysis, and provide more comprehensive and deep information support for clinical pathological diagnosis.

In the specific application field of immunohistochemical staining machine data acquisition analysis, the core characteristic of the cascade reaction prediction network algorithm is that the multi-layer neural network architecture can learn and model multi-hop propagation paths among cells, and the algorithm characteristic enables a system to identify indirect cell interaction modes which cannot be found by a traditional method, particularly in tumor microenvironment analysis, the algorithm characteristic can reveal complex interrelation among tumor cells, immune cells and stromal cells, and provides more accurate and comprehensive technical support for biomarker evaluation in accurate medical treatment. The multi-stage transmission chain among cells can be systematically identified and quantified through graph traversal and hierarchical processing by the path tracking algorithm characteristics of the multi-hop cell influence transmission sequence, and the unique value of the algorithm characteristics is that the algorithm characteristics can discover the cascade effect and amplification mechanism among cells, so that immunohistochemical analysis is expanded from static staining intensity evaluation to dynamic signal transmission analysis.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained based on these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of an embodiment of a data acquisition and analysis method of an immunohistochemical staining machine according to the present application;

FIG. 2 is a schematic diagram of the results of a cell population staining pattern zoning micro-environmental perception analysis according to an embodiment of the present application;

FIG. 3 is a schematic diagram of an embodiment of a data acquisition and analysis system of an immunohistochemical staining machine according to the present application;

Fig. 4 is a schematic block diagram of a data acquisition and analysis device of an immunohistochemical staining machine in the embodiment of the invention.

Detailed Description

The embodiment of the application provides a data acquisition and analysis method and system for an immunohistochemical staining machine. The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus.

For ease of understanding, the following describes a specific flow of an embodiment of the present application, referring to fig. 1, and an embodiment of a data acquisition and analysis method of an immunohistochemical staining machine according to the embodiment of the present application includes:

S101, performing multicellular cooperative staining data acquisition processing on an immunohistochemical section through neighborhood cell staining intensity correlation analysis to obtain an intercellular staining correlation data set;

step S102, carrying out quantitative modeling treatment on the signal transduction intensity between adjacent cells according to the spatial distribution mode in the intercellular staining correlation data set to obtain a signal transduction intensity matrix;

step S103, inputting the signal transmission intensity matrix into a cascade reaction prediction network, and performing cascade calculation processing on the intercellular dyeing state interaction mechanism to obtain a micro-environment influence coefficient;

Step S104, carrying out context correction processing on a single cell dyeing result based on the correction effect of the micro-environment influence coefficient on the dyeing state of the target cell to obtain a corrected dyeing evaluation value;

and step 105, performing microenvironment sensing analysis processing on the cell population staining pattern according to the corrected staining evaluation value to obtain a staining analysis result based on cell interaction.

It will be appreciated that the execution subject of the present application may be an immunohistochemical staining machine data acquisition and analysis system, and may also be a terminal or a server, which is not limited herein. The embodiment of the application is described by taking a server as an execution main body as an example.

Specifically, the correlation analysis of the staining intensity of the neighborhood cells refers to quantitative detection of the DAB staining specific intensity of each cell by a multispectral separation technology, wherein a single-cell multichannel staining intensity vector comprises intensity values of three channels of RGB and concentration information of DAB specific staining marks, then the neighborhood radius of the cell is dynamically calculated based on the DAB staining intensity values of the cell, the higher the DAB staining intensity, the larger the cell neighborhood radius is because the influence range of high-expression cells on the surrounding environment is wider, then the synergy degree of the staining intensity among the cells is calculated in a determined neighborhood range, the synergy degree is determined by comparing the similarity of the staining patterns of each cell in the neighborhood, the cells with similar staining patterns have higher synergy indexes, finally the synergy indexes are subjected to weight standardization treatment according to the space distance, and the closer the cell weights are, so that an intercellular staining correlation data set is formed, and the data set comprises the correlation intensity information of each cell and the neighborhood cells.

The spatial distribution mode is characterized in that two-dimensional spatial arrangement characteristics of cells in a tissue slice are identified by analyzing coordinate information in a cell-to-cell staining association data set, the signal transduction intensity quantitative modeling firstly needs to calculate a distance attenuation weight between the cells, the attenuation weight is exponentially decreased along with the increase of the distance between the cells, a propagation attenuation rule of biological signals in the tissue is simulated, then gradient changes of the cell staining intensity are analyzed according to the distance attenuation weight, the gradient of the staining intensity reflects the directionality of signal transduction, the gradient directions from high to low of the intensity represent the dominant directions of the signal transduction, a signal transduction direction vector is formed, then the influence intensity between adjacent cells is quantified based on the direction vector, the distance factor and the direction factor are comprehensively considered by the conduction coefficient, the cells which are close in distance and are in the main direction of the signal transduction have higher conduction coefficients, finally, the signal transduction values between the cells are arranged into a matrix form according to a spatial position index, the rows and columns of the matrix respectively correspond to different cells, and the matrix element values represent the signal transduction intensities between the corresponding cells.

