CN108362628A - The n cell flow-sorting methods of flow cytometer are imaged based on polarizing diffraction - Google Patents
The n cell flow-sorting methods of flow cytometer are imaged based on polarizing diffraction Download PDFInfo
- Publication number
- CN108362628A CN108362628A CN201810028502.3A CN201810028502A CN108362628A CN 108362628 A CN108362628 A CN 108362628A CN 201810028502 A CN201810028502 A CN 201810028502A CN 108362628 A CN108362628 A CN 108362628A
- Authority
- CN
- China
- Prior art keywords
- cell
- convolutional neural
- neural networks
- diffraction
- flow
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 45
- 230000010287 polarization Effects 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 10
- 230000001427 coherent effect Effects 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 6
- 238000005286 illumination Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 238000005388 cross polarization Methods 0.000 claims description 3
- 239000012530 fluid Substances 0.000 description 7
- 239000000919 ceramic Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 2
- 239000003550 marker Substances 0.000 description 2
- 206010052128 Glare Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000002771 cell marker Substances 0.000 description 1
- 239000011093 chipboard Substances 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000007850 fluorescent dye Substances 0.000 description 1
- 230000004313 glare Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000003475 lamination Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 238000011282 treatment Methods 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—Optical investigation techniques, e.g. flow cytometry
- G01N15/1425—Optical investigation techniques, e.g. flow cytometry using an analyser being characterised by its control 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
- G01N15/10—Investigating individual particles
- G01N15/14—Optical investigation techniques, e.g. flow cytometry
- G01N15/1429—Signal processing
-
- 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
- G01N15/01—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials specially adapted for biological cells, e.g. blood cells
-
- 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—Optical investigation techniques, e.g. flow cytometry
- G01N15/149—Optical investigation techniques, e.g. flow cytometry specially adapted for sorting particles, e.g. by their size or optical properties
-
- 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
Landscapes
- Chemical & Material Sciences (AREA)
- Dispersion Chemistry (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
A kind of n cell flow-sorting methods being imaged flow cytometer based on polarizing diffraction, including:Establish convolutional neural networks disaggregated model;The convolutional neural networks disaggregated model established is trained and is examined according to cell type to be selected;FPGA programming softwares are called to generate the FPGA code of convolutional neural networks disaggregated model;The fpga chip in realtime graphic recognition unit is written into the FPGA code;Sort cell.The n cell flow-sorting methods that flow cytometer is imaged based on polarizing diffraction of the present invention, it being capable of the unmarked cell diffraction image of quick obtaining, identification diffraction image corresponds to cell type in real time, and sorts cell in due course by micro fluidic device, reaches the purpose of n cell categorised collection.
Description
Technical field
The present invention relates to a kind of biological cell method for separating.Fluidic cell is imaged based on polarizing diffraction more particularly to one kind
The n cell flow-sorting methods of instrument.
Background technology
Being sorted in scientific research, clinical diagnosis and treatment for biological cell has important practical significance, fluidic cell
Sorter is cell sorting method important branch, and traditional flow cell sorter is conventional flow cytoanalyze technical foundation
On, increase sorting unit, according to the difference of cell marker, the instrument that cell classification is collected.Due to conventional cell streaming point
It selects the fluorescence types that instrument generally use cell is marked as tag along sort, is consequently belonging to markd cell sorting method.
It is to obtain single cell queue using streaming method that polarizing diffraction, which is imaged flow cytometer, under coherent light illumination,
The dual-polarization diffraction image of cell in queue is shot, and to instrument that diffraction image is analyzed.The analytical instrument is analyzed thin
Born of the same parents are consequently belonging to the analysis method of unmarked cell without carrying out the operations such as fluorescent marker.
Diffraction image is more abstract compared to common plane image, therefore the technique study of generally use machine learning its point
Class problem, convolutional neural networks (CNN) are exactly one of common methods, have obtained answering well in the classification of cell diffraction image
With.
Invention content
The technical problem to be solved by the invention is to provide it is a kind of reach n cell categorised collection purpose based on inclined
Shake the n cell flow-sorting methods of diffraction imaging flow cytometer.
The technical solution adopted in the present invention is:A kind of n cell stream being imaged flow cytometer based on polarizing diffraction
Formula method for separating, includes the following steps:
1) convolutional neural networks disaggregated model is established;
2) the convolutional neural networks disaggregated model established is trained and is examined according to cell type to be selected;
3) FPGA programming softwares is called to generate the FPGA code of convolutional neural networks disaggregated model;
4) FPGA code is written to the fpga chip in realtime graphic recognition unit;
5) cell is sorted.
