CN109580458A - Fluidic cell intelligent immunity classifying method, device and electronic equipment - Google Patents
Fluidic cell intelligent immunity classifying method, device and electronic equipment Download PDFInfo
- Publication number
- CN109580458A CN109580458A CN201811459148.6A CN201811459148A CN109580458A CN 109580458 A CN109580458 A CN 109580458A CN 201811459148 A CN201811459148 A CN 201811459148A CN 109580458 A CN109580458 A CN 109580458A
- Authority
- CN
- China
- Prior art keywords
- cell
- position coordinates
- category
- mass
- cell mass
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 230000036039 immunity Effects 0.000 title claims abstract description 47
- 210000004027 cell Anatomy 0.000 claims abstract description 550
- 230000002159 abnormal effect Effects 0.000 claims abstract description 99
- 239000000427 antigen Substances 0.000 claims abstract description 31
- 102000036639 antigens Human genes 0.000 claims abstract description 31
- 108091007433 antigens Proteins 0.000 claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims description 62
- 210000005036 nerve Anatomy 0.000 claims description 47
- 239000000203 mixture Substances 0.000 claims description 15
- 230000005484 gravity Effects 0.000 claims description 9
- 239000003086 colorant Substances 0.000 claims description 6
- 238000004043 dyeing Methods 0.000 claims description 6
- 238000013473 artificial intelligence Methods 0.000 abstract description 9
- 238000013394 immunophenotyping Methods 0.000 abstract description 7
- 238000003062 neural network model Methods 0.000 abstract description 6
- 239000000523 sample Substances 0.000 description 73
- 238000012549 training Methods 0.000 description 32
- 238000010586 diagram Methods 0.000 description 21
- 238000004458 analytical method Methods 0.000 description 20
- 238000004590 computer program Methods 0.000 description 15
- 230000003287 optical effect Effects 0.000 description 15
- 201000010099 disease Diseases 0.000 description 14
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 14
- 230000008569 process Effects 0.000 description 11
- 238000012545 processing Methods 0.000 description 7
- 230000002776 aggregation Effects 0.000 description 6
- 238000004220 aggregation Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 239000000975 dye Substances 0.000 description 5
- 239000006285 cell suspension Substances 0.000 description 4
- 238000000684 flow cytometry Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 239000007788 liquid Substances 0.000 description 3
- 239000003550 marker Substances 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 230000000712 assembly Effects 0.000 description 2
- 238000000429 assembly Methods 0.000 description 2
- 230000020411 cell activation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000002715 modification method Methods 0.000 description 2
- 208000019838 Blood disease Diseases 0.000 description 1
- 206010028980 Neoplasm Diseases 0.000 description 1
- 101710160107 Outer membrane protein A Proteins 0.000 description 1
- 230000001594 aberrant effect Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000001772 blood platelet Anatomy 0.000 description 1
- 210000001124 body fluid Anatomy 0.000 description 1
- 239000010839 body fluid Substances 0.000 description 1
- 210000002798 bone marrow cell Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000024245 cell differentiation Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 210000003850 cellular structure Anatomy 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 239000007850 fluorescent dye Substances 0.000 description 1
- 208000014951 hematologic disease Diseases 0.000 description 1
- 210000003958 hematopoietic stem cell Anatomy 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 210000000265 leukocyte Anatomy 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000001613 neoplastic effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
- 210000004881 tumor cell Anatomy 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N15/14—Electro-optical investigation, e.g. flow cytometers
- G01N15/1434—Electro-optical investigation, e.g. flow cytometers using an analyser being characterised by its optical arrangement
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/10—Investigating individual particles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N15/14—Electro-optical investigation, e.g. flow cytometers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N15/14—Electro-optical investigation, e.g. flow cytometers
- G01N15/1429—Electro-optical investigation, e.g. flow cytometers using an analyser being characterised by its signal processing
- G01N15/1431—Electro-optical investigation, e.g. flow cytometers using an analyser being characterised by its signal processing the electronics being integrated with the analyser, e.g. hand-held devices for on-site investigation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G01N15/01—
-
- G01N15/149—
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N2015/1006—Investigating individual particles for cytology
-
- G01N2015/1028—
-
- G01N2015/1029—
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N15/14—Electro-optical investigation, e.g. flow cytometers
- G01N2015/1477—Multiparameters
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N15/14—Electro-optical investigation, e.g. flow cytometers
- G01N2015/1493—Particle size
Abstract
The present invention provides a kind of fluidic cell intelligent immunity classifying method, device and electronic equipments, by obtaining position coordinates of each cell in the coordinate system using cell surface difference antigen molecular as reference axis in streaming sample；The cell in the streaming sample is divided into multiple cell masses according to the position coordinates；Identify the respective cell category of the multiple cell mass；Judge whether the position coordinates of cell in each cell mass are in preset range corresponding to the cell category of the cell mass；And when judging result is that there are the position coordinates of cell to be not in preset range corresponding to the cell category of the cell mass in the cell mass, determine the cell mass for abnormal group.The judgement for dividing group and neural network model to realize abnormal group that cell is realized using artificial intelligence, significantly reduces the labor intensity of professional, while improving the accuracy and efficiency of fluidic cell immunophenotyping.
Description
Technical field
The present invention relates to flow cyctometry fields, and in particular to a kind of fluidic cell intelligent immunity classifying method and device,
Electronic equipment, computer program product and computer readable storage medium.
Background technique
The cell surface antigen molecules that haemocyte appears or disappears during the different phase of differentiation and cell activation
It is referred to as cell differentiation group, there is different points in erythron, leukon, blood platelet, megakaryocytic series and non-hematopoietic cell
Change epitope cluster expression.When neoplastic hematologic disorder occurs, cell will lose serial specificity and the rule of differential period of normal cell
Rule property.Flow cytometry can be to determining its source and differential period, judge that the presence of minimal residue lesion and supposition prognosis are equal
It is valuable.
Existing streaming immunophenotyping method is quantitatively to carry out a point group to born of the same parents using flow cytometer, and pass through profession mostly
Personnel carry out normal or abnormal judgement to each cell colony from two-dimensional visual angle, and are pushed away according to the feature of aberrant cell populations
Survey type and the degree etc. of disease.
However, the prior art has the disadvantage in that
1. subjectivity is big.The experience of each professional is different and judgment basis is also not exactly the same, therefore, judges to tie
The judging result that fruit will also be deviated or even the same person provides under varying environment, state is also not necessarily identical.
2. large labor intensity.The personnel of judgement need certain technical foundation and working experience, and actual huge work
Work amount and the opposite professional lacked increase the labor intensity of staff.
3. low efficiency.It relies on relatively small number of professional to go to be accomplished manually huge workload, working efficiency is natural
It is not high.
Summary of the invention
In view of this, the embodiment of the present invention is dedicated to providing a kind of fluidic cell intelligent immunity classifying method and device, lead to
The mode for crossing artificial intelligence substitutes existing manual operation, and the subjectivity for solving the prior art is big, large labor intensity, Yi Jixiao
The problems such as rate is low.
According to an aspect of the present invention, a kind of fluidic cell intelligent immunity classifying method that one embodiment of the invention provides,
It include: to obtain position of each cell in the coordinate system using cell surface difference antigen molecular as reference axis in streaming sample
Coordinate；The cell in the streaming sample is divided into multiple cell masses according to the position coordinates；Identify the multiple cell mass
Respective cell category；Judge whether the position coordinates of cell in each cell mass are in the thin of the cell mass
In preset range corresponding to born of the same parents' type；And when judging result be the cell mass in there are the position coordinates of cell not
In preset range corresponding to cell category in cell mass described in this, determine the cell mass for abnormal group；Wherein, sentence
Whether the position coordinates of cell in each cell mass that break are in pre- corresponding to the cell category of the cell mass
If realized based on first nerves network model in range.
In one embodiment, whether the position coordinates for judging cell in each cell mass it is described be in this
Include: in preset range corresponding to the cell category of cell mass by the position coordinates of cell in individual cells group with it is described
The cell category of individual cells group inputs the first nerves network model, by described in first nerves network model judgement
The position coordinates in individual cells group with the presence or absence of cell are not at pre- corresponding to the cell category of the cell mass
If in range.