The cascade reaction prediction network is a deep neural network specially designed for processing multi-level interaction among cells, the network architecture comprises a plurality of hidden layers for capturing cascade effects of different levels, after a signal transmission intensity matrix is input into the network, a multi-layer forward propagation calculation process is carried out to obtain a primary influence response value among cells, the primary influence response value reflects first-order interaction intensity among directly adjacent cells, then cascade reaction path tracking is carried out on the primary influence response value, path tracking identifies multi-hop propagation paths among cells through a graph traversal algorithm, each path represents an intermediate cell sequence through which a signal propagates from a source cell to a target cell, layering grading is carried out according to path length, the first-hop path is directly adjacent cells, the second-hop path needs to pass through an intermediate cell, a layering propagation path structure is formed by analogy, an influence attenuation coefficient is calculated for each path, the longer attenuation is larger, a path attenuation weight sequence is obtained, then a multi-hop cell influence propagation sequence is formed according to propagation time sequence, accumulation calculation is carried out on cascade reaction intensity based on the sequence, the accumulation process considers all possible propagation paths and the accumulation influence intensities, finally the accumulation effect intensity is normalized according to the propagation sequence, and the environment influence on the micro-state is quantized by the micro-influence environment coefficient.

The context correction processing refers to a process of correcting a staining evaluation result according to the influence degree of a microenvironment where a cell is located, firstly, carrying out weighted fusion calculation on a microenvironment influence coefficient and the original staining intensity of a target cell, integrating the microenvironment information into single-cell staining information through weighted average in the fusion process, dynamically adjusting the weighting coefficient according to the significance of the microenvironment influence, wherein the more significant the microenvironment influence is, the higher the weighting is, obtaining a microenvironment correction factor, the factor reflects the staining state of the cell after the microenvironment influence is considered, then, carrying out threshold redefinition on a staining classification boundary according to the correction factor, the traditional fixed threshold classification method ignores the microenvironment difference, dynamically adjusting a personalized staining judgment threshold according to the microenvironment characteristics of each cell, correspondingly improving the cell threshold in an inhibitory microenvironment, properly reducing the cell threshold in an accelerating microenvironment, finally, applying the personalized judgment threshold to the target cell for reevaluation, and obtaining a corrected staining evaluation value, wherein the value more accurately reflects the real staining state of the cell in the tissue microenvironment compared with the traditional single-cell analysis result.

The micro-environment perception analysis processing firstly carries out clustering grouping on the correction dyeing evaluation value according to the cell population distribution area, a clustering algorithm divides the tissue slice into different functional areas according to the space position and the dyeing characteristics of cells, the cells in each area have similar dyeing modes and micro-environment characteristics to form cell population dyeing mode partitions, then analyzes the interaction intensity among the partitions, quantitatively evaluates the interaction degree among the partitions by calculating the signal transmission intensity and the influence range among the cells of the different partitions to obtain a interaction intensity matrix among the communities, the matrix describes the interaction relation among different cell populations, comprehensively evaluates the dyeing heterogeneity of the cell population based on the interaction intensity matrix, and the heterogeneity evaluation considers the dyeing consistency inside the partitions and the interaction difference among the partitions, the higher the heterogeneity index is, the more obvious the staining difference among different areas in the tissue is, the higher the complexity of the microenvironment is, finally, the heterogeneity index and the staining mode of the cell population are integrated in a partitioning way to form a staining analysis result based on cell interaction, the result not only comprises correction staining evaluation of each cell, but also provides microenvironment heterogeneity information of the whole tissue and the interaction mode among the cell populations, compared with the traditional single cell independent analysis method, the invention can identify synergistic effect and antagonistic effect among cells, find the staining rule hidden in the cell populations, for example, in the lung cancer tissue analysis, the PD-L1 staining intensity of a certain tumor cell is originally in a weak positive boundary, but the correction staining evaluation value of the cell is reclassified to be positive after considering the high expression state and the signaling influence of peripheral immune cells, it was also found that this region forms an immunosuppressive microenvironment with significant interactions between different cell populations.

In a specific embodiment, the process of executing step S101 may specifically include the following steps:

Carrying out synergy quantification treatment on the correlation of the intercellular staining intensity in the neighborhood based on the dynamic neighborhood range of the cells to obtain an intercellular staining synergy index;

Specifically, the multi-spectral staining intensity separation detection treatment refers to a technical process of carrying out spectral separation and quantitative determination on a staining signal of each cell in a tissue slice through an optical detection system of an immunohistochemical staining machine, the treatment firstly irradiates the staining slice by utilizing a multi-wavelength light source, different wavelengths of light and different staining reagents generate specific absorption and reflection, the composite staining signal is decomposed into independent signals of each staining component through a spectral analyzer, a single-cell multi-channel staining intensity vector comprises an optical density value of the cell in three basic channels of red, green and blue and a specific intensity value of a DAB staining mark, the DAB staining specific intensity value reflects the expression abundance of a target protein in the cell, the protein expression level is quantified through measuring the optical density of a DAB chromogenic product, the staining intensity vector form of each cell is a four-dimensional array, and the values respectively correspond to red channel intensity, green channel intensity, blue channel intensity and DAB specific intensity, and the values are directly acquired through an optical density analyzer and stored in a digital manner. The self-adaptive neighborhood radius calculation processing is an algorithm for dynamically determining the influence range of the DAB staining specific intensity value of each cell, the core logic of the algorithm is that the higher the DAB staining intensity is, the higher the concentration of signal molecules secreted by the cell is, the larger the influence range of the cell is on surrounding cells, the centroid coordinates of the cell are firstly extracted as a center point in the calculation process, then the neighborhood radius is determined according to the DAB intensity value of the cell through a linear mapping relation, the numerical range of the DAB intensity value is converted into a spatial distance range by the mapping relation, the cell neighborhood radius with lower intensity value is set as a smaller value, the cell neighborhood radius with higher intensity value is correspondingly increased, the cell dynamic neighborhood range is a circular area with the centroid of the cell as a center and the calculated radius value as a radius, and all cells in the area are considered to be directly influenced by the center cell.

