CN109580458B

CN109580458B - Flow-type cell intelligent immune typing method and device and electronic equipment

Info

Publication number: CN109580458B
Application number: CN201811459148.6A
Authority: CN
Inventors: 王志岗; 汝昆; 贺环宇
Original assignee: Suzhou Deep Analysis Intelligent Technology Co Ltd
Current assignee: Tianjin Shenxi Intelligent Technology Development Co.,Ltd.
Priority date: 2018-11-30
Filing date: 2018-11-30
Publication date: 2020-02-14
Anticipated expiration: 2038-11-30
Also published as: CN109580458A

Abstract

The invention provides a flow cell intelligent immunity typing method, a flow cell intelligent immunity typing device and electronic equipment, wherein position coordinates of each cell in a flow sample in a coordinate system with different cell surface antigen molecular weights as coordinate axes are obtained; dividing cells in the flow sample into a plurality of cell populations according to the location coordinates; identifying a cell type of each of the plurality of cell populations; judging whether the position coordinates of the cells in each cell population are within a preset range corresponding to the cell types of the cell population; and determining the cell group as an abnormal group when the judgment result shows that the position coordinates of the cells existing in the cell group are not in a preset range corresponding to the cell type of the cell group. The artificial intelligence is utilized to realize the grouping of the cells and the judgment of the abnormal groups by the neural network model, the labor intensity of professionals is greatly reduced, and the accuracy and the efficiency of flow cell immune typing are improved.

Description

Flow-type cell intelligent immune typing method and device and electronic equipment

Technical Field

The present invention relates to the field of flow cytometry, and in particular, to a flow cytometry intelligent immunophenotyping method and apparatus, an electronic device, a computer program product, and a computer-readable storage medium.

Background

Cell surface antigenic molecules that appear or disappear during the different stages of differentiation and cell activation of blood cells are collectively called cell differentiation population, and have different differentiation antigenic cluster expressions in erythroid cell lines, leukocyte lines, platelets, megakaryocytic cell lines and non-hematopoietic cells. When a hematologic neoplasm develops, the cells lose the serial specificity of normal cells and the regularity of the differentiation stages. Flow cytometry can be valuable for determining its source and differentiation stage, determining the presence of trace residual lesions, and inferring prognosis.

Most of the conventional flow-based immunotyping methods involve grouping cells quantitatively by a flow cytometer, judging whether each cell population is normal or abnormal from a two-dimensional viewpoint by a professional, and estimating the type, degree, and the like of a disease based on the characteristics of the abnormal cell population.

However, the prior art has the following disadvantages:

1. the subjectivity is high. The experience of each professional is different and the judgment basis is not completely the same, so that the judgment result is deviated, and even the judgment results given by the same person under different environments and states are not necessarily completely the same.

2. The labor intensity is high. The judgment personnel need a certain technical basis and working experience, and the actual huge workload and the relatively lacked professional personnel increase the labor intensity of the personnel.

3. The efficiency is low. Rely on relatively less professional to accomplish huge work load by manual work, its work efficiency is not high naturally.

Disclosure of Invention

In view of this, embodiments of the present invention provide a flow cytometry intelligent immunophenotyping method and apparatus, which replace the existing manual operation by an artificial intelligence manner, and solve the problems of the prior art, such as high subjectivity, high labor intensity, and low efficiency.

According to an aspect of the present invention, an embodiment of the present invention provides a flow cytometry intelligent immune typing method, including: acquiring the position coordinates of each cell in the flow sample in a coordinate system with coordinate axes of different antigen molecular weights on the cell surface; dividing cells in the flow sample into a plurality of cell populations according to the location coordinates; identifying a cell type of each of the plurality of cell populations; judging whether the position coordinates of the cells in each cell population are within a preset range corresponding to the cell types of the cell population; and determining the cell population as an abnormal population when the judgment result indicates that the position coordinates of the cells existing in the cell population are not within a preset range corresponding to the cell type of the cell population; wherein the determining whether the position coordinates of the cells in each of the cell populations are within a preset range corresponding to the cell type of the cell population is based on a first neural network model.

In an embodiment, the determining whether the position coordinates of the cells in each of the cell populations are within a predetermined range corresponding to the cell type of the cell population includes: inputting the position coordinates of the cells in the single cell population and the cell types of the single cell population into the first neural network model, and judging whether the position coordinates of the cells in the single cell population are not in a preset range corresponding to the cell types of the cell population through the first neural network model.

In one embodiment, the training process of the first neural network model includes: and taking the position coordinate of the single cell in the coordinate system and the cell type of the single cell as sample input, and taking the result of whether the position coordinate of the single cell is in a preset range corresponding to the cell type of the single cell as sample output to train the first neural network model.

In one embodiment, the obtaining the position coordinates of each cell in the flow sample in a coordinate system with the molecular weight of the different antigens on the cell surface on the coordinate axis comprises: selecting the molecular weight of a plurality of antigens on the cell surface as coordinate axes, and obtaining the position coordinates of each cell in a coordinate system formed by the coordinate axes.

In one embodiment, the acquiring the position coordinates of each cell in the coordinate system formed by the coordinate axes includes: and respectively selecting part of the coordinate axes to form a plurality of coordinate systems, and respectively obtaining the position coordinates of each cell in the coordinate systems.

In one embodiment, the dividing the cells in the flow sample into a plurality of cell populations according to the location coordinates comprises: inputting the position coordinates into a second neural network model, and dividing the cells in the flow sample into a plurality of cell groups through the second neural network model.

In an embodiment, the training process of the second neural network model includes: and training the second neural network model by taking the position coordinates of the single cell in the coordinate system as sample input and taking a cell group image corresponding to the single cell as sample output.

In an embodiment, the method further comprises: determining a degree of abnormality of the group of abnormalities.

In one embodiment, the determining the degree of abnormality of the abnormality group includes: and determining the abnormal degree of the abnormal group according to the difference between the position coordinate of the gravity center of the abnormal group in the coordinate system and the position coordinate of the normal cell corresponding to the cell type of the abnormal group in the coordinate system.

In one embodiment, the determining the degree of abnormality of the abnormality group includes: and inputting the position coordinates of the gravity center of the abnormal group in the coordinate system and the cell type of the abnormal group into a third neural network model, and determining the abnormal degree of the abnormal group through the third neural network model.