In one embodiment, the training process of the first nerves network model includes: by individual cells in the coordinate
The cell category of the position coordinates and the individual cells in system is as sample input, the position of the individual cells
Whether coordinate is in the result in preset range corresponding to the cell category of the individual cells as sample and exports training
The first nerves network model.
In one embodiment, each cell is being to sit with cell surface difference antigen molecular in the acquisition streaming sample
Position coordinates in the coordinate system of parameter include: choose the cell surface a variety of antigen moleculars be reference axis, obtain
Take the position coordinates in each cell coordinate system composed by the reference axis.
In one embodiment, the position obtained in each cell coordinate system composed by the reference axis is sat
Mark includes: that the partial coordinates axis chosen in the reference axis respectively forms multiple coordinate systems, obtains each cell respectively and exists
Position coordinates in the multiple coordinate system.
In one embodiment, described that the cell in streaming sample is divided into multiple cell mass packets according to the position coordinates
It includes: the position coordinates being inputted into nervus opticus network model, by the nervus opticus network model by the streaming sample
In cell be divided into multiple cell masses.
In one embodiment, the training process of the nervus opticus network model includes: by individual cells in the coordinate
The position coordinates in system export training institute as sample as sample input, the corresponding cell mass image of the individual cells
State nervus opticus network model.
In one embodiment, the method further includes: determine the intensity of anomaly of the abnormal group.
In one embodiment, the intensity of anomaly of the determination abnormal group includes: to be existed according to the center of gravity of the abnormal group
Position of the position coordinates normal cell corresponding with the exception cell category of group in the coordinate system in the coordinate system
The difference for setting coordinate determines the intensity of anomaly of the abnormal group.
In one embodiment, the intensity of anomaly of the determination abnormal group includes: by abnormal group's center of gravity described
The cell category of position coordinates and the abnormal group in coordinate system inputs third nerve network model, passes through the third nerve
Network model determines the intensity of anomaly of the abnormal group.
In one embodiment, the training process of the third nerve network model includes: by single abnormal cell described
The cell category of position coordinates in coordinate system and the single abnormal cell inputs as sample, the single abnormal cell
Intensity of anomaly is as the sample output training third nerve network model.
In one embodiment, the method also includes: the position coordinates are compensated, according to compensated institute's rheme
It sets coordinate and the cell in the streaming sample is divided into multiple cell masses.
In one embodiment, described compensate to the position coordinates includes: by the corresponding coordinate of the position coordinates
Vector obtains the compensated position coordinates, wherein the compensation matrix is to describe multiplied by the inverse matrix of compensation matrix
State the cell influence degree of various colors to each Color Channel in dyeing course.
In one embodiment, before being compensated to the position coordinates, further includes: whether judge the compensation matrix
Accurately；If judging result is the compensation matrix inaccuracy, the compensation matrix is corrected.
In one embodiment, described to judge whether the compensation matrix accurately includes: the position for calculating the cell
The distance between the coordinate straight line of point composition all equal with each coordinate components in coordinate system value；Calculate the streaming sample
Described in distance value be less than pre-determined distance threshold value cell quantity accounting；And if the quantity accounting is preset greater than first and is accounted for
When than threshold value, it is determined that the compensation matrix inaccuracy.
In one embodiment, the amendment compensation matrix includes: the position coordinates and the institute for calculating the cell
State the distance between all equal straight line of point composition of each coordinate components in coordinate system value；Calculate the distance value be less than it is default away from
The quantity accounting of the cell from threshold value；And the element value in the adjustment compensation matrix, so that the compensated number
Accounting is measured less than the second default accounting threshold value.
In one embodiment, each coordinate components are all in the position coordinates and the coordinate system for calculating the cell
The distance between the equal straight line of point composition value includes: to add to each coordinate components of the position coordinates of the cell
Power processing calculates the straight of weighting treated the position coordinates point composition all equal with each coordinate components in the coordinate system
The distance between line value.
In one embodiment, the multiple respective cell category of cell mass of identification includes: by laser irradiation institute
It states scattered light signal caused by multiple cell masses and identifies the multiple respective cell category of cell mass.
In one embodiment, the scattered light signal includes forward angle light scatter optical signal and lateral scattering optical signal.
According to another aspect of the present invention, a kind of fluidic cell intelligent immunity parting dress that one embodiment of the invention provides
It sets, comprising: coordinate obtaining module, be configured to obtain each cell in streaming sample is being with cell surface difference antigen molecular
Position coordinates in the coordinate system of reference axis；Grouping module, being configured to will be in the streaming sample according to the position coordinates
Cell is divided into multiple cell masses；Category identification module is configured to identify the respective cell category of the multiple cell mass；Abnormal group
Judgment module is configured to judge whether the position coordinates of cell in each cell mass are in the cell
In preset range corresponding to the cell category of group；And abnormal group's determining module, it is configured to when judging result be the cell
There are the position coordinates of cell to be not in preset range corresponding to the cell category of the cell mass in group, determines
The cell mass is abnormal group；Wherein, judge whether the position coordinates of cell in each cell mass are in the institute
State in preset range corresponding to the cell category of cell mass is realized based on first nerves network model.
In one embodiment, abnormal group's judgment module, which is configured that, sits the position of cell in individual cells group
Mark and the cell category of the individual cells group input the first nerves network model, pass through the first nerves network model
Judge that the position coordinates in the individual cells group with the presence or absence of cell are not at the cell category institute of the cell mass
In corresponding preset range.
In one embodiment, abnormal group's judgment module includes: the first training unit, is configured to individual cells in institute
The cell category of the position coordinates and the individual cells in coordinate system is stated as sample input, the institute of the individual cells
It is defeated as sample to state the result whether position coordinates are in preset range corresponding to the cell category of the individual cells
The first nerves network model is trained out.
In one embodiment, the coordinate obtaining module is configured that a variety of antigens point for choosing the cell surface
Son amount is reference axis, obtains the position coordinates in each cell coordinate system composed by the reference axis.
In one embodiment, the coordinate obtaining module is configured that the partial coordinates axis chosen in the reference axis respectively
Multiple coordinate systems are formed, obtain position coordinates of each cell in the multiple coordinate system respectively.
In one embodiment, the grouping module, which is configured that, inputs nervus opticus network model for the position coordinates, leads to
It crosses the nervus opticus network model and the cell in the streaming sample is divided into multiple cell masses.
In one embodiment, the grouping module includes: the second training unit, is configured to individual cells in the coordinate
The position coordinates in system export training institute as sample as sample input, the corresponding cell mass image of the individual cells
State nervus opticus network model.
In one embodiment, described device further comprises: intensity of anomaly determining module, is configured to determine the abnormal group
Intensity of anomaly.
In one embodiment, the intensity of anomaly determining module is configured that according to the center of gravity of the abnormal group in the seat
Position coordinates of the position coordinates normal cell corresponding with the exception cell category of group in the coordinate system in mark system
Difference determine the intensity of anomaly of the abnormal group.
In one embodiment, the intensity of anomaly determining module is configured that abnormal group's center of gravity in the coordinate system
In position coordinates and the cell category of the abnormal group input third nerve network model, pass through the third nerve network mould
Type determines the intensity of anomaly of the abnormal group.
In one embodiment, the intensity of anomaly determining module includes: third training unit, is configured to single exception is thin
The cell category of position coordinates of the born of the same parents in the coordinate system and the single abnormal cell inputs as sample, is described single different
The intensity of anomaly of normal cell is as the sample output training third nerve network model.
In one embodiment, described device further include: compensating module is configured to compensate the position coordinates, point
Cell in the streaming sample is divided into multiple cell masses according to the compensated position coordinates by group's module.
In one embodiment, the compensating module is configured that the corresponding coordinate vector of the position coordinates multiplied by compensation
Inverse of a matrix matrix obtains the compensated position coordinates, wherein the compensation matrix is being dyed to describe the cell
Influence degree of the various colors to each Color Channel in the process.
In one embodiment, the compensating module configuration includes: accuracy judging unit, is configured to sit to the position
Before mark compensates, judge whether the compensation matrix is accurate；Amending unit is configured to when judging result be the compensation square
Battle array inaccuracy, then correct the compensation matrix.