The synergy quantification treatment is a data processing method for evaluating the intercellular functional coordination by comparing the similarity degree of different intercellular staining patterns in the neighborhood, the treatment process firstly identifies all the neighbor cells in the dynamic neighborhood range of each cell, then calculates the correlation between the staining intensity vectors of the central cell and each neighbor cell, the correlation calculation adopts a vector similarity algorithm, the similarity degree is determined by comparing the intensity distribution patterns of two cells on four staining channels, the cells similar in staining patterns have similar protein expression patterns and cellular functional states, the fact that the synergetic relationship exists between the cells is shown, the higher the correlation value is the stronger the synergy degree, the intercellular staining synergetic index is obtained by carrying out weighted average on the correlation values of all the cell pairs in the neighborhood, the weight coefficient is determined according to the spatial distance between the cells, and the closer the cell weights are the higher the spatial neighboring cell interactions are more direct and strong. The correlation intensity normalization processing is a data preprocessing step of normalizing and normalizing the intercellular staining collaborative indexes according to the spatial distance weight, the processing firstly calculates Euclidean distance between each pair of cells, then converts the distance into weight coefficients through a distance attenuation function, the distance attenuation function adopts an exponential attenuation model, the farther the distance is, the more obvious the weight attenuation is, then multiplies the collaborative indexes by the corresponding distance weights to obtain weighted collaborative indexes, finally, the normalization processing is carried out on all the weighted collaborative indexes, the numerical range is uniformly mapped into a range from zero to one, the normalized numerical values form an intercellular staining correlation data set, the data set is stored in a matrix form, the rows and columns of the matrix respectively represent different cells, and the matrix element values represent the normalized correlation intensity between the corresponding cells.

In a specific embodiment, the process of executing step S102 may specifically include the following steps:

performing directional conduction judgment processing on the intercellular staining intensity gradient according to the distance attenuation weight to obtain a signal conduction direction vector;

conducting coefficient quantification treatment is conducted on the neighbor cell-to-cell staining influence intensity based on the signal transduction direction vector, and a cell-to-cell signal transduction value is obtained;

and (3) carrying out matrix arrangement processing on the intercellular signal transduction values according to the cell space position index to obtain a signal transduction intensity matrix.

Specifically, the distance attenuation function calculation processing is a mathematical processing method for simulating a propagation attenuation rule of a biological signal in a tissue based on a space distance between cells, the processing firstly extracts space coordinate information of each cell from a related data set of intercellular staining, the coordinate information comprises an abscissa position of a centroid of the cell in a tissue section, then calculates a Euclidean distance between any two cells, the distance calculation adopts a standard two-dimensional space distance formula, the distance value is substituted into an exponential attenuation function to perform weight calculation, the exponential attenuation function simulates a physical rule that the concentration of the biological signal molecule is exponentially decreased along with the increase of the distance when the biological signal molecule diffuses in the tissue, the lower the intercellular weight is, the weaker the influence degree is indicated, the numerical range of the distance attenuation weight is between zero and one, the weight approaches zero when the distance is zero, and the weight reflects the difference of interaction intensity between cells due to the difference of space positions. The directional conduction determination processing is an algorithm for determining a dominant direction of signal conduction by analyzing an intercellular staining intensity gradient, the processing carries out weighted calculation on a staining intensity difference between cells based on a distance attenuation weight, the staining intensity gradient refers to a change rate and a change direction of staining intensity between adjacent cells, the gradient calculation is determined by comparing the staining intensity difference between a central cell and a neighborhood cell, the gradient direction points to a cell with high staining intensity from a cell with low staining intensity, the potential direction of signal conduction is represented, the directional determination algorithm further considers the regulation effect of the distance weight on the gradient, the contribution of a cell with a smaller distance to the gradient calculation is larger, the contribution of a cell with a longer distance is correspondingly reduced, and a signal conduction direction vector is obtained by unitizing the gradient direction and comprises angle information of the conduction direction and size information of the conduction intensity.