In one embodiment, the training process of the third neural network model includes: and training the third neural network model by taking the position coordinates of the single abnormal cell in the coordinate system and the cell type of the single abnormal cell as sample input and the abnormal degree of the single abnormal cell as sample output.

In an embodiment, the method further comprises: and compensating the position coordinates, and dividing the cells in the flow sample into a plurality of cell groups according to the compensated position coordinates.

In an embodiment, said compensating said position coordinates comprises: and multiplying the coordinate vector corresponding to the position coordinate by an inverse matrix of a compensation matrix to obtain the compensated position coordinate, wherein the compensation matrix is used for describing the influence degree of each color on each color channel in the dyeing process of the cell.

In an embodiment, before compensating the position coordinates, the method further includes: judging whether the compensation matrix is accurate or not; and if the judgment result is that the compensation matrix is not accurate, correcting the compensation matrix.

In an embodiment, the determining whether the compensation matrix is accurate includes: calculating a distance value between the position coordinates of the cell and a straight line composed of points in the coordinate system, wherein the coordinate components of the points are equal; calculating the number ratio of the cells with the distance value smaller than a preset distance threshold value in the flow sample; and if the number fraction is greater than a first preset fraction threshold, determining that the compensation matrix is inaccurate.

In one embodiment, said modifying said compensation matrix comprises: calculating a distance value between the position coordinates of the cell and a straight line composed of points in the coordinate system, wherein the coordinate components of the points are equal; calculating the number ratio of the cells of which the distance value is smaller than a preset distance threshold value; and adjusting element values in the compensation matrix to enable the number ratio after compensation to be smaller than a second preset ratio threshold value.

In one embodiment, the calculating a distance value between the position coordinates of the cell and a straight line composed of points in the coordinate system whose coordinate components are equal includes: and performing weighting processing on each coordinate component of the position coordinates of the cell, and calculating a distance value between the weighted position coordinates and a straight line formed by points with equal coordinate components in the coordinate system.

In one embodiment, the identifying the cell type of each of the plurality of cell populations comprises: the cell types of the plurality of cell groups are identified by scattered light signals generated by irradiating the plurality of cell groups with laser light.

In one embodiment, the scattered light signals include forward angle scattered light signals and side scattered light signals.

According to another aspect of the present invention, an embodiment of the present invention provides a flow cytometry intelligent immunophenotyping device, including: the coordinate acquisition module is configured to acquire the position coordinates of each cell in the flow sample in a coordinate system with the different molecular weights of antigens on the cell surface as coordinate axes; a clustering module configured to cluster cells in the flow sample into a plurality of cell clusters according to the location coordinates; a species identification module configured to identify a cell species of each of the plurality of cell groups; an abnormal group judgment module configured to judge whether the position coordinates of the cells in each of the cell groups are within a preset range corresponding to a cell type of the cell group; and an abnormal group determination module configured to determine that the cell group is an abnormal group when the determination result indicates that the position coordinates of the cells existing in the cell group are not within a preset range corresponding to the cell type of the cell group; wherein the determining whether the position coordinates of the cells in each of the cell populations are within a preset range corresponding to the cell type of the cell population is based on a first neural network model.

In one embodiment, the anomaly group determination module is configured to: inputting the position coordinates of the cells in the single cell population and the cell types of the single cell population into the first neural network model, and judging whether the position coordinates of the cells in the single cell population are not in a preset range corresponding to the cell types of the cell population through the first neural network model.

In one embodiment, the anomaly group determination module includes: a first training unit configured to train the first neural network model as a sample output, a result of whether the position coordinates of the single cell in the coordinate system and a cell type of the single cell are input as a sample, and the position coordinates of the single cell are within a preset range corresponding to the cell type of the single cell.

In one embodiment, the coordinate acquisition module is configured to: selecting the molecular weight of a plurality of antigens on the cell surface as coordinate axes, and obtaining the position coordinates of each cell in a coordinate system formed by the coordinate axes.

In one embodiment, the coordinate acquisition module is configured to: and respectively selecting part of the coordinate axes to form a plurality of coordinate systems, and respectively obtaining the position coordinates of each cell in the coordinate systems.

In one embodiment, the clustering module is configured to: inputting the position coordinates into a second neural network model, and dividing the cells in the flow sample into a plurality of cell groups through the second neural network model.

In one embodiment, the clustering module comprises: a second training unit configured to train the second neural network model with the position coordinates of the single cell in the coordinate system as a sample input and a cell population image corresponding to the single cell as a sample output.

In an embodiment, the apparatus further comprises: an abnormality degree determination module configured to determine an abnormality degree of the abnormality group.

In one embodiment, the abnormality degree determination module is configured to: and determining the abnormal degree of the abnormal group according to the difference between the position coordinate of the gravity center of the abnormal group in the coordinate system and the position coordinate of the normal cell corresponding to the cell type of the abnormal group in the coordinate system.

In one embodiment, the abnormality degree determination module is configured to: and inputting the position coordinates of the gravity center of the abnormal group in the coordinate system and the cell type of the abnormal group into a third neural network model, and determining the abnormal degree of the abnormal group through the third neural network model.

In one embodiment, the abnormality degree determination module includes: a third training unit configured to train the third neural network model with the position coordinates of the single abnormal cell in the coordinate system and the cell type of the single abnormal cell as sample inputs and the degree of abnormality of the single abnormal cell as sample outputs.

In one embodiment, the apparatus further comprises: a compensation module configured to compensate the position coordinates, and a grouping module to group the cells in the flow sample into a plurality of cell groups according to the compensated position coordinates.

In one embodiment, the compensation module is configured to: and multiplying the coordinate vector corresponding to the position coordinate by an inverse matrix of a compensation matrix to obtain the compensated position coordinate, wherein the compensation matrix is used for describing the influence degree of each color on each color channel in the dyeing process of the cell.

In one embodiment, the compensation module configuration comprises: an accuracy judging unit configured to judge whether the compensation matrix is accurate before compensating the position coordinates; and the correcting unit is configured to correct the compensation matrix when the judgment result is that the compensation matrix is inaccurate.