In one embodiment, the accuracy judging unit is configured that the position coordinates for calculating the cell and institute
State the distance between all equal straight line of point composition of each coordinate components in coordinate system value；Calculate described in the streaming sample away from
Quantity accounting from the cell that value is less than pre-determined distance threshold value；And if the quantity accounting is greater than the first default accounting threshold value
When, it is determined that the compensation matrix inaccuracy.
In one embodiment, the amending unit is configured that the position coordinates and the coordinate for calculating the cell
The distance between straight line identical with the angle of all reference axis is worth in system；The distance value is calculated less than pre-determined distance threshold
The quantity accounting of the cell of value；And the element value in the adjustment compensation matrix, so that the compensated quantity accounts for
Than less than the second default accounting threshold value.
In one embodiment, the amending unit is configured that each coordinate components to the position coordinates of the cell
It is weighted processing, calculates weighting treated the position coordinates point group all equal with each coordinate components in the coordinate system
At the distance between straight line value.
In one embodiment, the category identification module is configured that by produced by the multiple cell mass of laser irradiation
Scattered light signal identify the multiple respective cell category of cell mass.
In one embodiment, the scattered light signal includes forward angle light scatter optical signal and lateral scattering optical signal.
According to another aspect of the present invention, a kind of electronic equipment that one embodiment of the invention provides, comprising: processor；It deposits
Reservoir；And the computer program instructions of storage in the memory, the computer program instructions are by the processor
The processor is made to execute as above described in any item methods when operation.
According to another aspect of the present invention, a kind of computer program product that one embodiment of the invention provides, including calculate
Machine program instruction, the computer program instructions execute the processor described in any one as above
Method.
According to another aspect of the present invention, a kind of computer readable storage medium that one embodiment of the invention provides, thereon
The step of being stored with computer program, as above any one the method realized when the computer program is executed by processor.
Fluidic cell intelligent immunity classifying method provided in an embodiment of the present invention, by obtaining each cell in streaming sample
Position coordinates in the coordinate system using cell surface difference antigen molecular as reference axis；It will be described according to the position coordinates
Cell in streaming sample is divided into multiple cell masses；Identify the respective cell category of the multiple cell mass；Judgement is each described
Whether the position coordinates of cell are in preset range corresponding to the cell category of the cell mass in cell mass；With
And when judging result is that there are the cell categories that the position coordinates of cell are not at the cell mass in the cell mass
In corresponding preset range, determine the cell mass for abnormal group；Wherein, judge the institute of cell in each cell mass
Whether state position coordinates and be in preset range corresponding to the cell category of the cell mass is based on first nerves network
Model realization.The judgement for dividing group and neural network model to realize abnormal group that cell is realized using artificial intelligence, is significantly reduced
The labor intensity of professional, while improving the accuracy and efficiency of fluidic cell immunophenotyping.
Detailed description of the invention
Fig. 1 show the flow chart of the fluidic cell intelligent immunity classifying method of one embodiment of the application offer.
Fig. 2 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.
Fig. 3 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.
Fig. 4 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.
Fig. 5 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.
Fig. 6 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.
Fig. 7 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.
Fig. 8 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.
Fig. 9 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.
Figure 10 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.
Figure 11 show the flow chart of the compensation matrix accuracy judgment method of one embodiment of the application offer.
Figure 12 show the flow chart of the compensation matrix modification method of one embodiment of the application offer.
Figure 13 show the structural schematic diagram of the fluidic cell intelligent immunity parting device of one embodiment of the application offer.
Figure 14 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
Figure 15 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
Figure 16 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
Figure 17 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
Figure 18 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
Figure 19 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
Figure 20 show the structural schematic diagram of the electronic equipment of one embodiment of the invention offer.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that the described embodiment is only a part of the embodiment of the present invention, instead of all the embodiments.Based on this
Embodiment in invention, every other reality obtained by those of ordinary skill in the art without making creative efforts
Example is applied, shall fall within the protection scope of the present invention.
Flow cytometry working principle is by monoclonal antibody on cellular and molecular level to individual cells or other lifes
Object particle carries out multi-parameter, quick quantitative analysis, it can be with high speed analysis a cells up to ten thousand, and can be simultaneously from a cell
Multiple parameters are measured, have the advantages that speed is fast, precision is high, accuracy is good, are the present age state-of-the-art cell quantitative technologies
One of.It is with the detection that flow cytometer carries out peripheral white blood cells, bone marrow cell and tumour cell etc. in clinic at present
The important component of clinical detection.
Flow cytometry mainly realizes that the working principle of flow cytometer is to make to float on a liquid on flow cytometer
The cell through fluorescent marker or particle of dispersion pass through sample cell one by one, while being captured fluorescence signal by fluorescent probe and being converted
At the electric impulse signal for respectively representing angle of scattering and different fluorescence intensities, corresponding point diagram, histogram are formed through computer processing
With add three-dimensional structure image to be analyzed.
Flow cytometer specifically includes that liquid fluid system, including flow chamber and liquid stream drive system, for realizing the stream of sample
It is dynamic；Optical system, including excitation light source and beam collection system, for realizing the fluorescent marker and collection of cell or particle；Electricity
Subsystem, including photoelectric converter and data processing system, for realizing the fluorescence information of cell is converted to electric signal and is obtained
To the related data information of cell.Wherein, it is single cell suspension for the sample of flow cytometer, can be blood, suspends carefully
Born of the same parents' culture solution, various body fluid, the single cell suspension of fresh solid tumor and single cell suspension of paraffin-embedded tissue etc..
After flow cytometry is point diagram for obtaining single cell suspension using flow cytometer etc. at present, then by professional couple
Point diagram carries out the abnormal judgement that cell divides group and cell mass, due to can only manually realize the judgement of two-dimensional coordinate, artificial to grasp
Need to judge in multiple two-dimensional coordinate systems the position characteristic of cell mass when making, and comprehensive analysis obtains final diagnostic result.
However, different personnel can generate diagnosis difference when operating, and the limited amount of professional, workload are huge, cause people
Work workload is huge, while low efficiency.
The application proposes that a kind of fluidic cell intelligent immunity classifying method, device and electronics are set for solving the above problems
It is standby, existing manual operation is substituted by artificial intelligence, realizes that cell divides the abnormal judgement of group and cell mass, to be diagnosed
As a result, avoid because experience and the judgment basis of different personnel are different and caused by diagnose difference, while being greatly reduced artificial
Workload, improve efficiency.
In the following, illustrate in conjunction with the drawings and the specific embodiments the fluidic cell intelligent immunity classifying method of the application, device and
The implementation of electronic equipment:
Fig. 1 show the flow chart of the fluidic cell intelligent immunity classifying method of one embodiment of the application offer.Such as Fig. 1 institute
Show, which includes the following steps:
Step 110: each cell is using cell surface difference antigen molecular as the seat of reference axis in acquisition streaming sample
Position coordinates in mark system.
As described above, haemocyte has different epitope cluster expression during the different phase of differentiation and cell activation
(i.e. antigen molecular) obtains each of the streaming sample for passing through fluorescent marker using different antigen moleculars as reference axis
Position coordinates of the cell in the coordinate system constituted with the reference axis.Coordinate value of the cell in reference axis indicates the cell at this
The carrying quantity of antigen molecule representated by reference axis, i.e., by the position coordinates of cell as the various antigen molecules entrained by it
Quantity determined.
Step 120: the cell in streaming sample being divided into multiple cell masses according to position coordinates.
All cells in streaming sample are divided into multiple cell masses according to the position coordinates of each cell, wherein each
Cell mass is same kind of cell.The case where antigen molecule as entrained by same kind of cell, is usually identical,
Even abnormal cell (being typically not individualism), can also have antigen molecule entrained by the cell of a group exception
Situation is identical, therefore, all cells in sample can be divided into multiple cell masses.
Step 130: identifying the respective cell category of multiple cell masses.
As described above, the case where antigen molecule entrained by cell in the same cell mass, is identical, therefore, according to each
The position coordinates of cell in cell mass, can each cell mass of automatic identification cell category.