The conduction coefficient quantification treatment is a calculation process for carrying out numerical assessment on the actual influence intensity between cells based on a signal conduction direction vector, the treatment combines the signal conduction direction vector with a biological interaction mode between cells, the calculation of the conduction coefficient considers three main factors, namely, the consistency of the conduction direction, the influence of a conduction distance and the magnitude of the conduction intensity, the consistency of the conduction direction is assessed by comparing the coincidence degree of the predicted signal conduction direction and the actually observed change direction of a dyeing mode, the higher the consistency is the more accurate the predicted signal conduction direction, the conduction coefficient is correspondingly increased, the influence of the conduction distance is embodied by the distance attenuation weight obtained by the previous calculation, the farther the distance is the smaller the conduction coefficient is, the magnitude of the conduction intensity is determined by the magnitude of a dyeing intensity gradient, the stronger the gradient magnitude is the conduction coefficient is correspondingly increased, the signal conduction value between cells is obtained by carrying out weighted fusion calculation on the three factors, and the numerical value quantitatively represents the influence degree of one cell on the dyeing state of the other cell. The matrixing arrangement treatment is a data organization process of orderly arranging signal transmission values among cells according to spatial position indexes of the cells and constructing a matrix structure, wherein the treatment firstly allocates unique position indexes for each cell in a tissue slice, the indexes are numbered according to the spatial distribution sequence of the cells in the slice, a grid numbering mode from left to right and from top to bottom is generally adopted, then the signal transmission values obtained by calculation among each pair of cells are filled in corresponding positions of a matrix, row indexes of the matrix represent signal transmitting cells, column indexes represent signal receiving cells, matrix element values are signal transmission values among corresponding cells, a signal transmission intensity matrix is a square matrix, the dimension of the matrix is equal to the total number of cells in the tissue slice, diagonal elements of the matrix are usually set to be zero, and because the transmission values of the cells have no practical significance, and non-diagonal elements reflect the mutual influence relationship among different cells.

In Ki67 immunohistochemical staining analysis of colorectal cancer tissues, two adjacent tumor cells are respectively positioned at a coordinate position A and a position B, the distance between cells is obtained by calculating Euclidean distance between the two points, the distance is substituted into an exponential decay function to calculate a distance decay weight, the weight reflects the influence degree of the space distance on the interaction between cells, ki67 staining intensity difference of the two cells is analyzed, the staining intensity of the cell A is lower, the staining intensity of the cell B is higher, a staining intensity gradient is obtained by calculating the intensity difference and the direction, the gradient direction points to the cell B from the cell A, the dominant direction of signal conduction is represented, the gradient is regulated by combining the distance decay weight to obtain a signal conduction direction vector, the vector contains comprehensive information of the conduction direction and the conduction intensity, the conduction coefficient is further calculated based on the direction vector, three factors of the direction consistency, the distance influence and the intensity are comprehensively considered, the signal conduction value of the cell A on the cell B is obtained, the influence degree of proliferation state of the cell A on the proliferation activity of the cell B is quantitatively represented, the signal conduction value between all the cell A is directly represented according to the space position index of the cell B, the signal conduction matrix is directly solved, the signal conduction relation between the cells is not influenced by the whole tissue, and the signal interaction between the cells is directly solved by the signal matrix, and the signal interaction is directly ignored.

In a specific embodiment, the process of executing step S103 may specifically include the following steps:

Specifically, the cascade reaction prediction network is a deep neural network architecture specially used for processing a multistage interaction relation among cells, the network comprises an input layer, a plurality of hidden layers and an output layer, the network is specially designed for capturing the cascade effect among cells, the multilayer forward propagation calculation processing refers to a process of taking a signal transmission intensity matrix as input data and carrying out layer-by-layer calculation through each layer of the neural network, the input layer receives values in the signal transmission intensity matrix, each matrix element represents a pair of direct signal transmission intensities among cells, the input layer converts the matrix data into a vector format required by network calculation, the first hidden layer carries out linear transformation and nonlinear activation processing on the input vector, the linear transformation is realized through matrix multiplication operation of a weight matrix and the input vector, nonlinear activation function introduces nonlinear characteristics, so that the network can learn a complex cell interaction mode, the second hidden layer continues to carry out similar processing on the output of the first layer, higher-level characteristic information is extracted layer by layer, finally, the output layer generates a primary influence response value among cells, the response value reflects the direct interaction intensity among cells after the network is considered, and compared with the primary influence on the signal transmission intensity and the primary influence value is more linear. The cascade reaction path tracking calculation process is an algorithm for identifying and analyzing multi-hop propagation paths among cells based on primary influence response values, the process models interaction relations among cells as directed graph structures, each cell serves as a node in a graph, the primary influence response values serve as side weights among the nodes, the path tracking algorithm systematically searches all possible propagation paths from any source cell to target cells by adopting graph traversal technology, one-hop paths represent the source cells to directly influence the target cells, two-hop paths represent the source cells to influence the target cells through one middle cell, so that the multi-hop propagation paths are formed by pushing, the algorithm records cell sequences and corresponding influence intensity values of each path in the path searching process, the path length limit is set according to biological rationality, and is generally not more than five hops because the overlong propagation paths have weak influence in a biological system, the multi-hop cell influence propagation sequences comprise complete propagation chains from the source cells to the target cells, and each sequence records the identification and the propagation intensity of each cell on the propagation paths.