In an embodiment, the accuracy determination unit is configured to: calculating a distance value between the position coordinates of the cell and a straight line composed of points in the coordinate system, wherein the coordinate components of the points are equal; calculating the number ratio of the cells with the distance value smaller than a preset distance threshold value in the flow sample; and if the number fraction is greater than a first preset fraction threshold, determining that the compensation matrix is inaccurate.

In an embodiment, the correction unit is configured to: calculating the distance value between the position coordinates of the cells and a straight line which has the same included angle with all the coordinate axes in the coordinate system; calculating the number ratio of the cells of which the distance value is smaller than a preset distance threshold value; and adjusting element values in the compensation matrix to enable the number ratio after compensation to be smaller than a second preset ratio threshold value.

In an embodiment, the correction unit is configured to: and performing weighting processing on each coordinate component of the position coordinates of the cell, and calculating a distance value between the weighted position coordinates and a straight line formed by points with equal coordinate components in the coordinate system.

In one embodiment, the category identification module is configured to: the cell types of the plurality of cell groups are identified by scattered light signals generated by irradiating the plurality of cell groups with laser light.

According to another aspect of the present invention, an embodiment of the present invention provides an electronic device, including: a processor; a memory; and computer program instructions stored in the memory, which when executed by the processor, cause the processor to perform a method as claimed in any one of the above.

According to another aspect of the invention, an embodiment of the invention provides a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the method as defined in any one of the above.

According to another aspect of the present invention, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the method according to any one of the above.

According to the flow cell intelligent immunity typing method provided by the embodiment of the invention, the position coordinates of each cell in a flow sample in a coordinate system with coordinate axes of different antigen molecular weights on the cell surface are obtained; dividing cells in the flow sample into a plurality of cell populations according to the location coordinates; identifying a cell type of each of the plurality of cell populations; judging whether the position coordinates of the cells in each cell population are within a preset range corresponding to the cell types of the cell population; and determining the cell population as an abnormal population when the judgment result indicates that the position coordinates of the cells existing in the cell population are not within a preset range corresponding to the cell type of the cell population; wherein the determining whether the position coordinates of the cells in each of the cell populations are within a preset range corresponding to the cell type of the cell population is based on a first neural network model. The artificial intelligence is utilized to realize the grouping of the cells and the judgment of the abnormal groups by the neural network model, the labor intensity of professionals is greatly reduced, and the accuracy and the efficiency of flow cell immune typing are improved.

Drawings

Fig. 1 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to an embodiment of the present disclosure.

Fig. 2 is a flowchart of a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure.

Fig. 3 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure.

Fig. 4 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure.

Fig. 5 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure.

Fig. 6 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure.

Fig. 7 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure.

Fig. 8 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure.

Fig. 9 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure.

Fig. 10 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure.

Fig. 11 is a flowchart illustrating a method for determining accuracy of a compensation matrix according to an embodiment of the present disclosure.

Fig. 12 is a flowchart illustrating a compensation matrix modification method according to an embodiment of the present application.

Fig. 13 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to an embodiment of the present disclosure.

Fig. 14 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure.

Fig. 15 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure.

Fig. 16 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure.

Fig. 17 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure.

Fig. 18 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure.

Fig. 19 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure.

Fig. 20 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The flow cytometry adopts the working principle of carrying out multi-parameter and rapid quantitative analysis on single cells or other biological particles through monoclonal antibodies on the cellular molecule level, can analyze tens of thousands of cells at high speed, can simultaneously measure a plurality of parameters from one cell, has the advantages of rapid speed, high precision and good accuracy, and is one of the most advanced cell quantitative analysis techniques of the present generation. At present, the detection of peripheral blood leucocyte, bone marrow cell, tumor cell and the like by using a flow cytometer is an important component of clinical detection in clinic.

The flow cytometry is mainly realized on a flow cytometer, and the working principle of the flow cytometer is that cells or particles which are suspended in liquid and are dispersed and marked by fluorescence pass through a sample cell one by one, fluorescence signals are captured by a fluorescence detector and converted into electric pulse signals respectively representing scattering angles and different fluorescence intensities, and a corresponding point diagram, a histogram and a three-dimensional structural image are formed for analysis through computer processing.

The flow cytometer mainly comprises: a flow system comprising a flow cell and a flow driving system for effecting flow of the sample; the optical system comprises an excitation light source and a light beam collecting system and is used for realizing the fluorescent labeling and collection of cells or particles; and the electronic system comprises a photoelectric converter and a data processing system and is used for converting the fluorescence information of the cells into electric signals and obtaining the related data information of the cells. Wherein, the sample used for the flow cytometer is single cell suspension, which can be blood, suspension cell culture solution, various body fluids, single cell suspension of fresh solid tumor, single cell suspension of paraffin-embedded tissue, and the like.

In the current flow cytometry, after a dot diagram of a single cell suspension is obtained by a flow cytometer, a professional performs cell grouping and abnormal judgment of a cell group on the dot diagram, and because the manual operation can only realize the judgment of two-dimensional coordinates, the position characteristics of the cell group need to be judged in a plurality of two-dimensional coordinate systems during the manual operation, and the final diagnosis result is obtained through comprehensive analysis. However, different people can generate diagnosis differences during operation, and the number of professionals is limited, the workload is huge, the manual workload is huge, and meanwhile, the efficiency is low.

In order to solve the problems, the application provides a flow cell intelligent immunity typing method, a flow cell intelligent immunity typing device and electronic equipment, the existing manual operation is replaced by artificial intelligence, the cell grouping and the abnormal judgment of the cell group are realized, so that a diagnosis result is obtained, the diagnosis difference caused by different experience and judgment of different personnel according to different conditions is avoided, meanwhile, the manual workload is greatly reduced, and the efficiency is improved.

The following describes implementation manners of the flow cytometry intelligent immunophenotyping method, apparatus, and electronic device according to the present application, with reference to the accompanying drawings and specific embodiments:

fig. 1 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to an embodiment of the present disclosure. As shown in fig. 1, the flow cytometry intelligent immunophenotyping method comprises the following steps:

step 110: and acquiring the position coordinates of each cell in the flow sample in a coordinate system with coordinate axes of different cell surface antigen molecular weights.