Step 140: judge the position coordinates of cell in each cell mass whether be in the cell mass cell category institute it is right
In the preset range answered.
It is normal although different human body or in the different growth courses of cell for the cell of each type
The case where antigen molecule that cell carries, will be different, still, the quantity of antigen molecule entrained by the cell of each type
Have a normal range.Therefore, the normal range can be used as judge cell whether Yi Chang foundation, that is, pass through judgement
The position coordinates of cell in each cell mass whether normal range corresponding to the cell category all in the cell mass, sentence
Whether extremely the cell mass break.For example, the normal range that the cell of a certain type carries the quantity of a certain antigen molecule is
200-500, then can be by judging the coordinate position of cell in the cell mass in the corresponding reference axis of this kind of antigen molecule
Whether component is in 200-500, to judge whether the cell mass is abnormal.
Step 150: when judging result is that there are the cell kinds that the position coordinates of cell are not at the cell mass in cell mass
In preset range corresponding to class, determine the cell mass for abnormal group.Wherein, judge the position coordinates of cell in each cell mass
Whether be in is to be realized in preset range corresponding to the cell category of the cell mass based on first nerves network model.
Since the cell of each cell mass is same kind of cell, wherein normally with abnormal same category cell
For different types, therefore, when there are a cell being abnormal cell in cell mass, it can determine that the cell mass is abnormal thin
Born of the same parents group.I.e. and when above-mentioned judging result is that there are the position coordinates of cell not in the normal range in the cell mass,
Determine the cell mass for abnormal group.It may be implemented to judge the position of the cell in each cell mass by first nerves network model
Coordinate is set whether in range normal corresponding to the cell category of the cell mass, and with this come determine the cell mass whether be
Abnormal group.
The embodiment of the present application divides group and mind using what the method and neural network model of artificial intelligence realized cell automatically
Realize that the judgement of abnormal group significantly reduces the labor intensity of professional instead of existing manual operation through network model,
The accuracy and efficiency of fluidic cell immunophenotyping are improved simultaneously.
Fig. 2 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.Such as Fig. 2
Shown, step 140 may include sub-step:
Step 145: by the first mind of the position coordinates of cell in individual cells group and the input of the cell category of individual cells group
Through network model, judge that the position coordinates in individual cells group with the presence or absence of cell are not at this by first nerves network model
In preset range corresponding to the cell category of cell mass.
Using the position coordinates of the cell in individual cells group and the cell category of individual cells group as first nerves network
The input of model, by first nerves network model export to obtain in individual cells group with the presence or absence of cell position coordinates not
In preset range corresponding to cell category in the cell mass.Each cell is judged automatically using neural network model realization
Whether group abnormal, significantly reduces the labor intensity of professional, at the same improve fluidic cell immunophenotyping accuracy and
Efficiency.
In one embodiment, the training process of first nerves network model may include: by individual cells in a coordinate system
Position coordinates and the cell categories of individual cells as the position coordinates of sample input, individual cells whether to be in this single thin
Result in preset range corresponding to the cell category of born of the same parents exports training first nerves network model as sample.
When training first nerves network model, by the cell of individual cells position coordinates in a coordinate system and individual cells
Whether type is in preset range corresponding to the cell category of the individual cells as the position coordinates of input, individual cells
Result as output, first nerves network model is trained with the sample of input-output.Pass through the instruction of big data sample
Practice, can to avoid the error of judging result caused by the subjective factor in manual operation, thus, improve the accuracy of judgement.
Fig. 3 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.Such as Fig. 3
Shown, step 110 may include sub-step:
Step 115: a variety of antigen moleculars for choosing cell surface are reference axis, obtain each cell in reference axis institute group
At coordinate system in position coordinates.
Cell in streaming sample needs that various antigen molecules could be distinguished after the dyeing of dyestuff, and cell surface is taken
The type of the antigen molecule of band is more, however the color category of dyestuff is limited and expensive.Therefore, for save cost and
The necessity of analysis considers, a variety of antigen moleculars for having an impact or being affected to analysis result are chosen according to the demand of analysis
As reference axis, and position coordinates of each cell in the coordinate system composed by the reference axis are obtained, according to the position coordinates
Cell in convection type sample carries out point group and abnormal group's judgement.
Fig. 4 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.Such as Fig. 4
Shown, step 115 may include sub-step:
Step 1151: the partial coordinates axis chosen in reference axis respectively forms multiple coordinate systems, obtains each cell respectively
Position coordinates in multiple coordinate systems.
As described above, the color category of dyestuff is limited, and the type that will appear color during analysis is not enough to prop up
Hold the type of antigen molecule required for analysis.In order to solve this problem, the embodiment of the present application will propose required for analyzing
Reference axis of the different antigen moleculars as multiple coordinate systems, is obtained finally by the combination of the judging result of multiple coordinate systems
Judging result, under the conditions of the color category of dyestuff is limited complete analysis work.
It, can be with it should be appreciated that be to be mutually related between each reference axis in the embodiment of the present application in each coordinate system
All or part of judging result is obtained by the combination of the reference axis, i.e., may be implemented by the combination of reference axis a certain
Specific judging result or a certain foundation for final judging result.
It should be appreciated that the embodiment of the present application can choose the coordinate system and seat of different number according to the different demands of analysis
The combination of different reference axis in mark system, and the same antigen molecular can be used as the reference axis in different coordinates, as long as
The combination of reference axis in selected coordinate system quantity and coordinate system can be realized analysis work, and the application is implemented
Example for the reference axis in coordinate system quantity and coordinate system combination without limitation.
Fig. 5 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.Such as Fig. 5
Shown, step 120 may include sub-step:
Step 125: position coordinates being inputted into nervus opticus network model, by nervus opticus network model by streaming sample
In cell be divided into multiple cell masses.
By establishing nervus opticus network model, the cell in streaming sample is divided into using nervus opticus network model more
A cell mass, that realizes cell divides group automatically, reduces manually-operated workload and improves the efficiency that cell divides group, simultaneously
Also data basis is provided for the abnormal judgement of subsequent cell mass.
In one embodiment, the training process of nervus opticus network model may include: by individual cells in a coordinate system
Position coordinates as sample input, the corresponding cell mass image of individual cells as the trained nervus opticus network mould of sample output
Type.
When training nervus opticus network model, using the position coordinates of individual cells in a coordinate system as input, it is single carefully
The corresponding cell mass image of born of the same parents is trained nervus opticus network model with the sample of input-output as output.By big
The training of data sample, can to avoid the error of grouping result caused by the subjective factor in manual operation, thus, improve point
The accuracy of group.
Fig. 6 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.Such as Fig. 6
Shown, the above method can further comprise:
Step 160: determining the intensity of anomaly of exception group.
After having determined a certain cell mass for abnormal group, it is also necessary to further determine that the intensity of anomaly of exception group.According to
The type of the abnormal available disease of cell mass, but it is unable to get the severity of disease, therefore, it is necessary to further to different
The intensity of anomaly of Chang Qun is determined, and to obtain the severity of disease, obtains final analysis result.
In one embodiment, the intensity of anomaly for determining abnormal group includes: the position according to the center of gravity of abnormal group in a coordinate system
The difference for setting the position coordinates of coordinate normal cell corresponding with the abnormal cell category of group in a coordinate system determines abnormal group's
Intensity of anomaly.
It is simple to be difficult to determine abnormal group by numerical value since everyone has its otherness in embodiments herein
Accurate intensity of anomaly can also judge exception group's by comparing the relative positional relationship of abnormal group and normal cell mass
Intensity of anomaly, in this way can error caused by the otherness to avoid different human body, improve the precision of analysis.
Fig. 7 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.Such as Fig. 7
Shown, step 160 may include sub-step:
Step 165: by abnormal group's center of gravity position coordinates in a coordinate system and the cell category input third mind of abnormal group
Through network model, the intensity of anomaly of abnormal group is determined by third nerve network model.
By establishing third nerve network model, the intensity of anomaly of abnormal group is determined using third nerve network model, from
And the severity of disease is obtained, it realizes automatically analyzing for disease, reduces manually-operated workload and improve disease point
The efficiency of analysis.