The cumulative superposition calculation processing is a data processing method for comprehensively calculating the influence intensities of a plurality of propagation paths, the processing firstly calculates the total influence intensity of the paths for each propagation path, the total influence intensity of the paths is obtained by carrying out continuous multiplication or weighted average on the propagation intensities of all sections on the paths, the continuous multiplication operation simulates the gradual attenuation of signals in the propagation process, the weighted average operation considers the relative importance of different propagation sections, then the superposition is carried out on all path intensities between the same pair of source cells and target cells, the superposition mode comprises direct summation or weighted summation, the weighting coefficient is determined according to the path length, the shorter path has higher weight because of higher propagation efficiency, the cumulative cascade influence intensity comprehensively reflects the total influence degree of the source cells on the target cells, the value considers the direct influence and the indirect influence through intermediate cells, and compared with the simple direct influence analysis, the cumulative influence intensity more comprehensively describes the complex interaction relation between the cells. The normalization weight distribution process is a data normalization step of converting the cumulative cascade influence intensity into a normalized micro-environment influence coefficient, the process firstly calculates statistical characteristics of the cumulative influence intensity among all cells, including a maximum value, a minimum value, a mean value and a standard deviation, then maps the influence intensity value into a standard interval from zero to one by adopting a maximum and minimum normalization method, the normalization formula subtracts the minimum value from the original value and divides the value range to ensure that all the coefficients are compared under the same scale, the weight distribution process adjusts according to the relative position and the functional importance of each cell in the tissue, the weight of the cells in the central area is generally higher than that of the cells in the edge area, because the central cells are influenced by more neighbor cells, the micro-environment influence coefficient finally forms a normalized numerical matrix, and the matrix element represents the comprehensive influence degree of the corresponding cells under the micro-environment.

In HER2 immunohistochemical staining analysis of gastric cancer tissue, signal transmission intensity matrix contains direct conduction relation among all cells, after the matrix is input into cascade reaction prediction network, the network recognizes complex interaction mode among cells through multilayer calculation processing, the output primary influence response value not only keeps direct interaction information, but also fuses indirect relevance learned by the network, path tracking is carried out based on the response values, a tumor cell with high expression of HER2 is found, not only directly influences adjacent tumor cells, but also indirectly influences cell populations with farther distance through middle matrix cells, path tracking algorithm recognizes a plurality of propagation paths including two-hop paths and three-hop paths, each path records cell propagation sequence and corresponding influence intensity, cumulative superposition calculation synthesizes the influence intensities of all paths, the cumulative influence intensity of the high expression cells on the long-distance cell population is found to be obviously higher than that of the simple direct influence, a significant cascade amplification effect exists among cells, finally, the cumulative influence intensity is converted into a microenvironment influence coefficient through normalization processing, the comprehensive influence degree of each cell under the microenvironment is quantitatively described, and the technical problem that the cell interaction in the traditional cascade analysis cannot be solved.

In one embodiment, the performing step for performing the cascade path-tracing calculation processing on the primary influence response value may specifically include the following steps:

Hierarchical grading treatment is carried out on the multi-hop propagation distances according to path length parameters in the intercellular propagation path set, so that a hierarchical propagation path structure is obtained;

and carrying out serialization arrangement treatment on the path attenuation weight sequence according to the propagation time sequence order to obtain the multi-hop cell influence propagation sequence.

Specifically, the graph traversal path search process is an algorithm for constructing a cell connection graph based on primary influence response values and systematically searching all possible propagation paths, the process converts the primary influence response values among cells into a directed graph structure, each cell serves as a node in the graph, the primary influence response values serve as directed edge weights of connection nodes, the cell node connection relation is determined through threshold screening, when the influence response values among two cells exceed a preset threshold, connection is established, the graph traversal algorithm adopts a strategy of combining depth-first search and breadth-first search, all available target cell nodes are recursively accessed from each source cell node, the algorithm records each propagation path in the search process, the path comprises a starting cell, an intermediate cell sequence and a terminating cell, the search depth limit is set to be the maximum five layers, the long-distance propagation path which is excessively complicated is avoided, the cell-to-cell propagation path set comprises all possible propagation sequences from any source cell to any target cell, each path records the cell identifier and the total length of the path set, and the data structure of the path set is stored in an adjacent table form, so that subsequent path analysis and processing operations are convenient. The hierarchical processing is a data structuring method for classifying and organizing a path set according to the hop length of a propagation path, the processing firstly counts the length parameter of each propagation path, the path length is equal to the number of cell nodes in the path minus one, namely the propagation hop number, then distributes all paths into different hierarchical levels according to the path length, one-hop paths are classified into a first layer and represent propagation among directly adjacent cells, two-hop paths are classified into a second layer and represent indirect propagation through one middle cell, the hierarchical propagation path structure is stored by adopting a multi-layer nested data structure, each layer comprises all propagation paths with the length, the intra-layer paths are ordered according to the influence intensity from high to low, the inter-layer paths are ordered according to the path length from short to long, and the structure is convenient for subsequent path analysis and weight calculation according to the hierarchy.