As described above, blood cells express different antigen clusters (i.e., antigen molecular weights) at different stages of differentiation and during cell activation, and position coordinates of each cell in a flow sample labeled with fluorescence in a coordinate system formed by the coordinate axes are obtained using the different antigen molecular weights as coordinate axes. The coordinate value of the cell on the coordinate axis represents the carrying number of the antigen molecules represented by the cell on the coordinate axis, namely, the position coordinate of the cell is determined by the number of various antigen molecules carried by the cell.

Step 120: and dividing the cells in the flow sample into a plurality of cell groups according to the position coordinates.

All cells in the flow sample are divided into a plurality of cell groups according to the position coordinates of each cell, wherein each cell group is the same type of cell. Since the antigen molecules carried by the same kind of cells are generally the same, and even if the cells are abnormal (not normally present alone), the antigen molecules carried by a group of abnormal cells are also the same, all the cells in the sample can be divided into a plurality of cell groups.

Step 130: the cell types of each of the plurality of cell populations are identified.

Since the antigen molecules carried by the cells in the same cell population are the same as described above, the cell type of each cell population can be automatically identified based on the position coordinates of the cells in each cell population.

Step 140: and judging whether the position coordinates of the cells in each cell group are within a preset range corresponding to the cell types of the cell group.

Although the antigen molecules carried by normal cells will vary from human to human or during different growth processes, the number of antigen molecules carried by each cell type will have a normal range. Therefore, the normal range can be used as a basis for determining whether the cells are abnormal, that is, whether the position coordinates of the cells in each cell group are within the normal range corresponding to the cell type of the cell group, thereby determining whether the cell group is abnormal. For example, if the normal range of the number of antigen molecules carried by a certain kind of cells is 200-500, it can be determined whether the cell population is abnormal by determining whether the component of the coordinate position of the cells in the cell population on the coordinate axis corresponding to the antigen molecules is within 200-500.

Step 150: and determining the cell group as an abnormal group when the judgment result indicates that the position coordinates of the cells in the cell group are not in the preset range corresponding to the cell type of the cell group. And judging whether the position coordinates of the cells in each cell group are within a preset range corresponding to the cell types of the cell group or not is realized based on the first neural network model.

Since the cells of each cell population are of the same type, and the normal and abnormal cells of the same type are of different types, when one cell in the cell population is an abnormal cell, the cell population can be determined to be an abnormal cell population. That is, when the position coordinates of the cells in the cell population are not within the normal range as a result of the above determination, the cell population is determined to be an abnormal population. The first neural network model can judge whether the position coordinates of the cells in each cell group are in a normal range corresponding to the cell types of the cell group, and therefore whether the cell group is an abnormal group can be determined.

According to the embodiment of the application, the artificial intelligence method and the neural network model are used for automatically realizing the grouping of the cells and judging the abnormal groups of the neural network model, the existing manual operation is replaced, the labor intensity of professionals is greatly reduced, and meanwhile the accuracy and the efficiency of flow cell immune typing are improved.

Fig. 2 is a flowchart of a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure. As shown in fig. 2, step 140 may include the sub-steps of:

step 145: inputting the position coordinates of the cells in the single cell group and the cell types of the single cell group into a first neural network model, and judging whether the position coordinates of the cells in the single cell group are not in a preset range corresponding to the cell types of the cell group through the first neural network model.

And taking the position coordinates of the cells in the single cell group and the cell types of the single cell group as the input of the first neural network model, and outputting through the first neural network model to obtain whether the position coordinates of the cells in the single cell group are not in the preset range corresponding to the cell types of the cell group. Whether each cell group is abnormal or not is automatically judged by utilizing the neural network model, the labor intensity of professionals is greatly reduced, and meanwhile, the accuracy and the efficiency of flow cytometry immune typing are improved.

In an embodiment, the training process of the first neural network model may include: and taking the position coordinates of the single cell in the coordinate system and the cell type of the single cell as sample input, and taking the result of whether the position coordinates of the single cell are in a preset range corresponding to the cell type of the single cell as sample output to train the first neural network model.

When the first neural network model is trained, the position coordinates of the single cell in the coordinate system and the cell type of the single cell are used as input, and the result of whether the position coordinates of the single cell are in the preset range corresponding to the cell type of the single cell is used as output, so that the input-output sample is used for training the first neural network model. Through the training of the big data sample, the error of the judgment result caused by subjective factors in manual operation can be avoided, and therefore the judgment accuracy is improved.

Fig. 3 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure. As shown in fig. 3, step 110 may include the sub-steps of:

step 115: selecting the molecular weights of multiple antigens on the cell surface as coordinate axes, and obtaining the position coordinates of each cell in a coordinate system formed by the coordinate axes.

Cells in a flow sample can be distinguished from various antigen molecules only after being dyed by a dye, the types of the antigen molecules carried on the surfaces of the cells are more, and the dye is limited in color types and expensive. Therefore, in consideration of cost saving and analysis necessity, a plurality of antigen molecular weights which have an influence or a large influence on an analysis result are selected as coordinate axes according to analysis requirements, position coordinates of each cell in a coordinate system formed by the coordinate axes are obtained, and grouping and abnormal group judgment are performed on the cells in the flow type sample according to the position coordinates.

Fig. 4 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure. As shown in fig. 4, step 115 may include the sub-steps of:

step 1151: and respectively selecting part of the coordinate axes to form a plurality of coordinate systems, and respectively obtaining the position coordinates of each cell in the plurality of coordinate systems.

As mentioned above, the dye has a limited variety of colors, and the variety of colors that may appear during the course of the assay is insufficient to support the variety of antigenic molecules required for the assay. In order to solve the problem, the embodiment of the present application proposes that different antigen molecular weights required for analysis are used as coordinate axes of a plurality of coordinate systems, a final judgment result is obtained by combining judgment results of the plurality of coordinate systems, and the analysis work is completed under the condition that the color types of dyes are limited.

It should be understood that, in the embodiment of the present application, coordinate axes in each coordinate system are correlated, and a part or all of the determination results may be obtained through combination of the coordinate axes, that is, a certain specific determination result or a certain basis for a final determination result may be achieved through a combination manner of the coordinate axes.