In one embodiment, the training process of third nerve network model include: by single abnormal cell in a coordinate system
Position coordinates and the cell category of single abnormal cell is inputted as sample, the intensity of anomaly of single abnormal cell is as sample
Export training third nerve network model.
It is when training third nerve network model, the position coordinates of single abnormal cell in a coordinate system and single exception are thin
The cell category of born of the same parents is used as output as the intensity of anomaly of input, single abnormal cell, with the sample of input-output to third mind
It is trained through network model.It, can be to avoid caused by the subjective factor in manual operation points by the training of big data sample
The error of result is analysed, thus, improve the accuracy of diseases analysis.
Fig. 8 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.Such as Fig. 8
Shown, the method for the embodiment of the present application can further comprise:
Step 170: position coordinates are compensated.
Certain interference can be generated under different use environments or during prolonged use due to flow cytometer,
Wherein, biggish interference is the cell influence of different colours to each Color Channel in dyeing course, for example, blue is to red
Channel should be do not have it is influential, but in actual application, blue may red channel generate one it is lesser dry
It disturbs, to produce interference to the position coordinates of cell, leading to the position coordinates of the cell is not its accurate position coordinates.Cause
This, before step 120 can also compensate position coordinates, to restore the accurate position coordinates of each cell, then
The cell in streaming sample is divided into multiple cell masses according to compensated position coordinates.
Fig. 9 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.Such as Fig. 9
Shown, step 170 may include sub-step:
Step 175: by the corresponding coordinate vector of position coordinates multiplied by the inverse matrix of compensation matrix, obtaining compensated position
Coordinate, wherein compensation matrix is to describe the cell influence degree of various colors to each Color Channel in dyeing course.
By obtaining accurate compensation matrix, carried out after compensation is calculated using initial position coordinates and compensation matrix
Position coordinates, the accurate location coordinate of cell can be obtained, simply to guarantee that subsequent accurate cell divides group and abnormal group
Judgement provides accurate data and supports.
Figure 10 show the flow chart for the fluidic cell intelligent immunity classifying method that another embodiment of the application provides.Such as figure
Shown in 10, before step 170, can also include:
Step 180: judging whether compensation matrix is accurate.
Step 190: when judging result is compensation matrix inaccuracy, then correction-compensation matrix.
It before compensating position coordinates, needs to judge the accuracy of compensation matrix, only accurately mend
Accurate position coordinates can just be obtained by repaying matrix, to guarantee that subsequent accurate cell divides group and abnormal group's judgement to provide accurately
Data are supported.
Figure 11 show the flow chart of the compensation matrix accuracy judgment method of one embodiment of the application offer.Such as Figure 11 institute
Show, step 180 may include sub-step:
Step 181: calculate the position coordinates of the cell point composition all equal with coordinate components each in coordinate system straight line it
Between distance value.
Step 182: calculating the quantity accounting that distance value in streaming sample is less than the cell of pre-determined distance threshold value.
Step 183: judging whether quantity accounting is greater than the first default accounting threshold value, accounted for when quantity accounting is default greater than first
When than threshold value, 184 are gone to step；Otherwise, 185 are gone to step.
Step 184: determining compensation matrix inaccuracy.
Step 185: determining that compensation matrix is accurate.
Pass through the cell aggregation degree judgement around the straight line of all equal point composition of each coordinate components in coordinates computed system
The accuracy of compensation matrix is realized the judgement of the cell aggregation degree of various dimensions using artificial intelligence, substitutes artificial multiple two dimensions
The comprehensive descision of degree further improves the efficiency of entire analytic process.
Figure 12 show the flow chart of the compensation matrix modification method of one embodiment of the application offer.As shown in figure 12, it walks
Rapid 190 may include sub-step:
Step 191: calculate the position coordinates of the cell point composition all equal with coordinate components each in coordinate system straight line it
Between distance value.
Step 192: calculating the quantity accounting that distance value is less than the cell of pre-determined distance threshold value.
Step 193: the element value in adjustment compensation matrix, so that compensated quantity accounting is less than the second default accounting threshold
Value.
By adjusting the element value in compensation matrix, so that the straight line that all equal point of each coordinate components forms in coordinate system
The cell aggregation degree of surrounding is minimum, that is, is gathered in all equal cell accounting put around the straight line formed of each coordinate components most
It is few.The adjustment that the compensation matrix of various dimensions is realized using artificial intelligence, substitutes the multiple adjustment of artificial multiple two-dimensions, into one
Step improves the efficiency and accuracy of entire analytic process.
In one embodiment, above-mentioned steps 181 and step 191 can include: each coordinate to the position coordinates of cell
Component is weighted processing, calculates weighting treated the position coordinates point composition all equal with coordinate components each in coordinate system
The distance between straight line value.
In one embodiment, step 130 may include: by scattered light signal caused by the multiple cell masses of laser irradiation
Identify the respective cell category of multiple cell masses.Preferably, scattered light signal includes forward angle light scatter optical signal and lateral scattering
Optical signal.
Laser irradiation cell can produce multiple scattered light signals, and forward angle light scatter optical signal can reflect the big of cell
Small, lateral scattering optical signal can reflect the inside labyrinth of cell, pass through forward angle light scatter optical signal and side scattered light
Signal can identify the classification of cell.It should be appreciated that the position that can be combined with cell in cell mass in the embodiment of the present application is sat
Mark and the comprehensive type for identifying cell mass of scattered light signal, to improve the precision of identification.
Figure 13 show the structural schematic diagram of the fluidic cell intelligent immunity parting device of one embodiment of the application offer.Such as
Shown in Figure 13, which includes: coordinate obtaining module 1, for obtaining in streaming sample each cell with cell surface difference
Antigen molecular is the position coordinates in the coordinate system of reference axis；Grouping module 2, being used for will be in streaming sample according to position coordinates
Cell be divided into multiple cell masses；Category identification module 3, for identification respective cell category of multiple cell masses；Abnormal group sentences
Disconnected module 4, for judging whether the position coordinates of cell in each cell mass are in corresponding to the cell category of the cell mass
In preset range；And abnormal group's determining module 5, for being that there are the position coordinates of cell not to locate in cell mass when judging result
In the preset range corresponding to the cell category of the cell mass, determine the cell mass for abnormal group；Wherein, judge each cell
It is based on first nerves that whether the position coordinates of cell, which are in preset range corresponding to the cell category of the cell mass, in group
Network model is realized.
The embodiment of the present application divides group and neural network mould using what fluidic cell intelligent immunity parting shape realized cell automatically
Type realizes that the judgement of abnormal group significantly reduces the labor intensity of professional, improve simultaneously instead of existing manual operation
The accuracy and efficiency of fluidic cell immunophenotyping.
In one embodiment, abnormal group's judgment module 4 can be configured to: by the position coordinates and list of cell in individual cells group
The cell category of a cell mass inputs first nerves network model, judged by first nerves network model be in individual cells group
The no position coordinates there are cell are not in preset range corresponding to the cell category of the cell mass.
Using the position coordinates of the cell in individual cells group and the cell category of individual cells group as first nerves network
The input of model, by first nerves network model export to obtain in individual cells group with the presence or absence of cell position coordinates not
In preset range corresponding to cell category in the cell mass.Each cell is judged automatically using neural network model realization
Whether group abnormal, significantly reduces the labor intensity of professional, at the same improve fluidic cell immunophenotyping accuracy and
Efficiency.
Figure 14 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
As shown in figure 14, abnormal group's judgment module 4 can include: the first training unit 41, for the position by individual cells in a coordinate system
Whether the cell category for setting coordinate and individual cells is in the individual cells as the position coordinates of sample input, individual cells
Result in preset range corresponding to cell category exports training first nerves network model as sample.
In one embodiment, coordinate obtaining module 1 can be configured to: a variety of antigen moleculars for choosing cell surface are coordinate
Axis obtains position coordinates of each cell in the coordinate system composed by reference axis.
The a variety of antigen moleculars for having an impact or being affected to analysis result are chosen as coordinate according to the demand of analysis
Axis, and position coordinates of each cell in the coordinate system composed by the reference axis are obtained, according to the position coordinates convection current style
Cell in this carries out point group and abnormal group's judgement.