The decreasing calculation process is an algorithm for calculating attenuation weights of each path based on the principle that biological signals are attenuated step by step in the multi-hop propagation process, the process calculates the influence attenuation coefficients of each path in the layered propagation path structure, the influence of the path length on the signal propagation effect is considered in the calculation of the attenuation coefficients, the longer the path is, the more serious the signal attenuation is, the attenuation calculation adopts an exponential attenuation model, the attenuation coefficients are equal to the path length power of an attenuation constant, the attenuation constant is set according to biological experience, the attenuation coefficients of different path lengths are usually valued between zero point five and zero point eight, the difference of the attenuation coefficients of the paths of different paths is obvious, the attenuation coefficients of the paths of one hop are highest, the attenuation coefficients of the paths of two hops are about half to three quarters of the paths of one hop, the three-hop paths are further decreased, the attenuation coefficients of each propagation path are recorded by a path attenuation weight sequence, the weight values in the sequence reflect the contribution degree of the paths to the final influence effect, the calculation of the attenuation weights also considers the product effect of the propagation intensities of the paths, and the propagation paths with weak strength can further reduce the effective weight of the whole path. The sequencing arrangement treatment is a data arrangement method for reorganizing a path attenuation weight sequence according to the time sequence of signal propagation, the treatment firstly determines the propagation time sequence of each propagation path, the propagation time sequence is determined based on the path length and the propagation direction, the paths with the same length are ordered according to the priority of the initial cells, the priority is comprehensively determined according to the dyeing intensity and the space position of the cells, the cells with high dyeing intensity and the cells positioned in the central area have higher priority, then all the paths are arranged according to the propagation time sequence to form an ordered propagation sequence, each element in the sequence comprises a path identifier, a propagation cell sequence, attenuation weights and time sequence positions, and the multi-hop cell influence propagation sequence reflects the complete time-space dynamics process of signal propagation among cells.

In a specific embodiment, the process of executing step S104 may specifically include the following steps:

Performing weighted fusion calculation processing on the micro-environment influence coefficient and the original staining intensity of the target cells to obtain a micro-environment correction factor;

Performing threshold redefinition treatment on the cell staining classification boundary according to the microenvironment correction factors to obtain a personalized staining judgment threshold;

Specifically, the weighted fusion calculation process is a mathematical operation process of integrating the micro-environment influence coefficient with the original staining intensity of the target cell, the process firstly extracts the original staining intensity value of the target cell, the value is directly derived from the optical density detection result of the immunohistochemical staining machine, the expression abundance of the target protein in the cell is reflected, then the micro-environment influence coefficient corresponding to the cell is obtained, the coefficient is calculated through the cascade reaction analysis, the comprehensive influence degree of the surrounding cell on the target cell is quantized, the weighted fusion adopts a linear weighted average method, the original staining intensity is multiplied by the original weight coefficient, the micro-environment influence coefficient is multiplied by the environment weight coefficient, then the two weighted results are added to obtain a fusion value, the distribution of the weight coefficient is determined according to the significance degree of the micro-environment influence, the environment weight is increased when the micro-environment influence coefficient is higher, the original weight is occupied when the micro-environment influence is weaker, the sum of the weight coefficient is always equal to the numerical stability of the fusion result, the micro-environment correction factor comprehensively reflects the actual staining state of the cell after the micro-environment influence is considered, and the real expression level of the cell in the tissue micro-environment is more accurately described compared with the original staining intensity. The threshold redefinition treatment is an algorithm for carrying out personalized adjustment on traditional fixed dyeing classification boundaries based on microenvironment correction factors, the treatment firstly analyzes the distribution characteristics of the microenvironment correction factors, calculates the mean value, variance, quantile and other statistical parameters of the correction factors, the traditional dyeing classification adopts a fixed threshold to divide cells into three types of negative, weak positive and strong positive, the fixed threshold ignores the adjustment effect of different microenvironments on cell expression, the personalized threshold adjustment is dynamically set according to the microenvironment characteristics of each cell, the classification threshold is properly reduced when the cells are in an accelerating microenvironment, the cells which are in a boundary state are more easily classified as positive, the classification threshold is correspondingly improved when the cells are in an inhibiting microenvironment, the cell error to be inhibited is prevented from being classified as strong positive, the personalized dyeing judgment threshold is calculated by analyzing the deviation between the correction factors and the traditional classification result, the cell with larger deviation needs more remarkable threshold adjustment, and the cell threshold adjustment amplitude with smaller deviation is relatively smaller.

The re-evaluation treatment is a decision process of applying a personalized staining judgment threshold value to target cells for classification re-judgment, the treatment compares micro-environment correction factors of each cell with the corresponding personalized judgment threshold value, when the correction factors exceed the personalized threshold value, the cells are reclassified into higher staining grades, when the correction factors are lower than the personalized threshold value, the cells are reclassified into lower staining grades, the re-classification treatment adopts a multi-level threshold system, the multi-level system comprises a negative and weak positive demarcation threshold value and a strong positive demarcation threshold value, each demarcation threshold value is subjected to personalized adjustment according to micro-environment characteristics of the cells, the classification result not only comprises the staining grade but also comprises a confidence score, the confidence is determined according to the difference between the correction factors and the threshold value, the larger the classification confidence is higher, the corrected staining evaluation value is the final classification result after re-evaluation, and the numerical value combines the reclassified staining grade and the corresponding confidence score to form a more accurate staining state description than the traditional single cell analysis.

In a specific embodiment, the process of executing step S105 may specifically include the following steps:

Clustering and grouping treatment is carried out on the correction dyeing evaluation values according to the cell population distribution areas, so that cell population dyeing mode partitions are obtained;

Specifically, the clustering grouping process is a data mining method for classifying cells in a tissue slice based on correction staining evaluation values, the process firstly extracts correction staining evaluation values of each cell as clustering features, the correction staining evaluation values comprehensively reflect staining intensity and microenvironment influence of the cells, then a multidimensional feature vector is constructed by combining spatial coordinate information of the cells, the feature vector comprises three dimensions of staining intensity, spatial position and microenvironment coefficient, a K-means clustering method is adopted by a clustering algorithm, the cells with similar features are classified into the same category through iterative calculation, the optimal clustering number is automatically determined by the algorithm in the clustering process, clustering effects are evaluated through a contour coefficient and intra-category inter-category distance ratio, a clustering number setting range is generally between three and eight, corresponding to different cell function groups, cell group distribution areas are spatially divided on the tissue slice according to clustering results, each area comprises cell groups with similar staining patterns, the cell group dyeing patterns partition reflects functional heterogeneous distribution inside the tissue, the different partitions represent cell groups with different biological features, and partition boundaries are determined through spatial continuity and staining similarity double constraint. The quantitative analysis treatment is a numerical evaluation method for calculating interaction intensity among different cell population partitions, the treatment firstly calculates the distribution characteristics of the staining intensity of each partition, the distribution characteristics comprise four statistical parameters of mean value, variance, skewness and kurtosis, the mean value reflects the average expression level of cells in the partition, the variance reflects the uniformity degree of expression, the skewness describes the asymmetry of the distribution, the kurtosis describes the sharpness degree of the distribution, then analyzes the difference of the staining intensity distribution among different partitions, the difference is quantified by the distance measurement among the distributions, the difference comprises the mean value difference, the distribution shape difference and the overlap degree difference, then calculates the interaction intensity among the partitions based on the signal transduction analysis result, the interaction intensity is determined by counting the number of signal transduction paths and the conduction intensity of the cross-partition, the larger the conduction path among the partitions is, the interaction intensity among the clusters is higher, the interaction intensity matrix records the interaction degree among any two partitions, the rows and the columns of the matrix respectively correspond to different partitions, and the element values of the matrix represent the interaction intensity among the corresponding partitions.

The comprehensive evaluation processing is an analysis method for quantitatively describing the complexity and heterogeneity of the whole tissue microenvironment based on an inter-population interaction intensity matrix, global characteristic parameters of the interaction matrix are calculated, the global characteristic comprises linear algebraic characteristics such as a spectral radius, a condition number, a rank and the like of the matrix, the spectral radius reflects the strongest interaction intensity, the condition number reflects the stability of interaction, the matrix rank reflects the number of independent interaction modes, then the heterogeneity degree of the matrix is analyzed, the heterogeneity is evaluated by calculating standard deviation and variation coefficients of matrix elements, the greater the standard deviation is to indicate that the interaction difference among different subareas is more remarkable, the variation coefficients reflect the degree of relative difference, then network connectivity indexes are calculated, the interaction matrix is regarded as a weighted undirected graph, the connectivity degree, the clustering coefficient, the average path length and the like of the graph are calculated, the connectivity degree reflects the density of interaction among subareas, the average path length reflects the information propagation efficiency, the microenvironment heterogeneity index is obtained by carrying out weighted comprehensive calculation on the plurality of parameters, and the higher the weight coefficient is to indicate that the heterogeneity of the tissue heterogeneity index is higher according to the importance of each parameter. The result integration treatment is a data synthesis method for carrying out association mapping on a micro-environment heterogeneity index and a cell population staining pattern partition, the heterogeneity index is used as an organization overall characteristic parameter, association analysis is carried out on the heterogeneity index and local characteristics of each partition, association analysis is carried out to identify which partitions have the largest contribution to overall heterogeneity and which interactions among the partitions are most obvious, then a multi-level analysis result structure is constructed, the structure comprises three layers of an organization overall layer, a partition population layer and a single cell layer, the overall layer provides the heterogeneity index and the overall staining pattern description, the population layer provides staining characteristics and interaction relations of each partition, the single cell layer provides a correction staining evaluation value and micro-environment influence information of each cell, and traditional single cell staining information and innovative intercellular interaction information are synthesized based on staining analysis results of cell interaction to form a tissue microenvironment analysis report.

As shown in fig. 2, after clustering the corrected staining evaluation values, the tissue section is divided into four cell population partitions of a high expression region, an expression region, a low expression region and a mixed region, wherein black square lines in the figure represent microenvironment heterogeneity indexes of each partition, gray triangle lines represent interaction intensities among the partitions, a biaxial display result shows that the high expression region and the mixed region have higher microenvironment heterogeneity indexes, namely 0.72 and 0.85 respectively, the heterogeneity index of the low expression region is at least 0.28, the interaction intensity among the partitions and the heterogeneity index show similar variation trend, wherein the interaction intensity of the mixed region reaches the highest value of 0.79, which indicates that complex inter-regulation relations exist among different types of cells in the region, and the analysis result effectively reveals a functional differentiation mode and an inter-cell dynamic interaction network inside an immunohistochemical staining tissue, so as to provide quantitative microenvironment feature description for staining analysis based on cell interaction.