Described in step 1) includes based on convolutional neural networks disaggregated model:1 input layer, 2 or more convolutional layers,
2 corresponding with convolutional layer or more pond layers, 1 full articulamentum and 1 output layer, the input layer are cell diffraction
Figure, the output layer have 3 output nodes, are named as A sortings channel, the sorting channels B and unsuccessful channel.
Step 2) includes that foundation waits for that the cell type of sorted sample picks out corresponding cell diffraction from diffraction pattern data library
Diffraction pattern and corresponding cell type label are divided into two groups by figure, and one group is training group, and another group is check groups, passes through instruction
Practice group data to be trained convolutional neural networks disaggregated model, obtains the network parameter of convolutional neural networks;Then, it will examine
The diffraction pattern of each cell inputs convolutional neural networks disaggregated model respectively in group, then convolutional neural networks disaggregated model tells institute
The cell of input belongs to the sorting channels A or the sorting channels B or unsuccessful channel, the standard of statistics convolutional neural networks disaggregated model classification
Exactness when accuracy reaches setting value, indicates that the convolutional neural networks disaggregated model training is completed, and otherwise, increases cell and spreads out
Penetrate figure quantity, re -training.
The diffraction pattern data library storage is the cell diffraction pattern containing cell type label.
The accuracy of the statistics convolutional neural networks disaggregated model classification, is with correct cell number and the cell of classifying
The ratio of sum.
Realtime graphic recognition unit described in step 4) includes fpga chip, and the fpga chip input is S-polarization
Diffraction pattern and P polarization diffraction pattern, the output end of the fpga chip connect the sorting driver for controlling cell flow direction,
The fpga chip is also connected with memory, and connects computer by communication interface.
The S-polarization diffraction pattern and P polarization diffraction pattern, when being that each cell flows through camera site, in the horizontal direction
Coherent light illumination under, the diffraction light of cell be diffracted as the polarization splitting prism in microscopic photography unit be divided into two it is orthogonal
Polarised light, to clap two width cross-polarizations diffraction image, be denoted as S-polarization diffraction pattern and P polarization diffraction pattern respectively, this two width
It is sent into realtime graphic recognition unit after the completion of figure shooting.
Step 5) includes:The channel belonging to cell selected according to realtime graphic recognition unit, driving sorting mechanism make carefully
Born of the same parents flow into corresponding channel.
The n cell flow-sorting methods that flow cytometer is imaged based on polarizing diffraction of the present invention, can quickly be obtained
Unmarked cell diffraction image is taken, identifies that diffraction image corresponds to cell type in real time, and sorted in due course carefully by micro fluidic device
Born of the same parents reach the purpose of n cell categorised collection.The present invention has the following advantages:
1, special marking need not be carried out to cell, such as fluorescent staining etc. is beneficial to cell repetition to cell fanout free region
It utilizes.
2, use cost is cheap, does not need special chemicals consumptive material, environmental-friendly.
3, wide adaptability, to the type of cell without particular/special requirement, as long as diffraction image can be acquired.
4, screening accuracy is high, and the three-dimensional structure information of diffraction image reacting cells, information content is more rich, to available
Higher accuracy rate.
Description of the drawings
Fig. 1 is the structural schematic diagram of polarizing diffraction imaging flow cytometer;
Fig. 2 is the structural schematic diagram of micro-fluidic chip plate;
Fig. 3 a are P polarization diffraction patterns;
Fig. 3 b are S-polarization diffraction patterns;
Fig. 4 is the structure diagram of realtime graphic recognition unit;
Fig. 5 is convolutional neural networks disaggregated model.
In figure
101:Diffraction image microscopic photography unit 102:P cameras
103:Realtime graphic recognition unit 104:Computer
105:S cameras 106:S mirrors
107:Sort driver 108:A sorts test tube
109:Unsuccessful test tube 110:B sorts test tube
111:Coherent light 113:Micro-fluidic chip plate
114:Sample bottle 115:Sheath fluid bottle
116:Microcobjective 117:Spike filter
118:Polarization splitting prism 119:P mirrors
201:Sample flow channel 202:Sheath fluid channel
203:Conical opening 204:Sense channel
205:Camera site 206:The single queue of cell to be sorted
207:Ceramic chips 208:Sort location
209:A sorts channel 210:Unsuccessful channel
211:B sorts channel 1:Fpga chip
2:Memory 3:Communication interface
4:S diffraction patterns 5:P diffraction patterns
Specific implementation mode
With reference to embodiment and attached drawing to the n cell for being imaged flow cytometer based on polarizing diffraction of the present invention
Flow-sorting methods are described in detail.
Polarizing diffraction is imaged flow cytometer as shown in Figure 1, including:Diffraction image for obtaining cell diffraction image is aobvious
Micro- shooting unit 101 is arranged in 101 lower end of diffraction image microscopic photography unit for guiding cell to exist at a given speed
Position appropriate is flowed, to facilitate, optical system is shot and the micro-fluidic chip plate 113 of sorting unit running, the diffraction image are aobvious
The output end of two Image Acquisition of micro- shooting unit 101 is by signal wire connection for cell category corresponding to diffraction image
The realtime graphic recognition unit 103 identified in real time, the realtime graphic recognition unit 103 be also respectively connected with computer 104 and
Signal input part for the sorting driver 107 for driving sorting actuating mechanism in micro-fluidic chip plate 113.The computer
104 include mating control software system, are given birth to for recognizer in the library management of cell diffraction pattern data, realtime graphic recognition unit
At and human-computer interaction, wherein cell diffraction pattern data library storage is cell diffraction pattern and its corresponding cell type label.Institute
The signal output end for stating sorting driver 107 is separately connected the pressure for driving cell flow direction of micro-fluidic chip plate 113
Electroceramics chip.The cell outflow end of the micro-fluidic chip plate 113 is separately connected A sortings test tube 108, unsuccessful test tube 109 and B
Sort test tube 110.The input terminal of the micro-fluidic chip plate 113 is separately connected sample bottle 114 and sheath fluid bottle 115.The miniflow
The middle part of control chip board 113 is provided with the coherent light 111 for shooting the illumination of visual field inner cell.
The diffraction image microscopic photography unit 101 include the microcobjective 116 set gradually from the bottom to top along light path,
Spike filter 117, polarization splitting prism 118, P mirrors 119 and P cameras 102, are successively set on the polarization splitting prism
S cylinders mirror 106 on the 118 another light separated and S cameras 105.Wherein, the output of the P cameras 102 and S cameras 105
End is separately connected the realtime graphic recognition unit 103.
As shown in Fig. 2, the micro-fluidic chip plate 113 include integrally formed conical opening 203, sense channel 204,
A sorts channel 209, the sorting channels 211 B and unsuccessful channel 210, wherein the sense channel 204 and the A sortings are logical
Road 209, the sorting channels 211 B and 210 junction of unsuccessful channel are provided with sort location 208, and the corresponding sort location 208 is set
The ceramic chips 207 for driving cell flow direction are equipped with, sample flow channel is formed in the conical opening 203
201 and sheath fluid channel 202, need to be sorted the single queue 206 of cell in the sense channel 204, and in sense channel 204
Place setting camera site 205.
The present invention's is imaged the n cell flow-sorting methods of flow cytometer based on polarizing diffraction, including walks as follows
Suddenly:
1) convolutional neural networks disaggregated model is established;
As shown in figure 5, described include based on convolutional neural networks disaggregated model:1 input layer, 2 or more volumes
Lamination, 2 corresponding with convolutional layer or more pond layers, 1 full articulamentum and 1 output layer, the input layer are cell
Diffraction pattern, the output layer have 3 output nodes, are named as A sortings channel, the sorting channels B and unsuccessful channel.
2) the convolutional neural networks disaggregated model established is trained and is examined according to cell type to be selected;Including:
According to waiting for that the cell type of sorted sample picks out corresponding cell diffraction pattern from diffraction pattern data library, described spreads out
Penetrate chart database storage is the cell diffraction pattern containing cell type label.The diffraction pattern and corresponding cell that will be singled out
Type label is divided into two groups, and one group is training group, and another group is check groups, is classified according to convolutional neural networks by set
Model is trained, and obtains the network parameter of convolutional neural networks;Then, the diffraction pattern of each cell in check groups is inputted respectively
Convolutional neural networks disaggregated model, then convolutional neural networks disaggregated model tell inputted cell and belong to A sortings channel or B
Sort channel or unsuccessful channel, the accuracy of statistics convolutional neural networks disaggregated model classification, the statistics convolutional Neural net
The accuracy of network disaggregated model classification is the ratio with classify correct cell number and total number of cells.Accuracy reaches setting value
When, it indicates that the convolutional neural networks disaggregated model training is completed, otherwise, increases cell diffraction pattern quantity, re -training.
Such as:Containing C1, C2, C3 in cell sample to be sorted, tetra- kinds of cells of C4 are present in diffraction pattern data library,
User's Pre-sorting goes out C1 and C2, then may specify that C1 enters A sortings channel 209, C2 enters B sortings channel 211, and C3 and C4, which enter, to be fallen
Gate road 210.Each 1000 of the C1~C4 cells that will be singled out, each cell contains a P polarization diffraction pattern and a S-polarization
Diffraction pattern shares 4000 pairs of diffraction patterns, is divided into that training group 3600 is right, and check groups 400 are right, soft using the TensorFlow of Google
Part is trained convolutional neural networks (CNN) grader, obtains CNN network parameters.The input requirements of general CNN graders
It is single image, therefore the combination picture together with the P polarization diffraction pattern of each cell is vertically connected with S-polarization diffraction pattern
As input, output is the sorting channel that C1~C4 is divided into.It is tested to trained CNN graders with check groups data,
If there is 90 C1 cells to be divided into A sortings channel in classification results, 92 C2 cells are divided into B sortings channel, 92 C3 cells point
Enter unsuccessful channel, 94 C4 are divided into unsuccessful channel, then accuracy is (90+92+92+94)/400=92%.If point of setting
Class accuracy rate is 90%, then the sorter model trained is available.
3) FPGA programming softwares is called to generate the FPGA code of convolutional neural networks disaggregated model;
Fpga chip manufacturer generally provides programming software, such as the ISE that provides of Xilinx companies develops software, and user can be with
According to the CNN network parameters generated in step 2), program is realized accordingly with Verilog language editors, is generated after ISE is compiled
FPGA code.
4) FPGA code is written to the fpga chip in realtime graphic recognition unit;
As shown in figure 4, the realtime graphic recognition unit includes fpga chip 1, the input of the fpga chip 1 is S
Polarizing diffraction Fig. 4 and P polarization diffraction pattern 5, the output end of the fpga chip 1 connect the sorting for controlling cell flow direction
Driver 107, the fpga chip 1 is also connected with memory 2, and connects computer 104 by communication interface 3.
The S-polarization diffraction pattern 4 and P polarization diffraction pattern 5, when being that each cell flows through camera site, in level side
To coherent light illumination under, the diffraction light of cell be diffracted as the polarization splitting prism in microscopic photography unit be divided into two it is orthogonal
Polarised light, to clap two width cross-polarizations diffraction image, be denoted as S-polarization diffraction pattern and P polarization diffraction pattern respectively, this two
It is sent into realtime graphic recognition unit after the completion of the shooting of width figure.
It is integrated with the programming software of fpga chip in the ISE softwares of Xilinx, fpga chip can be automatically identified
The fpga chip of realtime graphic recognition unit is written by downloader by model for the FPGA code generated in step 3).
5) cell is sorted.Including:
The channel belonging to cell selected according to realtime graphic recognition unit, driving sorting mechanism make corresponding to cell inflow
Channel.Specifically there is following process in conjunction with Fig. 1 and Fig. 2:
(1) sample introduction
Suspension containing C1~C4 cells is placed in sample bottle 114 by user, sheath fluid and sample bottle in sheath fluid bottle 115
Sample in 114 enters sheath fluid channel 202 and the sample flow channel of horizontal positioned micro-fluidic chip plate 113 under air pressure driving
201, after through conical opening 203 enter sense channel 204.The single queue 206 of cell to be sorted is formed to advance along the horizontal directions y.
(2) diffraction pattern is shot
When cell is by camera site 205, (the generally use 532nm single modes of coherent light 111 with the propagation of the horizontal directions z
Laser) it meets, resulting centered on x-axis, the scattering light in 116 aperture angular region of microcobjective is by microcobjective 116
Collect, filter out other wavelength veiling glares through spike filter 117 (532nm centre wavelengths, 10nm bandwidth), after through polarization spectro rib
Mirror 118 is divided to for two beam crossed polarized light of P and S, and P polarization light reaches P cameras 102 through P mirrors 119 and forms P polarization diffraction patterns, and S is inclined
The light that shakes reaches S cameras 105 through S mirrors 106 and forms S-polarization diffraction pattern, as shown in Figure 3a and Figure 3b shows.
(3) identification cell diffraction pattern corresponds to sorting channel in real time
To be transported to realtime graphic recognition unit 103 after the completion of diffraction pattern shooting, in realtime graphic recognition unit 103 by
The convolutional neural networks grader that fpga chip is realized determines the corresponding sorting channel of cell according to the diffraction image.
(4) driving sorting mechanism sorts cell
When the sorting channel that cell advances to sort location 208, and the fpga chip in realtime graphic recognition unit 103 is sent
Id signal reaches sorting driver 107, and sorting driver 107 exports voltage driving appropriate according to sorting gap marker signal
The sorting mechanism action being made of ceramic chips 207, cell are displaced to the sorting channels 209 A or B under the impact of fine motion piece
Channel 211 is sorted, when ceramic chips attonity, cell enters unsuccessful channel 210.Above procedure is achieved that sorting is special
Determine the purpose of cell types.
Claims (8)
1. a kind of n cell flow-sorting methods being imaged flow cytometer based on polarizing diffraction, which is characterized in that including
Following steps:
1) convolutional neural networks disaggregated model is established;
2) the convolutional neural networks disaggregated model established is trained and is examined according to cell type to be selected;
3) FPGA programming softwares is called to generate the FPGA code of convolutional neural networks disaggregated model;
4) FPGA code is written to the fpga chip in realtime graphic recognition unit;
5) cell is sorted.
2. the n cell flow-sorting methods according to claim 1 that flow cytometer is imaged based on polarizing diffraction,
It is characterized in that, including based on convolutional neural networks disaggregated model described in step 1):1 input layer, 2 or more convolution
Layer, 2 corresponding with convolutional layer or more pond layers, 1 full articulamentum and 1 output layer, the input layer spread out for cell
Figure is penetrated, the output layer there are 3 output nodes, is named as A sortings channel, the sorting channels B and unsuccessful channel.
3. the n cell flow-sorting methods according to claim 1 that flow cytometer is imaged based on polarizing diffraction,
It is characterized in that, step 2), which includes foundation, waits for that the cell type of sorted sample picks out corresponding cell from diffraction pattern data library and spreads out
Figure is penetrated, diffraction pattern and corresponding cell type label are divided into two groups, one group is training group, and another group is check groups, is passed through
Set obtains the network parameter of convolutional neural networks according to being trained to convolutional neural networks disaggregated model;Then, it will examine
The diffraction pattern for testing each cell in group inputs convolutional neural networks disaggregated model respectively, then convolutional neural networks disaggregated model is told
The cell inputted belongs to the sorting channels A or the sorting channels B or unsuccessful channel, statistics convolutional neural networks disaggregated model classification
Accuracy when accuracy reaches setting value, indicates that the convolutional neural networks disaggregated model training is completed, otherwise, increases cell
Diffraction pattern quantity, re -training.
4. the n cell flow-sorting methods according to claim 3 that flow cytometer is imaged based on polarizing diffraction,
It is characterized in that, the diffraction pattern data library storage is the cell diffraction pattern containing cell type label.
5. the n cell flow-sorting methods according to claim 3 that flow cytometer is imaged based on polarizing diffraction,
It is characterized in that, the accuracy of the described statistics convolutional neural networks disaggregated model classification, be with classify correct cell number with
The ratio of total number of cells.
6. the n cell flow-sorting methods according to claim 1 that flow cytometer is imaged based on polarizing diffraction,
It is characterized in that, the realtime graphic recognition unit described in step 4) includes fpga chip (1), the fpga chip (1) is defeated
It is S-polarization diffraction pattern (4) and P polarization diffraction pattern (5) to enter, and the output end of the fpga chip (1) is connected for controlling cell stream
The sorting driver (107) in dynamic direction, the fpga chip (1) are also connected with memory (2), and even by communication interface (3)
Connect computer (104).
7. the n cell flow-sorting methods according to claim 6 that flow cytometer is imaged based on polarizing diffraction,
It is characterized in that, the S-polarization diffraction pattern (4) and P polarization diffraction pattern (5), when being that each cell flows through camera site,
Under the coherent light illumination of horizontal direction, the diffraction light of cell is diffracted as the polarization splitting prism in microscopic photography unit is divided into two
A orthogonal polarised light, to clap two width cross-polarizations diffraction image, be denoted as S-polarization diffraction pattern and P polarization diffraction respectively
Figure is sent into realtime graphic recognition unit after the completion of the shooting of this two width figure.
8. the n cell flow-sorting methods according to claim 1 that flow cytometer is imaged based on polarizing diffraction,
It is characterized in that, step 5) includes:The channel belonging to cell selected according to realtime graphic recognition unit, driving sorting mechanism make
Cell flows into corresponding channel.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810028502.3A CN108362628A (en) | 2018-01-11 | 2018-01-11 | The n cell flow-sorting methods of flow cytometer are imaged based on polarizing diffraction |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810028502.3A CN108362628A (en) | 2018-01-11 | 2018-01-11 | The n cell flow-sorting methods of flow cytometer are imaged based on polarizing diffraction |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108362628A true CN108362628A (en) | 2018-08-03 |
Family
ID=63011076
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810028502.3A Pending CN108362628A (en) | 2018-01-11 | 2018-01-11 | The n cell flow-sorting methods of flow cytometer are imaged based on polarizing diffraction |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108362628A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110060729A (en) * | 2019-03-28 | 2019-07-26 | 广州序科码生物技术有限责任公司 | A method of cell identity is annotated based on unicellular transcript profile cluster result |
CN110220836A (en) * | 2019-06-26 | 2019-09-10 | 袁家彬 | A kind of vitro detection stream type cell analyzer |
GB2576191A (en) * | 2018-08-08 | 2020-02-12 | Sphere Fluidics Ltd | Droplet processing methods and systems |
EP3828761A1 (en) * | 2019-11-29 | 2021-06-02 | Sysmex Corporation | Cell analysis method, cell analysis device, and cell analysis system |
CN113607628A (en) * | 2021-09-02 | 2021-11-05 | 清华大学 | Method for processing cell image stream by nerve morphology calculation driving image flow cytometer |
CN113740235A (en) * | 2021-09-02 | 2021-12-03 | 清华大学 | Image flow cytometer driven by neuromorphic calculation |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104136907A (en) * | 2011-12-07 | 2014-11-05 | Imec公司 | Analysis and sorting of objects in flow |
CA2944829A1 (en) * | 2014-05-23 | 2015-11-26 | Ting Chen | Systems and methods for detection of biological structures and/or patterns in images |
CN105589938A (en) * | 2015-12-13 | 2016-05-18 | 公安部第三研究所 | Image retrieval system and retrieval method based on FPGA |
CN106228240A (en) * | 2016-07-30 | 2016-12-14 | 复旦大学 | Degree of depth convolutional neural networks implementation method based on FPGA |
WO2017053592A1 (en) * | 2015-09-23 | 2017-03-30 | The Regents Of The University Of California | Deep learning in label-free cell classification and machine vision extraction of particles |
WO2017214572A1 (en) * | 2016-06-10 | 2017-12-14 | The Regents Of The University Of California | Image-based cell sorting systems and methods |
-
2018
- 2018-01-11 CN CN201810028502.3A patent/CN108362628A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104136907A (en) * | 2011-12-07 | 2014-11-05 | Imec公司 | Analysis and sorting of objects in flow |
CA2944829A1 (en) * | 2014-05-23 | 2015-11-26 | Ting Chen | Systems and methods for detection of biological structures and/or patterns in images |
WO2017053592A1 (en) * | 2015-09-23 | 2017-03-30 | The Regents Of The University Of California | Deep learning in label-free cell classification and machine vision extraction of particles |
CN105589938A (en) * | 2015-12-13 | 2016-05-18 | 公安部第三研究所 | Image retrieval system and retrieval method based on FPGA |
WO2017214572A1 (en) * | 2016-06-10 | 2017-12-14 | The Regents Of The University Of California | Image-based cell sorting systems and methods |
CN106228240A (en) * | 2016-07-30 | 2016-12-14 | 复旦大学 | Degree of depth convolutional neural networks implementation method based on FPGA |
Non-Patent Citations (5)
Title |
---|
KE DONG 等: "Label-free classification of cultured cells through diffraction imaging", 《BIOMEDICAL OPTICS EXPRESS》 * |
PHILIPP EULENBERG 等: "Deep learning for imaging flow cytometry: Cell cycle analysus of Jurkat cells", 《BIORXIV》 * |
SA YU 等: "A novel method of diffraction imaging flow cytometry for sizing microspheres cytometry for sizing microspheres", 《OPTICS EXPRESS》 * |
YUANMING FENG 等: "Polarization Imaging and Classification of Jurkat T and Ramos B Cells Using a Flow Cytometer", 《CYTOMETRY PART A》 * |
陶源 等: "基于卷积神经网络的细胞识别", 《中国医学物理杂志》 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2576191A (en) * | 2018-08-08 | 2020-02-12 | Sphere Fluidics Ltd | Droplet processing methods and systems |
GB2576191B (en) * | 2018-08-08 | 2022-11-16 | Sphere Fluidics Ltd | Droplet processing methods and systems |
CN110060729A (en) * | 2019-03-28 | 2019-07-26 | 广州序科码生物技术有限责任公司 | A method of cell identity is annotated based on unicellular transcript profile cluster result |
CN110060729B (en) * | 2019-03-28 | 2020-02-28 | 广州序科码生物技术有限责任公司 | Method for annotating cell identity based on single cell transcriptome clustering result |
CN110220836A (en) * | 2019-06-26 | 2019-09-10 | 袁家彬 | A kind of vitro detection stream type cell analyzer |
CN110220836B (en) * | 2019-06-26 | 2021-12-21 | 国科赛赋(深圳)新药研发科技有限公司 | Flow cytometry analyzer for in-vitro detection |
EP3828761A1 (en) * | 2019-11-29 | 2021-06-02 | Sysmex Corporation | Cell analysis method, cell analysis device, and cell analysis system |
CN112967785A (en) * | 2019-11-29 | 2021-06-15 | 希森美康株式会社 | Cell analysis method, device, system, and program |
US12020492B2 (en) | 2019-11-29 | 2024-06-25 | Sysmex Corporation | Cell analysis method, cell analysis device, and cell analysis system |
CN113607628A (en) * | 2021-09-02 | 2021-11-05 | 清华大学 | Method for processing cell image stream by nerve morphology calculation driving image flow cytometer |
CN113740235A (en) * | 2021-09-02 | 2021-12-03 | 清华大学 | Image flow cytometer driven by neuromorphic calculation |
CN113607628B (en) * | 2021-09-02 | 2023-02-10 | 清华大学 | Method for processing cell image stream by nerve morphology calculation driving image flow cytometer |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108362628A (en) | The n cell flow-sorting methods of flow cytometer are imaged based on polarizing diffraction | |
US10801944B2 (en) | High accuracy 5-part differential with digital holographic microscopy and untouched leukocytes from peripheral blood | |
Jayakody et al. | Microscope image based fully automated stomata detection and pore measurement method for grapevines | |
EP2936116B1 (en) | System and method for classification of particles in a fluid sample | |
US20060177348A1 (en) | Cell sorter chip having gel electrodes | |
CA2969912A1 (en) | Automated flow cytometry analysis method and system | |
JP2021522491A (en) | Characterization and sorting for particle analyzers | |
CN1031422A (en) | Utilize many parts discriminatory analysis device of light scattering technique | |
KR102122068B1 (en) | Image analyzing system and method thereof | |
JP2022512633A (en) | Adaptive sorting for particle analyzers | |
US11143587B2 (en) | Compensation editor | |
CN112161913A (en) | Analysis method and equipment for flow type fluorescence analysis system | |
US20220383629A1 (en) | Label-free cell classification and screening system based on hybrid transfer learning | |
US11788946B2 (en) | Optimized sorting gates | |
CN111652095A (en) | CTC image identification method and system based on artificial intelligence | |
CN114332856A (en) | Feature extraction method and device for blood type card image and blood type interpretation system | |
CN113588522A (en) | Circulating tumor detection and sorting method and system based on micro-fluidic and image recognition | |
EP4300437A2 (en) | System and method for classifying microscopic particles | |
EP3230887A1 (en) | Automated flow cytometry analysis method and system | |
CN112924452A (en) | Blood examination auxiliary system | |
CN114778419B (en) | High-magnification optical amplification imaging flow cytometer | |
US12008825B2 (en) | Classification workflow for flexible image based particle sorting | |
EP4396792A1 (en) | System and method for identifying and counting biological species | |
CN114441480A (en) | Nucleated red blood cell analysis device and analysis method | |
JP7557067B2 (en) | Classification workflows for flexible image-based particle sorting |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180803 |