In one embodiment, coordinate obtaining module 1 can be configured to: the partial coordinates axis composition chosen in reference axis respectively is more
A coordinate system obtains position coordinates of each cell in multiple coordinate systems respectively.
Final judging result is obtained by the combination of the judging result of multiple coordinate systems, it is limited in the color category of dyestuff
Under conditions of complete analysis work.
In one embodiment, grouping module 2 can be configured to: position coordinates be inputted nervus opticus network model, by the
Cell in streaming sample is divided into multiple cell masses by two neural network models.
By establishing nervus opticus network model, the cell in streaming sample is divided into using nervus opticus network model more
A cell mass, that realizes cell divides group automatically, reduces manually-operated workload and improves the efficiency that cell divides group, simultaneously
Also data basis is provided for the abnormal judgement of subsequent cell mass.
Figure 15 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
As shown in figure 15, grouping module 2 can include: the second training unit 21, for the position coordinates by individual cells in a coordinate system
Training nervus opticus network model is exported as sample as sample input, the corresponding cell mass image of individual cells.
When training nervus opticus network model, using the position coordinates of individual cells in a coordinate system as input, it is single carefully
The corresponding cell mass image of born of the same parents is trained nervus opticus network model with the sample of input-output as output.By big
The training of data sample, can to avoid the error of grouping result caused by the subjective factor in manual operation, thus, improve point
The accuracy of group.
Figure 16 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
As shown in figure 16, device can further include: intensity of anomaly determining module 6, for determining the intensity of anomaly of abnormal group.?
After having determined a certain cell mass for abnormal group, it is also necessary to be further determined to the intensity of anomaly of abnormal group, to obtain the disease
The severity of disease, obtains final analysis result.
In one embodiment, intensity of anomaly determining module 6 is configured that the position according to the center of gravity of abnormal group in a coordinate system
The difference of the position coordinates of coordinate normal cell corresponding with the abnormal cell category of group in a coordinate system determines that abnormal group's is different
Chang Chengdu.
In one embodiment, intensity of anomaly determining module 6 is configured that the position coordinates by abnormal group's center of gravity in a coordinate system
And the cell category of abnormal group inputs third nerve network model, and the abnormal journey of abnormal group is determined by third nerve network model
Degree.
By establishing third nerve network model, the intensity of anomaly of abnormal group is determined using third nerve network model, from
And the severity of disease is obtained, it realizes automatically analyzing for disease, reduces manually-operated workload and improve disease point
The efficiency of analysis.
Figure 17 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
As shown in figure 17, intensity of anomaly determining module 6 can include: third training unit 61 is used for single abnormal cell in coordinate system
In position coordinates and the cell category of single abnormal cell is inputted as sample, the intensity of anomaly of single abnormal cell is as sample
Third nerve network model is trained in this output.
It is when training third nerve network model, the position coordinates of single abnormal cell in a coordinate system and single exception are thin
The cell category of born of the same parents is used as output as the intensity of anomaly of input, single abnormal cell, with the sample of input-output to third mind
It is trained through network model.It, can be to avoid caused by the subjective factor in manual operation points by the training of big data sample
The error of result is analysed, thus, improve the accuracy of diseases analysis.
Figure 18 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
As shown in figure 18, device may also include that compensating module 7, for compensating to position coordinates.Position coordinates are compensated,
To restore the accurate position coordinates of each cell, then the cell in streaming sample is divided into according to compensated position coordinates
Multiple cell masses.
In one embodiment, compensating module 7 can be configured to: by the corresponding coordinate vector of position coordinates multiplied by compensation matrix
Inverse matrix obtains compensated position coordinates, wherein compensation matrix to describe cell in dyeing course various colors to each
The influence degree of Color Channel.
By obtaining accurate compensation matrix, carried out after compensation is calculated using initial position coordinates and compensation matrix
Position coordinates, the accurate location coordinate of cell can be obtained, simply to guarantee that subsequent accurate cell divides group and abnormal group
Judgement provides accurate data and supports.
Figure 19 show the structural schematic diagram for the fluidic cell intelligent immunity parting device that another embodiment of the application provides.
As shown in figure 19, compensating module 7 can include: accuracy judging unit 71, for judging before being compensated to position coordinates
Whether compensation matrix is accurate；Amending unit 72 is compensation matrix inaccuracy for working as judging result, then correction-compensation matrix.
It before compensating position coordinates, needs to judge the accuracy of compensation matrix, only accurately mend
Accurate position coordinates can just be obtained by repaying matrix, to guarantee that subsequent accurate cell divides group and abnormal group's judgement to provide accurately
Data are supported.
In one embodiment, accuracy judging unit 71 is configured that in the position coordinates and coordinate system that calculate cell and respectively sits
Mark the distance between all equal straight line of point composition of component value；It calculates distance value in streaming sample and is less than pre-determined distance threshold value
The quantity accounting of cell；And if quantity accounting be greater than the first default accounting threshold value when, it is determined that compensation matrix inaccuracy.
Pass through the cell aggregation degree judgement around the straight line of all equal point composition of each coordinate components in coordinates computed system
The accuracy of compensation matrix is realized the judgement of the cell aggregation degree of various dimensions using artificial intelligence, substitutes artificial multiple two dimensions
The comprehensive descision of degree further improves the efficiency of entire analytic process.
In one embodiment, amending unit 72 is configured that in the position coordinates and coordinate system for calculating cell and all coordinates
The distance between identical straight line of the angle of axis value；Calculate the quantity accounting that distance value is less than the cell of pre-determined distance threshold value；With
And the element value in adjustment compensation matrix, so that compensated quantity accounting is less than the second default accounting threshold value.
By adjusting the element value in compensation matrix, so that the straight line that all equal point of each coordinate components forms in coordinate system
The cell aggregation degree of surrounding is minimum, that is, is gathered in all equal cell accounting put around the straight line formed of each coordinate components most
It is few.The adjustment that the compensation matrix of various dimensions is realized using artificial intelligence, substitutes the multiple adjustment of artificial multiple two-dimensions, into one
Step improves the efficiency and accuracy of entire analytic process.
In one embodiment, amending unit 72, which is configured that, is weighted place to each coordinate components of the position coordinates of cell
Reason calculates the distance between the straight line of weighting treated the position coordinates point composition all equal with coordinate components each in coordinate system
Value.
In one embodiment, category identification module 3 is configured that by scattering light caused by the multiple cell masses of laser irradiation
The respective cell category of the multiple cell masses of signal identification.Preferably, scattered light signal includes forward angle light scatter optical signal and lateral
Scattered light signal.Laser irradiation cell can produce multiple scattered light signals, and forward angle light scatter optical signal can reflect cell
Size, lateral scattering optical signal can reflect the inside labyrinth of cell, by forward angle light scatter optical signal and lateral dissipate
The classification of cell can be identified by penetrating optical signal.
Figure 20 show the structural schematic diagram of the electronic equipment of one embodiment of the invention offer.As shown in figure 20, the electronics
Equipment can be the online electronic equipment of such as medicine detector device equipped with cell type analytical equipment etc thereon, can also
Capable of being communicated with online electronic equipment to transmit the offline electronic equipment of trained machine learning model to it.
Figure 20 illustrates the block diagram of the electronic equipment according to the embodiment of the present application.
As shown in figure 20, electronic equipment 9 includes one or more processors 91 and memory 92.
Processor 20 can be central processing unit (CPU) or have data-handling capacity and/or instruction execution capability
Other forms processing unit, and can control the other assemblies in electronic equipment 9 to execute desired function.
Memory 92 may include one or more computer program products, and the computer program product may include each
The computer readable storage medium of kind form, such as volatile memory and/or nonvolatile memory.The volatile storage
Device for example may include random access memory (RAM) and/or cache memory (cache) etc..It is described non-volatile to deposit
Reservoir for example may include read-only memory (ROM), hard disk, flash memory etc..It can be deposited on the computer readable storage medium
One or more computer program instructions are stored up, processor 91 can run described program instruction, to realize this Shen described above
The area marking method of each embodiment please and/or other desired functions.In the computer readable storage medium
In can also store the various contents such as the position coordinates of cell, coordinate system, training sample in streaming sample.