The method for collecting and analyzing data of an immunohistochemical staining machine in the embodiment of the present application is described above, and the system for collecting and analyzing data of an immunohistochemical staining machine in the embodiment of the present application is described below, referring to fig. 3, an embodiment of the system for collecting and analyzing data of an immunohistochemical staining machine in the embodiment of the present application includes:

The data acquisition and analysis system of the middle-immune histochemical dyeing machine in the embodiment of the invention is described in detail from the perspective of a modularized functional entity in fig. 3, and the data acquisition and analysis equipment of the middle-immune histochemical dyeing machine in the embodiment of the invention is described in detail from the perspective of hardware processing.

Referring to fig. 4, in an embodiment of the present invention, an immunohistochemical staining machine data acquisition and analysis device is further provided, where the immunohistochemical staining machine data acquisition and analysis device may be a server, and an internal structure of the immunohistochemical staining machine data acquisition and analysis device may be as shown in fig. 4. The data acquisition and analysis equipment of the immunohistochemical dyeing machine comprises a processor, a memory, a display screen, an input device, a network interface and a database which are connected through a system bus. Wherein the computer is configured to provide computing and control capabilities. The memory of the data acquisition and analysis equipment of the immunohistochemical staining machine comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the data acquisition and analysis equipment of the immunohistochemical staining machine is used for storing the corresponding data in the embodiment. The network interface of the data acquisition and analysis equipment of the immunohistochemical staining machine is used for communicating with an external terminal through network connection. Which computer program, when being executed by a processor, carries out the above-mentioned method.

It will be appreciated by those skilled in the art that the structure shown in fig. 4 is merely a block diagram of a portion of the structure associated with the present invention and is not intended to limit the data acquisition and analysis apparatus of an immunohistochemical staining machine to which the present invention is applied.

The present invention also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and may also be a volatile computer readable storage medium, where instructions are stored in the computer readable storage medium, where the instructions, when executed on a computer, cause the computer to perform the steps of the immunohistochemical staining machine data collection analysis method.

It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, systems and units may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or partly in the form of a software product, or all or part of the technical solution, stored in a storage medium, comprising instructions for causing an immunohistochemical staining machine data acquisition analysis device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present invention. The storage medium includes a U disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, an optical disk, or other various media capable of storing program codes.

The foregoing embodiments are merely for illustrating the technical solution of the present invention, but not for limiting the same, and although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that modifications may be made to the technical solution described in the foregoing embodiments or equivalents may be substituted for parts of the technical features thereof, and that such modifications or substitutions do not depart from the spirit and scope of the technical solution of the embodiments of the present invention in essence.

Claims

1. An immunohistochemical staining machine data acquisition analysis method, which is characterized by comprising the following steps:

2. The method for collecting and analyzing data of an immunohistochemical staining machine according to claim 1, wherein the step of performing multi-cell collaborative staining data collection processing on the immunohistochemical section through neighborhood cell staining intensity correlation analysis to obtain an intercellular staining correlation data set comprises the steps of:

3. The method for collecting and analyzing data of an immunohistochemical staining machine according to claim 1, wherein the performing quantitative modeling processing on the signal transduction intensity between adjacent cells according to the spatial distribution pattern in the data set of the correlation of the staining between cells to obtain a signal transduction intensity matrix comprises:

4. The method for collecting and analyzing data of an immunohistochemical staining machine according to claim 1, wherein inputting the signal transmission intensity matrix into a cascade reaction prediction network performs cascade calculation processing on a mechanism of interaction of staining states among cells to obtain a micro-environment influence coefficient, and comprises the following steps:

5. The method for collecting and analyzing data of an immunohistochemical staining machine according to claim 4, wherein the step of performing a cascade reaction path tracking calculation process on the primary influence response value to obtain a multi-hop cell influence propagation sequence comprises:

6. The method for collecting and analyzing data of an immunohistochemical staining machine according to claim 1, wherein the performing a context correction process on the single cell staining result based on the correction effect of the micro-environmental influence coefficient on the staining state of the target cell to obtain a corrected staining evaluation value comprises:

7. The method according to claim 1, wherein the performing the microenvironment sensing analysis on the cell population staining pattern according to the corrected staining evaluation value to obtain a staining analysis result based on cell interaction comprises:

8. An immunohistochemical staining machine data acquisition and analysis system for implementing the method for the data acquisition and analysis of an immunohistochemical staining machine according to any one of claims 1 to 7, comprising:

9. An immunohistochemical staining machine data acquisition and analysis apparatus comprising a memory and a processor, the memory storing a computer program executable on the processor, the processor implementing the method of the immunohistochemical staining machine data acquisition and analysis of any one of claims 1 to 7 when executing the computer program.

10. A computer readable storage medium having stored thereon a computer program, characterized in that the computer program, when run by a processor, causes the processor to perform the immunohistochemical staining machine data acquisition analysis method of any of claims 1 to 7.