It should be understood that, in the embodiment of the present application, different numbers of coordinate systems and combinations of different coordinate axes in the coordinate systems may be selected according to different requirements of analysis, and the same antigen molecular weight may be used as the coordinate axes in the different coordinate systems, as long as the number of the selected coordinate systems and the combination manner of the coordinate axes in the coordinate systems can implement analysis, and the number of the coordinate systems and the combination manner of the coordinate axes in the coordinate systems are not limited in the embodiment of the present application.

Fig. 5 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure. As shown in fig. 5, step 120 may include the sub-steps of:

step 125: inputting the position coordinates into a second neural network model, and dividing the cells in the flow sample into a plurality of cell groups through the second neural network model.

By establishing the second neural network model, the cells in the flow sample are divided into a plurality of cell groups by using the second neural network model, so that automatic grouping of the cells is realized, the workload of manual operation is reduced, the cell grouping efficiency is improved, and a data basis is provided for subsequent abnormal judgment of the cell groups.

In an embodiment, the training process of the second neural network model may include: and (3) taking the position coordinates of the single cell in the coordinate system as sample input and taking the cell group image corresponding to the single cell as sample output to train a second neural network model.

When the second neural network model is trained, the position coordinates of the single cell in the coordinate system are used as input, the cell group image corresponding to the single cell is used as output, and the input-output sample is used for training the second neural network model. Through training of the big data samples, errors of grouping results caused by subjective factors in manual operation can be avoided, and therefore accuracy of grouping is improved.

Fig. 6 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure. As shown in fig. 6, the method may further include:

step 160: the degree of abnormality of the abnormality group is determined.

After a certain cell population is determined to be an abnormal population, it is necessary to further determine the degree of abnormality of the abnormal population. The type of disease is obtained from the abnormal cell population, but the severity of the disease is not obtained, and therefore, it is necessary to further determine the abnormality degree of the abnormal cell population to obtain the severity of the disease and obtain the final analysis result.

In one embodiment, determining the degree of abnormality of the abnormality group includes: and determining the abnormal degree of the abnormal group according to the difference between the position coordinate of the gravity center of the abnormal group in the coordinate system and the position coordinate of the normal cell corresponding to the cell type of the abnormal group in the coordinate system.

In the embodiment of the application, because each person has the difference, the accurate abnormal degree of the abnormal group is difficult to determine simply through numerical values, and the abnormal degree of the abnormal group can be judged by comparing the relative position relationship between the abnormal group and the normal cell group, so that errors caused by the differences of different human bodies can be avoided, and the analysis precision is improved.

Fig. 7 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure. As shown in fig. 7, step 160 may include the sub-steps of:

step 165: and inputting the position coordinates of the gravity center of the abnormal group in the coordinate system and the cell type of the abnormal group into a third neural network model, and determining the abnormal degree of the abnormal group through the third neural network model.

By establishing the third neural network model and determining the abnormal degree of the abnormal group by using the third neural network model, the severity of the disease is obtained, the automatic analysis of the disease is realized, the workload of manual operation is reduced, and the disease analysis efficiency is improved.

In one embodiment, the training process of the third neural network model includes: and (3) training a third neural network model by taking the position coordinates of the single abnormal cell in the coordinate system and the cell type of the single abnormal cell as sample input and the abnormal degree of the single abnormal cell as sample output.

When the third neural network model is trained, the position coordinates of the single abnormal cell in the coordinate system and the cell type of the single abnormal cell are used as input, the abnormal degree of the single abnormal cell is used as output, and the input-output sample is used for training the third neural network model. Through the training of the big data sample, errors of an analysis result caused by subjective factors in manual operation can be avoided, and therefore the accuracy of disease analysis is improved.

Fig. 8 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure. As shown in fig. 8, the method of the embodiment of the present application may further include:

step 170: the position coordinates are compensated.

Because the flow cytometer may generate certain interference under different use environments or during long-term use, where a large interference is an influence of different colors of the cell on each color channel during the staining process, for example, blue should have no influence on a red color channel, but in practical application, blue may generate a small interference on the red color channel, thereby generating interference on the position coordinates of the cell, resulting in that the position coordinates of the cell are not accurate. Therefore, the position coordinates may also be compensated before step 120 to restore the exact position coordinates of each cell, and then the cells in the flow sample are divided into a plurality of cell groups according to the compensated position coordinates.

Fig. 9 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure. As shown in fig. 9, step 170 may include the sub-steps of:

step 175: and multiplying the coordinate vector corresponding to the position coordinate by an inverse matrix of the compensation matrix to obtain the compensated position coordinate, wherein the compensation matrix is used for describing the influence degree of each color on each color channel in the dyeing process of the cells.

The accurate position coordinates of the cells can be simply obtained by obtaining the accurate compensation matrix and calculating by using the initial position coordinates and the compensation matrix to obtain the compensated position coordinates, so that accurate data support is provided for ensuring the subsequent accurate cell grouping and abnormal group judgment.

Fig. 10 is a flowchart illustrating a flow cytometry intelligent immunophenotyping method according to another embodiment of the present disclosure. As shown in fig. 10, before step 170, the method may further include:

step 180: and judging whether the compensation matrix is accurate or not.

Step 190: and when the judgment result is that the compensation matrix is not accurate, correcting the compensation matrix.

Before the position coordinates are compensated, the accuracy of the compensation matrix needs to be judged, and only the accurate compensation matrix can obtain the accurate position coordinates, so that accurate data support is provided for ensuring the subsequent accurate cell grouping and abnormal group judgment.

Fig. 11 is a flowchart illustrating a method for determining accuracy of a compensation matrix according to an embodiment of the present disclosure. As shown in fig. 11, step 180 may include the sub-steps of:

step 181: and calculating the distance value between the position coordinates of the cell and a straight line consisting of points with equal coordinate components in the coordinate system.

Step 182: and calculating the number ratio of the cells with the distance value smaller than the preset distance threshold value in the flow sample.

Step 183: judging whether the number ratio is greater than a first preset ratio threshold value, and turning to step 184 when the number ratio is greater than the first preset ratio threshold value; otherwise, go to step 185.

Step 184: the determination of the compensation matrix is inaccurate.

Step 185: and the compensation matrix is accurately determined.

The accuracy of the compensation matrix is judged by calculating the cell aggregation degree around a straight line formed by points with equal coordinate components in a coordinate system, the judgment of the cell aggregation degree in multiple dimensions is realized by using artificial intelligence, the artificial comprehensive judgment of multiple two dimensions is replaced, and the efficiency of the whole analysis process is further improved.

Fig. 12 is a flowchart illustrating a compensation matrix modification method according to an embodiment of the present application. As shown in fig. 12, step 190 may include the sub-steps of:

step 191: and calculating the distance value between the position coordinates of the cell and a straight line consisting of points with equal coordinate components in the coordinate system.

Step 192: and calculating the number ratio of the cells with the distance value smaller than the preset distance threshold value.

Step 193: and adjusting element values in the compensation matrix to enable the compensated quantity ratio to be smaller than a second preset ratio threshold value.

By adjusting the element values in the compensation matrix, the cell aggregation degree around the straight line composed of the points with the same coordinate components in the coordinate system is minimized, that is, the cell aggregation ratio around the straight line composed of the points with the same coordinate components is minimized. The adjustment of the multi-dimensional compensation matrix is realized by using artificial intelligence, multiple manual two-dimensional adjustments are replaced, and the efficiency and the accuracy of the whole analysis process are further improved.

In an embodiment, the step 181 and the step 191 may include: weighting each coordinate component of the position coordinates of the cell, and calculating a distance value between the weighted position coordinates and a straight line composed of points in the coordinate system, each coordinate component of which is equal to the weighted position coordinates.

In one embodiment, step 130 may include: the scattered light signals generated by irradiating the plurality of cell groups with laser light identify the cell types of the plurality of cell groups, respectively. Preferably, the scattered light signals include forward angle scattered light signals and side scattered light signals.

The laser irradiation of the cell can generate a plurality of scattered light signals, the forward angle scattered light signal can reflect the size of the cell, the side scattered light signal can reflect the internal complex structure of the cell, and the category of the cell can be identified through the forward angle scattered light signal and the side scattered light signal. It should be understood that, in the embodiment of the present application, the position coordinates of the cells in the cell group and the scattered light signals may be combined to identify the type of the cell group, so as to improve the accuracy of the identification.

Fig. 13 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to an embodiment of the present disclosure. As shown in fig. 13, the apparatus includes: the coordinate acquisition module 1 is used for acquiring the position coordinates of each cell in the flow sample in a coordinate system with different cell surface antigen molecular weights as coordinate axes; the grouping module 2 is used for grouping the cells in the flow sample into a plurality of cell groups according to the position coordinates; a species identification module 3 for identifying a cell species of each of the plurality of cell groups; the abnormal group judgment module 4 is used for judging whether the position coordinates of the cells in each cell group are within a preset range corresponding to the cell types of the cell group; the abnormal group determining module 5 is used for determining the cell group as an abnormal group when the judgment result shows that the position coordinates of the cells in the cell group are not in the preset range corresponding to the cell type of the cell group; and judging whether the position coordinates of the cells in each cell group are within a preset range corresponding to the cell types of the cell group or not is realized based on the first neural network model.

According to the embodiment of the application, the grouping of the cells and the judgment of the abnormal groups by the neural network model are automatically realized by utilizing the intelligent immune typing state of the flow cells, the existing manual operation is replaced, the labor intensity of professionals is greatly reduced, and the accuracy and the efficiency of the flow cell immune typing are improved.

In one embodiment, the anomaly group determination module 4 may be configured to: inputting the position coordinates of the cells in the single cell group and the cell types of the single cell group into a first neural network model, and judging whether the position coordinates of the cells in the single cell group are not in a preset range corresponding to the cell types of the cell group through the first neural network model.

Fig. 14 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure. As shown in fig. 14, the abnormality group determination module 4 may include: the first training unit 41 is configured to train the first neural network model by using a result that the position coordinates of the single cell in the coordinate system and the cell type of the single cell are input as a sample, and whether the position coordinates of the single cell are within a preset range corresponding to the cell type of the single cell is output as a sample.

In an embodiment, the coordinate acquisition module 1 may be configured to: selecting the molecular weights of multiple antigens on the cell surface as coordinate axes, and obtaining the position coordinates of each cell in a coordinate system formed by the coordinate axes.

Selecting multiple antigen molecular weights which have influence or have great influence on an analysis result as coordinate axes according to analysis requirements, acquiring the position coordinates of each cell in a coordinate system formed by the coordinate axes, and performing grouping and abnormal group judgment on the cells in the flow sample according to the position coordinates.

In an embodiment, the coordinate acquisition module 1 may be configured to: and respectively selecting part of the coordinate axes to form a plurality of coordinate systems, and respectively obtaining the position coordinates of each cell in the plurality of coordinate systems.

And obtaining a final judgment result through the combination of the judgment results of a plurality of coordinate systems, and completing analysis work under the condition that the color types of the dyes are limited.

In an embodiment, the clustering module 2 may be configured to: inputting the position coordinates into a second neural network model, and dividing the cells in the flow sample into a plurality of cell groups through the second neural network model.

Fig. 15 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure. As shown in fig. 15, the grouping module 2 may include: and a second training unit 21, configured to train a second neural network model by using the position coordinates of the single cell in the coordinate system as a sample input and using the cell group image corresponding to the single cell as a sample output.

Fig. 16 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure. As shown in fig. 16, the apparatus may further include: and an abnormality degree determination module 6 for determining the abnormality degree of the abnormality group. After determining that a certain cell population is an abnormal population, the abnormal degree of the abnormal population needs to be further determined to obtain the severity of the disease and obtain the final analysis result.

In one embodiment, the degree of abnormality determination module 6 is configured to: and determining the abnormal degree of the abnormal group according to the difference between the position coordinate of the gravity center of the abnormal group in the coordinate system and the position coordinate of the normal cell corresponding to the cell type of the abnormal group in the coordinate system.

In one embodiment, the degree of abnormality determination module 6 is configured to: and inputting the position coordinates of the gravity center of the abnormal group in the coordinate system and the cell type of the abnormal group into a third neural network model, and determining the abnormal degree of the abnormal group through the third neural network model.

Fig. 17 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure. As shown in fig. 17, the abnormality degree determination module 6 may include: and a third training unit 61, configured to train a third neural network model by using the position coordinates of the single abnormal cell in the coordinate system and the cell type of the single abnormal cell as sample inputs and using the abnormal degree of the single abnormal cell as sample outputs.

Fig. 18 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure. As shown in fig. 18, the apparatus may further include: and the compensation module 7 is used for compensating the position coordinates. The position coordinates are compensated to recover accurate position coordinates for each cell, and then the cells in the flow sample are divided into a plurality of cell populations according to the compensated position coordinates.

In an embodiment, the compensation module 7 may be configured to: and multiplying the coordinate vector corresponding to the position coordinate by an inverse matrix of the compensation matrix to obtain the compensated position coordinate, wherein the compensation matrix is used for describing the influence degree of each color on each color channel in the dyeing process of the cells.

Fig. 19 is a schematic structural diagram of a flow cytometry intelligent immunophenotyping device according to another embodiment of the present disclosure. As shown in fig. 19, the compensation module 7 may include: an accuracy judgment unit 71, configured to judge whether the compensation matrix is accurate before compensating the position coordinate; and a correcting unit 72, configured to correct the compensation matrix if the determination result is that the compensation matrix is not accurate.

In one embodiment, the accuracy determining unit 71 is configured to: calculating the distance value between the position coordinates of the cells and a straight line formed by points with equal coordinate components in a coordinate system; calculating the number ratio of cells with the distance value smaller than a preset distance threshold value in the flow sample; and determining that the compensation matrix is inaccurate if the number fraction is greater than a first preset fraction threshold.

In one embodiment, the modification unit 72 is configured to: calculating the distance value between the position coordinate of the cell and a straight line which has the same included angle with all coordinate axes in the coordinate system; calculating the number ratio of the cells with the distance value smaller than a preset distance threshold value; and adjusting element values in the compensation matrix to enable the compensated quantity ratio to be smaller than a second preset ratio threshold value.

In one embodiment, the modification unit 72 is configured to: weighting each coordinate component of the position coordinates of the cell, and calculating a distance value between the weighted position coordinates and a straight line composed of points in the coordinate system, each coordinate component of which is equal to the weighted position coordinates.

In an embodiment, the category identification module 3 is configured to: the scattered light signals generated by irradiating the plurality of cell groups with laser light identify the cell types of the plurality of cell groups, respectively. Preferably, the scattered light signals include forward angle scattered light signals and side scattered light signals. The laser irradiation of the cell can generate a plurality of scattered light signals, the forward angle scattered light signal can reflect the size of the cell, the side scattered light signal can reflect the internal complex structure of the cell, and the category of the cell can be identified through the forward angle scattered light signal and the side scattered light signal.

Fig. 20 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. As shown in fig. 20, the electronic device may be an online electronic device such as a medical detection instrument equipped with a cell type analysis apparatus thereon, or an offline electronic device capable of communicating with the online electronic device to transmit the trained machine learning model thereto.

FIG. 20 illustrates a block diagram of an electronic device in accordance with an embodiment of the present application.

As shown in fig. 20, the electronic device 9 includes one or more processors 91 and a memory 92.

The processor 20 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 9 to perform desired functions.

Memory 92 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, Random Access Memory (RAM), cache memory (cache), and/or the like. The non-volatile memory may include, for example, Read Only Memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium and executed by the processor 91 to implement the region labeling methods of the various embodiments of the present application described above and/or other desired functions. Various contents such as position coordinates of cells in the streaming sample, a coordinate system, a training sample, etc. may also be stored in the computer-readable storage medium.

In one example, the electronic device 9 may further include: an input device 93 and an output device 94, which are interconnected by a bus system and/or other form of connection mechanism (not shown).

For example, the input device 93 may be a data acquisition device for acquiring positional coordinates of cells in a streaming sample, and the acquired positional coordinate information may be stored in the memory 92 for use by other components. Of course, other integrated or discrete data acquisition devices may be utilized to acquire the positional coordinate information of the cells in the streaming sample and transmit it to the electronic device 9. The input device 93 may also include, for example, a keyboard, a mouse, and a communication network and a remote input device connected thereto.

The output device 94 may output various information to the outside (e.g., a user or a machine learning model), including the determined cell type, cell population information, abnormal population information, training samples, etc. The output devices 94 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, among others.

Of course, for the sake of simplicity, only some of the components related to the present application in the electronic device 9 are shown in fig. 20, and components such as a bus, an input/output interface, and the like are omitted. In addition, the electronic device 9 may comprise any other suitable components, depending on the specific application.

In addition to the above-described methods and apparatus, embodiments of the present application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps in the methods according to the various embodiments of the present application described in the present specification.

The computer program product may be written with program code for performing the operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.

Furthermore, embodiments of the present application may also be a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform the steps in the methods according to the various embodiments of the present application described in the present specification.

The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The foregoing describes the general principles of the present application in conjunction with specific embodiments, however, it is noted that the advantages, effects, etc. mentioned in the present application are merely examples and are not limiting, and they should not be considered essential to the various embodiments of the present application. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the foregoing disclosure is not intended to be exhaustive or to limit the disclosure to the precise details disclosed.

The block diagrams of devices, apparatuses, and apparatuses referred to in this application are only given as illustrative examples and are not intended to require or imply that the connections, arrangements, and configurations must be made in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. The words "or" and "as used herein mean, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".

It should also be noted that in the devices, apparatuses, and methods of the present application, the components or steps may be decomposed and/or recombined. These decompositions and/or recombinations are to be considered as equivalents of the present application.

The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The foregoing description has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit embodiments of the application to the form disclosed herein. While various embodiment aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A flow cytometric intelligent immunophenotyping method, comprising:

obtaining the position coordinates of each cell in the flow sample in a coordinate system taking the number of different antigen molecules on the cell surface as coordinate axes;

judging whether the compensation matrix is accurate or not;

if the judgment result is that the compensation matrix is not accurate, correcting the compensation matrix;

if the judgment result is that the compensation matrix is accurate, compensating the position coordinate;

dividing cells in the flow sample into a plurality of cell groups according to the compensated position coordinates;

identifying a cell type of each of the plurality of cell populations;

judging whether the position coordinates of the cells in each cell population are within a preset range corresponding to the cell types of the cell population; and

determining the cell population as an abnormal population when the judgment result indicates that the position coordinates of the cells existing in the cell population are not in a preset range corresponding to the cell species of the cell population;

the determination of whether the position coordinates of the cells in each cell population are within a preset range corresponding to the cell type of the cell population is realized based on a first neural network model, and the specific method is as follows: inputting the position coordinates of the cells in the single cell population and the cell types of the single cell population into the first neural network model, and judging whether the position coordinates of the cells in the single cell population are not in a preset range corresponding to the cell types of the cell population through the first neural network model;

wherein the training process of the first neural network model comprises:

taking the position coordinates of the single cell in the coordinate system and the cell type of the single cell as sample input, and taking the result of whether the position coordinates of the single cell are in a preset range corresponding to the cell type of the single cell as sample output to train the first neural network model;

the specific method for judging whether the compensation matrix is accurate comprises the following steps:

calculating a distance value between the position coordinates of the cell and a straight line composed of points in the coordinate system, wherein the coordinate components of the points are equal;

calculating the number ratio of the cells with the distance value smaller than a preset distance threshold value in the flow sample; and

and if the number occupation ratio is larger than a first preset occupation ratio threshold value, determining that the compensation matrix is inaccurate.

2. The method of claim 1, wherein obtaining positional coordinates of each cell in the flow sample in a coordinate system having a different number of antigen molecules on the cell surface as coordinate axes comprises:

selecting the number of the multiple antigen molecules on the cell surface as a coordinate axis, and acquiring the position coordinate of each cell in a coordinate system formed by the coordinate axes.

3. The method of claim 2, wherein said obtaining the position coordinates of each cell in the coordinate system formed by the coordinate axes comprises:

and respectively selecting part of the coordinate axes to form a plurality of coordinate systems, and respectively obtaining the position coordinates of each cell in the coordinate systems.

4. The method of claim 1, wherein the dividing the cells in the flow sample into a plurality of cell populations according to the compensated position coordinates comprises:

inputting the compensated position coordinates into a second neural network model, and dividing the cells in the flow sample into a plurality of cell groups through the second neural network model.

5. The method of claim 4, wherein the training process of the second neural network model comprises:

and training the second neural network model by taking the compensated position coordinates of the single cell in the coordinate system as sample input and taking the cell group image corresponding to the single cell as sample output.

6. The method of claim 1, further comprising:

determining a degree of abnormality of the group of abnormalities.

7. The method of claim 6, wherein said determining a degree of abnormality of said group of abnormalities comprises:

and determining the abnormal degree of the abnormal group according to the difference between the position coordinate of the gravity center of the abnormal group in the coordinate system and the position coordinate of the normal cell corresponding to the cell type of the abnormal group in the coordinate system.

8. The method of claim 6, wherein said determining a degree of abnormality of said group of abnormalities comprises:

and inputting the position coordinates of the gravity center of the abnormal group in the coordinate system and the cell type of the abnormal group into a third neural network model, and determining the abnormal degree of the abnormal group through the third neural network model.

9. The method of claim 8, wherein the training process of the third neural network model comprises:

and training the third neural network model by taking the position coordinates of the single abnormal cell in the coordinate system and the cell type of the single abnormal cell as sample input and the abnormal degree of the single abnormal cell as sample output.

10. The method of claim 1, wherein the compensating the position coordinates comprises:

and multiplying the coordinate vector corresponding to the position coordinate by an inverse matrix of a compensation matrix to obtain the compensated position coordinate, wherein the compensation matrix is used for describing the influence degree of each color on each color channel in the dyeing process of the cell.

11. The method of claim 1, wherein the modifying the compensation matrix comprises:

calculating the number ratio of the cells of which the distance value is smaller than a preset distance threshold value; and

adjusting element values in the compensation matrix so that the number ratio after compensation is smaller than a second preset ratio threshold.

12. The method of claim 11, wherein the calculating a distance value between the position coordinates of the cell and a straight line composed of points in the coordinate system having equal coordinate components comprises:

and performing weighting processing on each coordinate component of the position coordinates of the cell, and calculating a distance value between the weighted position coordinates and a straight line formed by points with equal coordinate components in the coordinate system.

13. The method of claim 1, wherein the identifying the cell type of each of the plurality of cell populations comprises:

the cell types of the plurality of cell groups are identified by scattered light signals generated by irradiating the plurality of cell groups with laser light.

14. The method of claim 13, wherein the scattered light signals comprise forward angle scattered light signals and side scattered light signals.

15. A flow cytometric intelligent immunophenotyping device, comprising:

the coordinate acquisition module is used for acquiring the position coordinates of each cell in the flow sample in a coordinate system taking the number of different antigen molecules on the cell surface as coordinate axes;

the compensation module is used for judging whether the compensation matrix is accurate or not; if the judgment result is that the compensation matrix is not accurate, correcting the compensation matrix; if the judgment result is that the compensation matrix is accurate, compensating the position coordinate;

a clustering module configured to cluster the cells in the flow sample into a plurality of cell clusters according to the compensated position coordinates;

a species identification module configured to identify a cell species of each of the plurality of cell groups;

an abnormal group judgment module configured to judge whether the position coordinates of the cells in each of the cell groups are within a preset range corresponding to the cell type of the cell group; and

an abnormal group determination module configured to determine that the cell group is an abnormal group when the determination result indicates that the position coordinates of the cells existing in the cell group are not within a preset range corresponding to the cell type of the cell group;

wherein the determining whether the position coordinates of the cells in each of the cell populations are within a preset range corresponding to the cell type of the cell population is performed based on a first neural network model, and the abnormal population determining module is further configured to: inputting the position coordinates of the cells in the single cell population and the cell types of the single cell population into the first neural network model, and judging whether the position coordinates of the cells in the single cell population are not in a preset range corresponding to the cell types of the cell population through the first neural network model;

the abnormal group judgment module comprises a first training unit and a second training unit, wherein the first training unit is used for inputting the position coordinates of the single cell in the coordinate system and the cell type of the single cell as samples and outputting a result of whether the position coordinates of the single cell are in a preset range corresponding to the cell type of the single cell as samples to train the first neural network model;

wherein the compensation module is further configured to:

16. An electronic device, comprising:

a processor;

a memory; and

computer program instructions stored in the memory, which, when executed by the processor, cause the processor to perform the method of any of claims 1-14.

17. A computer product comprising computer program instructions stored on a memory, which, when executed by a processor, cause the processor to perform the method of any of claims 1-14.

18. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 14.