In one example, electronic equipment 9 can also include: input unit 93 and output device 94, these components pass through
The interconnection of bindiny mechanism's (not shown) of bus system and/or other forms.
For example, the input unit 93 can be data acquisition means, the position for acquiring the cell in streaming sample is sat
Mark, location coordinate information collected can be stored in memory 92 for other assemblies use.It is of course also possible to utilize
Other integrated or discrete data acquisition means acquire the location coordinate information of cell in the streaming sample, and it is sent
To electronic equipment 9.In addition, the input equipment 93 can also include such as keyboard, mouse and communication network and its be connected
Remote input equipment etc..
The output device 94 can export various information, including determination to external (for example, user or machine learning model)
Cell type, cell mass information, abnormal group's information, training sample out etc..The output equipment 94 may include such as display,
Loudspeaker, printer and communication network and its remote output devices connected etc..
Certainly, to put it more simply, illustrated only in Figure 20 it is some in component related with the application in the electronic equipment 9,
The component of such as bus, input/output interface etc. is omitted.In addition to this, according to concrete application situation, electronic equipment 9 is also
It may include any other component appropriate.
Other than the above method and equipment, embodiments herein can also be computer program product comprising meter
Calculation machine program instruction, the computer program instructions execute the processor in this specification to retouch
The step in the method according to the various embodiments of the application stated.
The computer program product can be write with any combination of one or more programming languages for holding
The program code of row the embodiment of the present application operation, described program design language includes object oriented program language, such as
Java, C++ etc. further include conventional procedural programming language, such as " C " language or similar programming language.Journey
Sequence code can be executed fully on the user computing device, partly execute on a user device, be independent soft as one
Part packet executes, part executes on a remote computing or completely in remote computing device on the user computing device for part
Or it is executed on server.
In addition, embodiments herein can also be computer readable storage medium, it is stored thereon with computer program and refers to
It enables, it is described in this specification according to this that the computer program instructions execute the processor
Apply for the step in the method for various embodiments.
The computer readable storage medium can be using any combination of one or more readable mediums.Readable medium can
To be readable signal medium or readable storage medium storing program for executing.Readable storage medium storing program for executing for example can include but is not limited to electricity, magnetic, light, electricity
Magnetic, the system of infrared ray or semiconductor, device or device, or any above combination.Readable storage medium storing program for executing it is more specific
Example (non exhaustive list) includes: the electrical connection with one or more conducting wires, portable disc, hard disk, random access memory
Device (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc
Read-only memory (CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.
The basic principle of the application is described in conjunction with specific embodiments above, however, it is desirable to, it is noted that in this application
The advantages of referring to, advantage, effect etc. are only exemplary rather than limitation, must not believe that these advantages, advantage, effect etc. are the application
Each embodiment is prerequisite.In addition, detail disclosed above is merely to exemplary effect and the work being easy to understand
With, rather than limit, it is that must be realized using above-mentioned concrete details that above-mentioned details, which is not intended to limit the application,.
Device involved in the application, device, equipment block diagram only as illustrative example and be not intended to require
Or hint must be attached in such a way that box illustrates, arrange, configure.As the skilled person will recognize,
It can be connected by any way, arrange, configure these devices, device, equipment, system.Such as "include", "comprise", " having "
Etc. word be open vocabulary, refer to " including but not limited to ", and can be used interchangeably with it.Vocabulary "or" used herein above
Refer to vocabulary "and/or" with "and", and can be used interchangeably with it, unless it is not such that context, which is explicitly indicated,.It is used herein above
Vocabulary " such as " refers to phrase " such as, but not limited to ", and can be used interchangeably with it.
It may also be noted that each component or each step are can to decompose in the device of the application, device and method
And/or reconfigure.These decompose and/or reconfigure the equivalent scheme that should be regarded as the application.
The above description of disclosed aspect is provided so that any person skilled in the art can make or use this
Application.Various modifications in terms of these are readily apparent to those skilled in the art, and are defined herein
General Principle can be applied to other aspect without departing from scope of the present application.Therefore, the application is not intended to be limited to
Aspect shown in this, but according to principle disclosed herein and the consistent widest range of novel feature.
In order to which purpose of illustration and description has been presented for above description.In addition, this description is not intended to the reality of the application
It applies example and is restricted to form disclosed herein.Although multiple embodiment aspect and embodiment already discussed above, this field
Its certain modifications, modification, change, addition and sub-portfolio will be recognized in technical staff.
Claims (10)
1. a kind of fluidic cell intelligent immunity classifying method characterized by comprising
Obtain position of each cell in the coordinate system using cell surface difference antigen molecular as reference axis in streaming sample
Coordinate；
The cell in the streaming sample is divided into multiple cell masses according to the position coordinates；
Identify the respective cell category of the multiple cell mass；
Judge the position coordinates of cell in each cell mass whether be in the cell mass cell category institute it is right
In the preset range answered；And
When judging result is that there are the cell kinds that the position coordinates of cell are not at the cell mass in the cell mass
In preset range corresponding to class, determine the cell mass for abnormal group；
Wherein, judge whether the position coordinates of cell in each cell mass are in the cell category of the cell mass
It is to be realized based on first nerves network model in corresponding preset range.
2. the method according to claim 1, wherein the institute's rheme for judging cell in each cell mass
It sets coordinate and whether is in preset range corresponding to the cell category of the cell mass and include:
By the position coordinates of cell in individual cells group and the input of the cell category of the individual cells group first mind
Through network model, the position that whether there is cell in the individual cells group is judged by the first nerves network model
Coordinate is not in preset range corresponding to the cell category of the cell mass.
3. the method according to claim 1, wherein further comprising:
Determine the intensity of anomaly of the abnormal group.
4. according to the method described in claim 3, it is characterized in that, the intensity of anomaly of the determination abnormal group includes:
It is corresponding with the abnormal cell category of group according to position coordinates of the center of gravity of the abnormal group in the coordinate system
The difference of position coordinates of the normal cell in the coordinate system determines the intensity of anomaly of the abnormal group.
5. the method according to claim 1, wherein the method also includes:
The position coordinates are compensated, are divided into the cell in the streaming sample according to the compensated position coordinates
Multiple cell masses.
6. according to the method described in claim 5, it is characterized in that, described compensate to the position coordinates includes:
By the corresponding coordinate vector of the position coordinates multiplied by the inverse matrix of compensation matrix, obtains the compensated position and sit
Mark, wherein the compensation matrix is to describe the cell influence journey of various colors to each Color Channel in dyeing course
Degree.
7. according to the method described in claim 6, it is characterized in that, before being compensated to the position coordinates, further includes:
Judge whether the compensation matrix is accurate；
If judging result is the compensation matrix inaccuracy, the compensation matrix is corrected.
8. the method according to the description of claim 7 is characterized in that it is described judge the compensation matrix whether accurately include:
Calculate the position coordinates of the cell point composition all equal with each coordinate components in the coordinate system straight line it
Between distance value；
Calculate quantity accounting of the distance value described in the streaming sample less than the cell of pre-determined distance threshold value；And
If the quantity accounting is greater than the first default accounting threshold value, it is determined that the compensation matrix inaccuracy.
9. the method according to the description of claim 7 is characterized in that the amendment compensation matrix includes:
Calculate the position coordinates of the cell point composition all equal with each coordinate components in the coordinate system straight line it
Between distance value；
Calculate quantity accounting of the distance value less than the cell of pre-determined distance threshold value；And
The element value in the compensation matrix is adjusted, so that the compensated quantity accounting is less than the second default accounting threshold value.
10. a kind of fluidic cell intelligent immunity parting device characterized by comprising
Coordinate obtaining module, be configured to obtain streaming sample in each cell using cell surface difference antigen molecular as coordinate
Position coordinates in the coordinate system of axis；
Grouping module is configured to that the cell in the streaming sample is divided into multiple cell masses according to the position coordinates；
Category identification module is configured to identify the respective cell category of the multiple cell mass；
Abnormal group's judgment module is configured to judge whether the position coordinates of cell in each cell mass are in this described
In preset range corresponding to the cell category of cell mass；And
Abnormal group's determining module is configured to when judging result be that there are the position coordinates of cell to be not in the cell mass
In preset range corresponding to the cell category of the cell mass, determine the cell mass for abnormal group；
Wherein, judge whether the position coordinates of cell in each cell mass are in the cell category of the cell mass
It is to be realized based on first nerves network model in corresponding preset range.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811459148.6A CN109580458B (en) | 2018-11-30 | 2018-11-30 | Flow-type cell intelligent immune typing method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811459148.6A CN109580458B (en) | 2018-11-30 | 2018-11-30 | Flow-type cell intelligent immune typing method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109580458A true CN109580458A (en) | 2019-04-05 |
| CN109580458B CN109580458B (en) | 2020-02-14 |
Family
ID=65925964
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811459148.6A Active CN109580458B (en) | 2018-11-30 | 2018-11-30 | Flow-type cell intelligent immune typing method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109580458B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113188982A (en) * | 2021-04-30 | 2021-07-30 | 天津深析智能科技发展有限公司 | Method for effectively removing interference of mononuclear cells in lymphocyte subpopulation automatic analysis |
| CN113218848A (en) * | 2021-04-30 | 2021-08-06 | 天津深析智能科技发展有限公司 | Method for judging non-specific cell population |
| CN114018789A (en) * | 2021-10-08 | 2022-02-08 | 武汉大学 | Acute leukemia typing method based on imaging flow cytometry detection and machine learning |
| CN114418995A (en) * | 2022-01-19 | 2022-04-29 | 生态环境部长江流域生态环境监督管理局生态环境监测与科学研究中心 | Cascade algae cell statistical method based on microscope image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0658345A1 (en) * | 1989-05-12 | 1995-06-21 | Cytogam, Inc. | Sex-associated membrane proteins and methods for increasing the probability that offspring will be of a desired sex |
| WO2005108963A1 (en) * | 2004-05-06 | 2005-11-17 | Nanyang Technological University | Microfluidic cell sorter system |
| CN106290877A (en) * | 2016-09-08 | 2017-01-04 | 北京海思特临床检验所有限公司 | Acute myeloid leukemia G-Ag expression detection kit and detection method |
| CN106644901A (en) * | 2015-10-14 | 2017-05-10 | 北京信息科技大学 | Fluorescence compensation method for flow cytometer |
| CN108693099A (en) * | 2018-03-16 | 2018-10-23 | 中国科学院苏州生物医学工程技术研究所 | Full-automatic fluorescence compensation method for flow cytometer |
-
2018
- 2018-11-30 CN CN201811459148.6A patent/CN109580458B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0658345A1 (en) * | 1989-05-12 | 1995-06-21 | Cytogam, Inc. | Sex-associated membrane proteins and methods for increasing the probability that offspring will be of a desired sex |
| WO2005108963A1 (en) * | 2004-05-06 | 2005-11-17 | Nanyang Technological University | Microfluidic cell sorter system |
| CN106644901A (en) * | 2015-10-14 | 2017-05-10 | 北京信息科技大学 | Fluorescence compensation method for flow cytometer |
| CN106290877A (en) * | 2016-09-08 | 2017-01-04 | 北京海思特临床检验所有限公司 | Acute myeloid leukemia G-Ag expression detection kit and detection method |
| CN108693099A (en) * | 2018-03-16 | 2018-10-23 | 中国科学院苏州生物医学工程技术研究所 | Full-automatic fluorescence compensation method for flow cytometer |
Non-Patent Citations (3)
| Title |
|---|
| OLIVER THEWS 等: ""Computer-Assisted Interpretation of Flow Cytometry Data in Hematology"", 《CYTOMETRY》 * |
| RAVI KOTHARI 等: ""Neural Network Analysis of Flow Cytometry"", 《IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING》 * |
| 王书奎，周振英 主编: "《实用流式细胞术彩色图谱》", 30 April 2004, 第二军医大学出版社 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113188982A (en) * | 2021-04-30 | 2021-07-30 | 天津深析智能科技发展有限公司 | Method for effectively removing interference of mononuclear cells in lymphocyte subpopulation automatic analysis |
| CN113218848A (en) * | 2021-04-30 | 2021-08-06 | 天津深析智能科技发展有限公司 | Method for judging non-specific cell population |
| CN113188982B (en) * | 2021-04-30 | 2022-05-10 | 天津深析智能科技发展有限公司 | Method for effectively removing interference of mononuclear cells in lymphocyte subpopulation automatic analysis |
| CN113218848B (en) * | 2021-04-30 | 2022-08-12 | 天津深析智能科技发展有限公司 | Method for judging non-specific cell population |
| CN114018789A (en) * | 2021-10-08 | 2022-02-08 | 武汉大学 | Acute leukemia typing method based on imaging flow cytometry detection and machine learning |
| CN114418995A (en) * | 2022-01-19 | 2022-04-29 | 生态环境部长江流域生态环境监督管理局生态环境监测与科学研究中心 | Cascade algae cell statistical method based on microscope image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109580458B (en) | 2020-02-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2020213396B2 (en) | Systems and methods for panel design in flow cytometry | |
| CN109580458A (en) | Fluidic cell intelligent immunity classifying method, device and electronic equipment | |
| US9880155B2 (en) | System, method, and article for detecting abnormal cells using multi-dimensional analysis | |
| JP2018505392A (en) | Automated flow cytometry analysis method and system | |
| US8391575B2 (en) | Automatic image analysis and quantification for fluorescence in situ hybridization | |
| US11639936B2 (en) | System, method, and article for detecting abnormal cells using multi-dimensional analysis | |
| JPH0352573B2 (en) | ||
| CN103592262B (en) | cell analysis apparatus and cell analysis method | |
| CN109636782A (en) | A kind of cell type analysis model training method, device and analysis method | |
| Gill et al. | Stratification of White Blood Cells Using Optimized DenseNet201 Model | |
| CN108520249A (en) | A kind of construction method of cell sorter, apparatus and system | |
| CN109554432A (en) | A kind of cell type analysis method, analytical equipment and electronic equipment | |
| CN105986003A (en) | White blood cell count method, white blood cell count device and cell analyzer | |
| CN109507406A (en) | A kind of cellular morphology intelligent extract method, device and analysis method | |
| US20210343009A1 (en) | Distance-based tissue state determination | |
| CN109490176B (en) | FCM analysis data processing method, device and classifying method, device | |
| CN104297135B (en) | Recognizing method and recognizing system of particles in blood sample and blood cell analytic instrument | |
| EP3775897B1 (en) | Biomarker analysis for high-throughput diagnostic multiplex data | |
| JP5314145B2 (en) | Method and apparatus for classification, visualization and search of biological data | |
| CN107430587A (en) | Automate flow cytometry method and system | |
| US10274412B2 (en) | Flow cytometry data segmentation result evaluation systems and methods | |
| JP6514818B1 (en) | System, program and method | |
| Takenga et al. | A clinical tool for automated flow cytometry based on machine learning methods | |
| Monteiro et al. | An efficient bioinformatics algorithm for healthcare: Detection and counting of leukocytes, blasts, and erythrocytes | |
| Novák | Studying lymphocyte development using mass cytometry |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address |
Address after: No. 5018f, Zhongnan business building, 21-b, innovation and entrepreneurship Park, 4668 Xinbei Road, Tanggu Ocean Science Park, Binhai New Area, Tianjin, 300451 Patentee after: Tianjin Shenxi Intelligent Technology Development Co.,Ltd. Address before: 215021 b6-2f, r2011, bio nano Park, 218 Xinghu Road, Suzhou Industrial Park, Suzhou City, Jiangsu Province Patentee before: SUZHOU SHENXI INTELLIGENT TECHNOLOGY Co.,Ltd. |
|
| CP03 | Change of name, title or address | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Flow cytometry intelligent immune typing method, device, and electronic device Granted publication date: 20200214 Pledgee: Bank of Shanghai Limited by Share Ltd. Tianjin branch Pledgor: Tianjin Shenxi Intelligent Technology Development Co.,Ltd. Registration number: Y2024120000006 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |