WO2022089552A1 - 细胞杀伤效力和/或免疫活性的检测方法、系统及其应用 - Google Patents

细胞杀伤效力和/或免疫活性的检测方法、系统及其应用 Download PDF

Info

Publication number
WO2022089552A1
WO2022089552A1 PCT/CN2021/127199 CN2021127199W WO2022089552A1 WO 2022089552 A1 WO2022089552 A1 WO 2022089552A1 CN 2021127199 W CN2021127199 W CN 2021127199W WO 2022089552 A1 WO2022089552 A1 WO 2022089552A1
Authority
WO
WIPO (PCT)
Prior art keywords
cell
target
image
cells
culture
Prior art date
Application number
PCT/CN2021/127199
Other languages
English (en)
French (fr)
Inventor
罗浦文
姜晶
陈凯
范伟亚
Original Assignee
上海睿钰生物科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Priority claimed from CN202011173457.4A external-priority patent/CN112301086A/zh
Priority claimed from CN202011173498.3A external-priority patent/CN112285082A/zh
Priority claimed from CN202011173508.3A external-priority patent/CN112285083B/zh
Priority claimed from CN202011173496.4A external-priority patent/CN112285081B/zh
Priority claimed from CN202011176203.8A external-priority patent/CN112304851B/zh
Priority claimed from CN202110127301.0A external-priority patent/CN112813133B/zh
Priority claimed from CN202110595121.5A external-priority patent/CN113237818A/zh
Application filed by 上海睿钰生物科技有限公司 filed Critical 上海睿钰生物科技有限公司
Priority to EP21885279.6A priority Critical patent/EP4212853A4/en
Publication of WO2022089552A1 publication Critical patent/WO2022089552A1/zh
Priority to US18/301,247 priority patent/US20230273188A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/5005—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells
    • G01N33/5008—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics
    • G01N33/5044—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics involving specific cell types
    • G01N33/5047—Cells of the immune system
    • 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
    • G01N15/10—Investigating individual particles
    • G01N15/14—Optical investigation techniques, e.g. flow cytometry
    • G01N15/1429—Signal processing
    • G01N15/1433—Signal processing using image recognition
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G01N21/645—Specially adapted constructive features of fluorimeters
    • G01N21/6456—Spatial resolved fluorescence measurements; Imaging
    • G01N21/6458—Fluorescence microscopy
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/5005—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells
    • G01N33/5008—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics
    • G01N33/5011—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics for testing antineoplastic activity
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/5005—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells
    • G01N33/5008—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics
    • G01N33/5014—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics for testing toxicity
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
    • G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
    • G01N33/5005—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells
    • G01N33/5008—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics
    • G01N33/502—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving human or animal cells for testing or evaluating the effect of chemical or biological compounds, e.g. drugs, cosmetics for testing non-proliferative effects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G06T7/0014—Biomedical image inspection using an image reference approach
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • 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
    • 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
    • G01N2015/1486—Counting the 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—Optical investigation techniques, e.g. flow cytometry
    • G01N2015/1493—Particle size
    • 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
    • G01N2015/1497—Particle shape
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G01N2021/6417—Spectrofluorimetric devices
    • G01N2021/6419—Excitation at two or more wavelengths
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G01N2021/6417—Spectrofluorimetric devices
    • G01N2021/6421—Measuring at two or more wavelengths
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G01N21/6428—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes"
    • G01N2021/6439—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes" with indicators, stains, dyes, tags, labels, marks
    • G01N2021/6441—Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes" with indicators, stains, dyes, tags, labels, marks with two or more labels
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2500/00—Screening for compounds of potential therapeutic value
    • G01N2500/10—Screening for compounds of potential therapeutic value involving cells
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10056—Microscopic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10064—Fluorescence image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20212—Image combination
    • G06T2207/20221—Image fusion; Image merging
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30024—Cell structures in vitro; Tissue sections in vitro

Definitions

  • the present specification relates to the field of image processing, in particular, to a method, system and application for the detection of cell killing efficacy and/or immune activity.
  • the detection of cell killing efficacy is of great significance for the quality control of immune cell therapy products. Due to the large individual differences in the starting point of immune cell therapy products, the low degree of scale of the preparation process, most of the preparations are living cell products, and the mechanism of action is not very clear, etc. Due to the characteristics of poor comparability of products, the quality control research of immune cell therapy products is relatively complicated, and the quality control of killing efficacy is one of the difficulties.
  • One of the embodiments of the present specification provides a method for detecting cell killing efficacy and/or immune activity, characterized by comprising: acquiring multiple microscopic images of a fixed area of a co-culture sample, wherein the co-culture sample is a target cell with A cell sample obtained by co-cultivation of effector cells, the fixed area contains a plurality of targets, the plurality of targets are cell groups comprising cells with a variety of different properties, and each of the targets has image-identifiable features,
  • the cellular properties of each of the objects are characterized by a collection of feature information showing the image-identifiable features of the object on a plurality of the microscopic images; imaging a plurality of the microscopic images Overlay synthesis analysis or image fusion analysis, obtain the cell properties of a plurality of the targets, and statistically correlate the cell parameters with the cell properties; based on the cell parameters, evaluate the killing efficacy and/or immunity of the effector cells active.
  • the detection system includes the following modules: a microscopic imaging module for acquiring multiple A microscopic image, wherein the co-culture sample is a cell sample obtained by co-cultivation of target cells and effector cells, the fixed area of the co-culture sample contains a plurality of targets, and the plurality of targets are composed of a plurality of targets.
  • a microscopic imaging module for acquiring multiple A microscopic image, wherein the co-culture sample is a cell sample obtained by co-cultivation of target cells and effector cells, the fixed area of the co-culture sample contains a plurality of targets, and the plurality of targets are composed of a plurality of targets.
  • Cell groups of cells with different attributes each of the objects has image-recognizable features, and the cell attributes of each of the objects can be identified in a plurality of the microscopes through the image-recognizable features of the object.
  • the set of feature information displayed on the image is characterized; the image analysis module is used to perform image overlay synthesis analysis or image fusion analysis based on a plurality of the microscopic images, obtain the cell properties of the plurality of targets, and count A cellular parameter associated with the properties of the cell; an evaluation module for evaluating the killing potency and/or immune activity of the effector cell based on the cellular parameter.
  • One of the embodiments of the present specification provides an apparatus for detecting cell killing efficacy and/or immune activity, comprising a processor for performing a method for detecting cell killing efficacy and/or immune activity.
  • One of the embodiments of the present specification provides a computer-readable storage medium, the storage medium stores computer instructions, and after the computer reads the computer instructions in the storage medium, the computer executes a method for detecting cell killing efficacy and/or immune activity.
  • One of the embodiments of the present specification provides the application of a detection method or detection system for cell killing potency and/or immune activity in detecting cell killing potency, effector cell immune activity, preparing immune products, quality control of immune products or evaluating individual immune function.
  • FIG. 1 is a schematic diagram of an application scenario of a detection system for cell killing efficacy and/or immune activity according to some embodiments of the present specification
  • FIG. 2 is an exemplary flow chart of a method for detecting cell killing efficacy and/or immune activity according to some embodiments of the present specification
  • FIG. 3 is an exemplary flowchart of image overlay synthesis analysis of multiple microscopic images according to some embodiments of the present specification
  • FIG. 4 is an exemplary flowchart of extracting the contour of an object in a microscopic image according to some embodiments of the present specification
  • FIG. 5 is an exemplary flowchart of acquiring a fused image according to some embodiments of the present specification
  • FIG. 6 is an exemplary flowchart of analyzing a fused image according to some embodiments of the present specification
  • FIG. 7 is an exemplary schematic diagram of an image recognition model according to some embodiments of the present specification.
  • FIG. 8 is a fluorescence microscope image collected by the FL1 channel in Example 1 of this specification.
  • Fig. 9 is the fluorescence microscope image collected by FL2 channel in Example 1 of this specification.
  • Fig. 10 is a fluorescence microscope image collected by FL3 channel in Example 1 of this specification.
  • FIG. 11 is a superimposed image of the fluorescence microscopic images collected by the FL1, FL2, and FL3 channels in Example 1 of the present specification.
  • system means for distinguishing different components, elements, parts, parts or assemblies at different levels.
  • device means for converting components, elements, parts, parts or assemblies to different levels.
  • FIG. 1 is a schematic diagram of an application scenario of the detection system for cell killing efficacy and/or immune activity according to some embodiments of the present specification.
  • the detection system 100 may include a server 110 , a network 120 , a storage device 130 and an image acquisition device 140 .
  • Server 110 may be used to manage resources and process data and/or information from at least one component of detection system 100 or an external data source (eg, a cloud data center). For example, image overlay composite analysis of multiple microscopic images (bright field microscopic image and at least one fluorescence microscopic image). In another example, image fusion analysis is performed on a plurality of microscopic images.
  • server 110 may acquire data (eg, one or more of a plurality of microscopic images) from storage device 140 or save data (eg, cellular properties of objects, cellular parameters) to storage device 140, and also Data (eg, brightfield microscopic images and/or at least one fluorescence microscopic image) may be read from other sources, such as image acquisition device 140 , via network 120 .
  • data eg, one or more of a plurality of microscopic images
  • data eg, cellular properties of objects, cellular parameters
  • Data eg, brightfield microscopic images and/or at least one fluorescence microscopic image
  • server 110 may be a single server or a group of servers.
  • the server group may be centralized or distributed (for example, server 110 may be a distributed system), dedicated or concurrently served by other devices or systems.
  • server 110 may be regional or remote.
  • server 110 may be implemented on a cloud platform, or provided in a virtual fashion.
  • the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc., or any combination thereof.
  • server 110 may include a processing device.
  • Processing devices may process data and/or information obtained from other devices or system components.
  • the processor may execute program instructions based on such data, information and/or processing results to perform one or more of the functions described herein.
  • a processing device may include one or more sub-processing devices (eg, a single-core processing device or a multi-core multi-core processing device).
  • a processing device may include a central processing unit (CPU), an application specific integrated circuit (ASIC), an application specific instruction processor (ASIP), a graphics processor (GPU), a physical processor (PPU), a digital signal processor (DSP) ), field programmable gate array (FPGA), programmable logic circuit (PLD), controller, microcontroller unit, reduced instruction set computer (RISC), microprocessor, etc. or any combination of the above.
  • CPU central processing unit
  • ASIC application specific integrated circuit
  • ASIP application specific instruction processor
  • GPU graphics processor
  • PPU physical processor
  • DSP digital signal processor
  • FPGA field programmable gate array
  • PLD programmable logic circuit
  • controller microcontroller unit, reduced instruction set computer (RISC), microprocessor, etc. or any combination of the above.
  • the network 120 may connect various components of the detection system 100 and/or connect the system to external resource components.
  • the network 120 enables communication between the various components and with other components outside the system, facilitating the exchange of data and/or information.
  • the network 120 may be any one or more of a wired network or a wireless network.
  • the network 120 may include a cable network, a fiber optic network, a telecommunications network, the Internet, a local area network (LAN), a wide area network (WAN), a wireless local area network (WLAN), a metropolitan area network (MAN), a public switched telephone network (PSTN) , Bluetooth network, ZigBee network (ZigBee), near field communication (NFC), intra-device bus, intra-device line, cable connection, etc. or any combination thereof.
  • the network connection between the various parts can be in one of the above-mentioned ways, and can also be in a variety of ways.
  • the network may be in point-to-point, shared, centralized, etc. various topologies or a combination of multiple topologies.
  • network 120 may include one or more network access points.
  • network 120 may include wired or wireless network access points, such as base stations and/or network exchange points, through which one or more components of system 100 may connect to network 120 to exchange data and/or information.
  • Storage device 130 may be used to store data (eg, brightfield microscopy images and at least one fluorescence microscopy image) and/or instructions. Storage device 130 is implemented in a single central server, multiple servers connected by communication links, or multiple personal devices. In some embodiments, storage device 130 may include mass memory, removable memory, volatile read-write memory (eg, random access memory RAM), read only memory (ROM), the like, or any combination thereof. Illustratively, mass storage may include magnetic disks, optical disks, solid state disks, and the like. In some embodiments, storage device 130 may be implemented on a cloud platform.
  • data eg, brightfield microscopy images and at least one fluorescence microscopy image
  • Storage device 130 is implemented in a single central server, multiple servers connected by communication links, or multiple personal devices.
  • storage device 130 may include mass memory, removable memory, volatile read-write memory (eg, random access memory RAM), read only memory (ROM), the like, or any combination thereof.
  • mass storage may include magnetic disks, optical disks, solid state
  • Image acquisition device 140 may be used to acquire multiple microscopic images (bright field microscopic images and at least one fluorescence microscopic image) of a fixed region of the co-culture sample.
  • the image acquisition device that acquires the different microscopic images (bright field microscopic image and at least one fluorescence microscopic image) may be the same.
  • the image acquisition device 140 may be a metallographic microscope.
  • the image acquisition devices that acquire different microscopic images may be different.
  • image acquisition device 140 may include a brightfield microscope for acquiring brightfield microscopy images and a fluorescence microscope for acquiring at least one fluorescence microscopy image.
  • the detection system 100 may also include a terminal device (not shown).
  • Terminal devices may include input devices (eg, keyboard, mouse) and/or output devices (eg, display screens, speakers).
  • the user can interact with the processing device 110, the image acquisition device 140 and other devices through the terminal device.
  • the user can view the microscopic image acquired by the image acquisition device 140 through the terminal device.
  • the user can directly observe the image analysis result processed by the processing device through the terminal device.
  • detection system 100 may include a microscopic imaging module, an image analysis module, and an evaluation module.
  • Microscopic imaging modules can be used to obtain microscopic images of cell samples.
  • the microscopic images may include brightfield microscopic images and at least one fluorescence microscopic image.
  • cell samples can include co-culture samples and control samples.
  • Co-culture samples are cell samples obtained by co-culture of target cells and effector cells.
  • Control samples are cell samples obtained by culturing target cells and/or effector cells alone.
  • step 210 For more description of the microscopic imaging module, reference may be made to step 210, which will not be repeated here.
  • the image analysis module can be used to perform image overlay synthesis analysis based on multiple microscopic images, to obtain the cellular properties of multiple targets, and to correlate cellular parameters of the cellular properties of multiple targets. Further, in some embodiments, the image analysis module can extract multiple target object regions and corresponding contour information in each microscopic image.
  • the target area is an image area containing a single target with a closed outline.
  • the image analysis module may determine the coincidence of the target objects based on the target object regions of the multiple microscopic images and the corresponding contour information, and obtain the coincidence determination result. In some embodiments, the image analysis module may obtain the cell properties of the corresponding target based on the coincidence determination result.
  • the image analysis module can classify, count and count a plurality of targets based on cell attributes to obtain cell parameters.
  • the image analysis module can also be used to perform image fusion analysis based on multiple microscopic images, to obtain cellular properties of multiple targets, and cell parameters associated with the cellular properties of multiple targets.
  • the image analysis module may obtain a fused image based on the plurality of microscopic images. Further, in some embodiments, the image analysis module may extract feature points of each of the microscopic images. In some embodiments, the image analysis module can register the plurality of microscopic images based on corresponding feature points of the plurality of microscopic images. In some embodiments, the image analysis module may fuse the registered multiple microscopic images based on transparency and/or chromaticity to obtain a fused image.
  • the image analysis module may analyze the fused image to obtain cellular properties of the plurality of objects, and cellular parameters associated with the cellular properties of the plurality of objects. Further, in some embodiments, the image analysis module may acquire multiple target image blocks based on the fused image. An object image block is an image block containing a single object. In some embodiments, the image analysis module may extract color features and shape features of a plurality of object image patches. In some embodiments, the image analysis module may acquire the cell attributes of the multiple objects and the cell parameters associated with the cell attributes of the multiple objects based on the color features and shape features of the image blocks of the multiple objects.
  • the image analysis module may further process the fused image based on the image recognition model, obtain the cellular properties of the multiple objects, and the cellular parameters associated with the cellular properties of the multiple objects.
  • the image recognition model may be a machine learning model.
  • the evaluation module can be used to evaluate the killing potency and/or immune activity of effector cells based on cellular parameters. In some embodiments, the evaluation module can evaluate the killing efficacy and/or immune activity of effector cells based on one or more of target cell mortality, cell-specific killing rate, and effector cell self-injury rate among cellular parameters.
  • the detection system 100 may also include a sample stage module and an automatic sample exchange module.
  • the stage module can be used to hold cell culture plates.
  • the cell culture plate has multiple sample wells for carrying co-culture samples with at least two effect-to-target ratios. Multiple microscopic images of fixed regions of co-culture samples.
  • the automatic sample exchange module can be used to exchange cell culture plates.
  • the microscopic imaging module images the fixed areas of the co-culture sample on the replaced cell culture plate respectively, and obtains multiple microscopic images of the fixed areas on the replaced cell culture plate.
  • the above description of the detection system and its modules is only for the convenience of description, and does not limit the description to the scope of the illustrated embodiments. It can be understood that for those skilled in the art, after understanding the principle of the system, various modules may be combined arbitrarily, or a subsystem may be formed to connect with other modules without departing from the principle.
  • the microscopic imaging module, the image analysis module, and the evaluation module disclosed in FIG. 1 may be different modules in a system, or a module may implement the functions of the above two or more modules.
  • the microscopic imaging module and the fluorescence microscopic imaging module can be the same module, which can acquire both bright-field microscopic images of the co-cultured samples and fluorescence microscopic images of the co-cultured samples.
  • each module may share one storage module, and each module may also have its own storage module. Such deformations are all within the protection scope of this specification.
  • FIG. 2 is an exemplary flowchart of a method for detecting cell killing efficacy and/or immune activity according to some embodiments of the present specification. As shown in FIG. 2 , the process 200 includes steps 210 to 230 .
  • Step 210 is the step of acquiring a microscopic image.
  • a plurality of microscopic images of the fixed region of the co-culture sample are acquired.
  • the microscopic imaging module may perform step 210 .
  • the microscopic imaging module may acquire a plurality of microscopic images via the image acquisition device 140.
  • the microscopic imaging module may obtain a plurality of pre-acquired microscopic images from the storage device 140 via the network 120 .
  • Co-culture samples are cell samples obtained by co-culture of target cells and effector cells.
  • the fixed area of the co-culture sample contains a plurality of targets, and the plurality of targets are cell groups containing cells with different properties, each target has image-recognizable features, and the cell properties of each target are passed through the An image of an object can be characterized by a collection of feature information that identifies features displayed on multiple microscopic images.
  • the fixed area may be the entire image acquisition area of the co-culture sample that includes all targets.
  • the immobilization region may be a portion of the image acquisition region of the co-culture sample that contains a portion of the target.
  • Target cells refer to various tumor cells or virus-infected cells corresponding to immune cells.
  • the target cells are virus-infected cells and/or tumor cells.
  • Tumor cells and virus-infected cells that can be target cells include, but are not limited to, K562 cells, Daudi cells, Jurkat cells, MCF-7 cells, A549 cells, HepG2 cells, and the like.
  • Effector cells refer to immune cells or engineered cells that participate in removing foreign antigens and performing effector functions in an immune response.
  • the effector cells are immune cells and/or engineered cells.
  • Immune cells and engineered cells that can be used as effector cells include but are not limited to PBMC cells, NK cells, T cells, CTL cells, LAK cells, CIK cells, TIL cells, DC cells, CAR-T cells, CAR-NK cells, NK92MI -CD16a cells, etc.
  • Cell properties can be the manifestation of one or more of a series of cell life phenomena (eg, growth, development, proliferation, differentiation, inheritance, metabolism, stress, exercise, aging, and death).
  • cell properties may include cell type and cell survival state.
  • cell types can include target cells and effector cells; cell survival states can include live and dead cells.
  • living cells are cells that can carry out metabolism, reproduction and replication, and are mainly characterized by complete cell membrane and selective permeability.
  • Dead cells are cells that cannot normally perform biological functions, metabolism, reproduction and replication, mainly manifested as cell membrane damage and loss of selective permeability.
  • cell death methods can include dead cells produced by cell death processes such as apoptosis and necrosis, ferroptosis, pyroptosis, and autophagy.
  • cell types can include target cells, effector cells, cell debris, and impurities; and cell survival states can include living cells, apoptotic cells, and necrotic cells.
  • apoptotic cells are cells that die autonomously and orderly under the control of genes in order to maintain the stability of the internal environment.
  • Necrotic cells refer to cells that have been damaged and died by extreme physical, chemical factors or severe pathological stimuli.
  • the plurality of targets within the immobilized area are live target cells, dead target cells, live effector cells, and cell groups of dead effector cells.
  • Objects with different cellular properties have different image-recognizable features, so that the image-recognizable features represent the cellular properties of the object.
  • the image identifiable features of the target can have many different specific types, and different microscopic image combinations are used to perform image analysis on the different types of image identifiable features.
  • the image identifiable features include fluorescent marker features.
  • the cell properties of the target can be characterized by different combinations of fluorescent markers .
  • the cellular properties of the target are characterized by a combination of three fluorescent labels.
  • the co-culture sample is obtained by labeling with three fluorescent labels, and the steps of obtaining the co-culture sample include: co-culture based on target cells carrying preset fluorescent labels and effector cells without fluorescent labels to obtain a co-culture; co-culture; After culturing for a predetermined period of time, the co-cultures were labeled with fluorescent markers of total cells and fluorescent markers of dead cells, respectively, to obtain co-culture samples.
  • the identifiable features of the image characterizing the cell attributes of the target are specifically: in the fixed area of the co-culture sample, the target carrying the preset fluorescent label and the total cell fluorescent label is a living target cell, and the target carrying the preset fluorescent label and the total cell
  • the fluorescently labeled and dead cell fluorescently labeled targets are dead target cells, the targets that only carry total cells fluorescently labeled are live effector cells, and the targets that carry total cells fluorescently labeled and dead cell fluorescently labeled are dead effector cells.
  • the preset fluorescent labels may be fluorescent proteins or cell dyes.
  • the fluorescent protein that can be used as a preset fluorescent label is green fluorescent protein (GFP) or red fluorescent protein (RFP).
  • the cell dye that can be used as a preset fluorescent marker is carboxyfluorescein diacetate succinimidyl ester (CFSE) or calcein AM (Calcein-AM). It should be noted that for target cells labeled with cell dyes that are prone to background fluorescence, the target cells need to be washed before co-culture with effector cells.
  • Total Cell Fluorescence Labeling can label all cells in a co-culture sample.
  • the total cellular fluorescent marker can be a nuclear dye.
  • Nuclear dyes that can be used as fluorescent markers for total cells include, but are not limited to, Hoechst33342, DAPI, and the like.
  • Dead cells fluorescent labeling can only label dead cells in co-culture samples.
  • the dead cell fluorescent label can be any dead cell labeling dye.
  • Dead cell labeling dyes that can be used as fluorescent markers of dead cells include but are not limited to Annexin-V (Annexin-V), SYTOX Green cyanine (SYTOX Green), propidium bromide (PI) and 7-aminoactinomycin D (7-AAD) et al.
  • the cellular properties of the target are characterized by a combination of two fluorescent labels.
  • the co-culture sample is obtained by labeling with two fluorescent labels, and the steps of obtaining the co-culture sample include: co-culture based on target cells carrying preset fluorescent labels and effector cells without fluorescent labels to obtain a co-culture; co-culture; After culturing for a predetermined time, the co-culture is labeled with a dead cell fluorescent marker to obtain the co-culture sample.
  • the identifiable features of the image characterizing the cell attributes of the target are specifically: in the fixed area of the co-culture sample, the target carrying only the preset fluorescent label is the live target cell, and the target carrying the preset fluorescent label and the dead cell fluorescent label is the target.
  • the target is dead target cells, the target without fluorescent label is live effector cells, and the target only carrying the fluorescent label of dead cells is dead effector cells.
  • the image identifiable features include fluorescent marker features and cell diameter features.
  • the cell attribute of each target in the fixed area is one of live target cells, dead target cells, live effector cells and dead effector cells.
  • the cell attribute of the target can be determined by using the difference between the fluorescent label feature and the cell diameter feature. represented by different combinations.
  • the cellular properties of the target are characterized by a combination of a fluorescent label and different cell diameters of effector/target cells.
  • the co-culture sample is obtained by labeling with a fluorescent label, and the steps of obtaining the co-culture sample include: co-culturing based on target cells without fluorescent label and effector cells without fluorescent label to obtain a co-culture; co-culture predetermined The co-culture samples were obtained by labeling the co-cultures with a dead cell fluorescent label after time.
  • the identifiable features of the image that characterize the cell properties of the target are specifically: in the fixed area of the co-culture sample, the target without fluorescent label and having a diameter greater than or equal to the minimum diameter of the target cell is a live target cell, carrying a dead cell fluorescent label and having a diameter of Targets with a diameter greater than or equal to the minimum diameter of target cells are dead target cells, targets without fluorescent labels and diameters smaller than the maximum diameter of effector cells are live effector cells, and targets with fluorescent markers of dead cells and diameters smaller than the maximum diameter of effector cells are dead effectors cell.
  • step 210 further includes:
  • Step 211 acquiring a bright-field microscopic image of the fixed area of the co-culture sample
  • Step 212 determining image identifiable features that characterize cell properties of multiple targets in the co-culture sample
  • Step 213 acquiring at least one fluorescence microscopic image of the fixed region of the co-culture sample, and the imaging parameters of the at least one fluorescence microscopic image are determined based on the identifiable features of the images of the multiple objects.
  • the brightfield microscopic image is an image acquired by the image acquisition device 140 irradiating the cell sample with a brightfield light source.
  • the background of the field of view in brightfield microscopy images is bright, while the edges of cells in the cell sample are dark.
  • the fluorescence microscope image is an image acquired after the image acquisition device 140 irradiates the cell sample with an excitation light source to cause the cell sample to emit fluorescence. Fluorescence microscopy images reflect the shape and location of cells in a cell sample.
  • the format of the microscopic image may include the Joint Photographic Experts Group (JPEG) image format, the Tagged Image File Format (TIFF) image format, the Graphics Interchange Format (GIF) image format, the Kodak Flash PiX (FPX) image format And Digital Imaging and Communications in Medicine (DICOM) image format, etc.
  • JPEG Joint Photographic Experts Group
  • TIFF Tagged Image File Format
  • GIF Graphics Interchange Format
  • FPX Kodak Flash PiX
  • DICOM Digital Imaging and Communications in Medicine
  • the imaging parameters of the at least one fluorescence microscopy image include fluorescence channel type and excitation light wavelength.
  • the type of the fluorescence channel and the corresponding excitation light wavelength of the fluorescence microscopic image used by the detection system 100 for image analysis are determined according to the image identifiable features of the cell properties of the multiple targets in the co-culture sample.
  • the imaging parameters of at least one fluorescence microscopic image are determined from specific image-identifiable features of the cellular properties of the plurality of targets in the co-culture sample.
  • the image identifiable feature characterizing the cellular property of the target is a combination of a preset fluorescent marker, a total cell fluorescent marker, and a dead cell fluorescent marker
  • the at least one fluorescent microscopic image includes the first fluorescent microscopic image , a second fluorescence microscope image, and a third fluorescence microscope image.
  • the fluorescence channel for collecting the first fluorescence microscopic image and the excitation light wavelength of the fluorescence channel match the preset fluorescent label
  • the fluorescence channel for collecting the second fluorescence microscopic image and the excitation light wavelength of the fluorescence channel match the total cell fluorescence label
  • the fluorescence channel for collecting the third fluorescence microscopic image and the excitation light wavelength of the fluorescence channel match the dead cell fluorescent label.
  • the imaging parameters of at least one fluorescence microscopic image are determined from specific image-identifiable features of the cellular properties of the plurality of targets in the co-culture sample.
  • the identifiable feature of the image characterizing the cell property of the target is a combination of a preset fluorescent marker and a dead cell fluorescent marker
  • the at least one fluorescence microscope image includes a first fluorescence microscope image and a third fluorescence microscope image. Micro image.
  • the fluorescence channel for collecting the first fluorescence microscopic image and the excitation light wavelength of the fluorescence channel match the preset fluorescent marker
  • the fluorescence channel for collecting the third fluorescence microscopic image and the excitation light wavelength of the fluorescence channel match the dead cell fluorescent marker
  • the imaging parameters of the at least one fluorescence microscopic image are determined from specific image-identifiable features of the cellular properties of the plurality of targets in the co-culture sample.
  • the image identifiable feature characterizing the cellular property of the target is a combination of fluorescent markers of dead cells and different types of cell diameters
  • the at least one fluorescence microscopic image is a third fluorescence microscopic image.
  • the fluorescence channel for collecting the third fluorescence microscopic image and the excitation light wavelength of the fluorescence channel match the dead cell fluorescent label.
  • step 210 further includes the step of obtaining a plurality of control group microscopic images of a fixed region of the control group effector cell sample.
  • the effector cell sample of the control group is a cell sample obtained by culturing the effector cells alone, and the fixed area of the target cell sample of the control group contains a plurality of first control objects with identifiable features of images.
  • the image identifiable features characterizing the cellular properties of the second control target are consistent with the image identifiable features characterizing the cellular properties of the target.
  • step 210 further includes the step of obtaining a plurality of control group microscopic images of a fixed region of the control group target cell sample.
  • the target cell sample of the control group is a cell sample obtained by culturing the target cells alone, and the fixed area of the target cell sample of the control group includes a plurality of second control objects with identifiable features in images.
  • the image identifiable features characterizing the cellular properties of the second control target are consistent with the image identifiable features characterizing the cellular properties of the target.
  • Step 220 is a step of performing image analysis.
  • image overlay synthesis analysis or image fusion analysis is performed on a plurality of microscopic images to obtain cell attributes of multiple targets, and cell parameters associated with the cell attributes are counted.
  • the image analysis module may perform step 220 .
  • image overlay synthesis analysis is performed on a plurality of microscopic images to obtain cellular properties of a plurality of objects, and cell parameters associated with the cellular properties are counted.
  • FIG. 3 For a specific description of performing image overlay synthesis analysis on multiple microscopic images, reference may be made to FIG. 3 and related descriptions thereof, and details are not repeated here.
  • image fusion analysis is performed on a plurality of microscopic images, the cellular properties of the plurality of objects are obtained, and the cellular parameters associated with the cellular properties are counted.
  • performing image fusion analysis on a plurality of microscopic images to obtain cell properties of a plurality of targets, and statistical cell parameters associated with the cell properties further include:
  • Analyze the fused images obtain the cellular properties of multiple targets, and count the cellular parameters associated with the cellular properties.
  • FIG. 5 For a specific description of acquiring a fused image, reference may be made to FIG. 5 and related descriptions, which will not be repeated here.
  • FIG. 6 For a specific description of analyzing the fused image, reference may be made to FIG. 6 and related descriptions, which will not be repeated here.
  • the cellular parameter may include a first cellular parameter that correlates to cellular properties of the plurality of targets of the co-culture sample.
  • the first cell parameter can be obtained by classifying, counting and counting a plurality of targets based on cell properties.
  • the first cell parameter may include total number of target cells and effector cells, total number of target cells, total number of live target cells, total number of dead target cells, target cell death rate, total number of effector cells, total number of live effector cells, total number of dead effector cells One or more of total number and effector cell mortality.
  • the target cell death rate and the effector cell death rate were calculated using the following formula:
  • Target cell death rate total number of dead target cells / total number of target cells ⁇ 100%;
  • Effector cell death rate total number of dead effector cells/total number of effector cells ⁇ 100%.
  • step 220 further includes performing an overlay synthesis analysis based on multiple control group microscopic images of the target cell samples of the control group, and obtaining The steps of comparing the cellular properties of a plurality of first control objects and statistically correlating the cellular properties of the cellular properties.
  • the cellular parameter may further include a second cellular parameter that correlates to cellular properties of the plurality of first control targets of the control target cell sample.
  • the second cell parameter can be obtained by classifying and counting the plurality of first control targets based on cell properties.
  • the second cell parameter may include one or more of the total number of target cells in the control group, the total number of live target cells in the control group, the total number of dead target cells in the control group, the death rate of the control group target cells, and the cell-specific killing rate. .
  • the target cell death rate and cell-specific killing rate in the control group were calculated using the following formula:
  • the death rate of target cells in the control group the total number of dead target cells in the control group/the total number of target cells in the control group ⁇ 100%;
  • Cell-specific killing rate target cell death rate-target cell death rate in the control group.
  • step 220 further includes performing overlay synthesis analysis based on multiple control group microscopic images of the control group effector cell sample, and obtaining Steps of performing a plurality of second control cellular properties of the target and statistically correlating the cellular parameters of the cellular properties.
  • the cellular parameter may further include a third cellular parameter that correlates to cellular properties of the plurality of second control targets of the control effector cell sample.
  • the third cell parameter can be obtained by classifying and counting the plurality of second control targets based on cell properties.
  • the third cell parameter may include one or more of the total number of effector cells in the control group, the total number of live effector cells in the control group, the total number of dead effector cells in the control group, the death rate of the control group effector cells, and the self-injury rate of the effector cells. kind. Specifically, the effector cell death rate and effector cell self-injury rate in the control group were calculated using the following formula:
  • the death rate of effector cells in the control group the total number of dead effector cells in the control group/the total number of effector cells in the control group ⁇ 100%;
  • Effector cell self-injury rate effector cell death rate-controller effector cell death rate.
  • Step 230 is a step of evaluating cell killing efficacy and/or immune activity. In step 230, based on the cellular parameters, the killing potency and/or immune activity of the effector cells is assessed. In some embodiments, the evaluation module may perform step 230 .
  • a combination of one or more of target cell mortality, effector cell mortality, cell-specific killing rate, and effector cell self-injury rate among cellular parameters can be used to characterize the killing efficacy and/or immunity of effector cells active.
  • one cell parameter or a combination of multiple cell parameters can intuitively reflect the effector cell killing potency level and/or immune activity level by comparing with the parameter threshold.
  • the effector cells of the sample to be tested can be evaluated by comparing the corresponding cell parameter or combination of cell parameters of the sample to be tested with the parameter threshold
  • the level of killing efficacy and/or immune activity relative to the control sample, the control sample is selected according to the different evaluation purposes.
  • the corresponding cell parameter of the sample to be tested or the combination of cell parameters and the control group can evaluate the effector cell killing efficacy and/or immune activity of the sample to be tested relative to the control group, and the control group is selected according to the different evaluation purposes.
  • evaluating the killing efficacy and/or immune activity of effector cells based on the cellular parameters may further include: comparing target cell mortality with a mortality threshold, wherein the mortality threshold includes an upper limit and a lower limit, and evaluating the effect according to the comparison result Killing potency and/or immune activity of cells.
  • tumor cells target cells
  • natural killer cells effector cells
  • Target cell death was detected in co-culture samples from blank control mice.
  • the average range of target cell death rate was obtained as the death rate threshold, which was used to represent the overall average value of target cell death rate in normal mice.
  • the tumor cells and the natural killer cells of the experimental group mice were co-cultured, and the above detection method or detection system was used to detect the target cell death rate of the co-cultured samples of the experimental group mice.
  • Comparative analysis of target cell mortality and mortality threshold of mice in the experimental group if the target cell death rate of the mice in the experimental group is higher than the upper limit of the mortality threshold, the natural killer cell killing efficacy and/or immune activity of the mice in the experimental group Higher than the normal level, indicating that the drug to be tested can improve the killing efficacy and/or immune activity of natural killer cells in mice; if the target cell death rate of the mice in the experimental group is between the upper and lower limits of the mortality threshold, the experimental group is small.
  • the killing efficacy and/or immune activity of natural killer cells in mice are at normal levels, indicating that the tested drug does not improve the killing efficacy and/or immune activity of natural killer cells in mice; if the target cell death rate of mice in the experimental group is lower than the death rate The lower limit of the threshold, the natural killer cell killing efficacy and/or immune activity of the mice in the experimental group is lower than the normal level, indicating that the drug to be tested can reduce the natural killer cell killing efficacy and/or immune activity of mice.
  • FIG. 3 is an exemplary flowchart of image overlay composite analysis of multiple microscopic images according to some embodiments of the present specification.
  • process 300 may be performed by an image analysis module. As shown in FIG. 3 , the process 300 includes steps 310 to 340 .
  • Step 310 is the step of extracting the contour of the object in the microscopic image.
  • multiple object regions and corresponding contour information in each microscopic image are extracted.
  • the target area refers to the image area containing a single target in the microscopic image, and the edge of the target area is the outer contour of the corresponding target.
  • the contour information may be related information representing the contour features of the target object.
  • the profile information may include one or more of target object location information, target object size information, and target object fluorescence information.
  • the target object position information includes, but is not limited to, the coordinate information of the outer contour pixel points of the target object on the microscopic image, the coordinate information of the target object contour feature points (such as center and centroid), and the like.
  • the target object size information includes, but is not limited to, diameter information, contour area information, and the like.
  • the target object fluorescence information includes, but is not limited to, fluorescence intensity information, color information, and the like.
  • Step 320 is a step of determining the coincidence of the objects for the plurality of microscopic images.
  • the target object coincidence determination is performed based on the target object regions of a plurality of microscopic images and the corresponding contour information, and the coincidence determination result is obtained.
  • the determination of the coincidence of the target objects includes a primary coincidence determination based on the calculation of the feature point coordinate distance and a secondary coincidence determination based on the calculation of the intersection ratio. Further, step 320 includes:
  • the determination result of the initial coincidence determination is the determination result of the current round of target overlap determination
  • a secondary coincidence determination will be made based on the compared two target object areas, and the judgment result of the secondary coincidence determination is the determination of the current round of target object coincidence determination. result.
  • the determination of the coincidence of the target objects includes a primary coincidence determination based on the calculation of the feature point coordinate distance and a secondary coincidence determination based on the calculation of the intersection ratio.
  • the step of primary coincidence determination further includes:
  • the feature points are the feature pixels of the target area on the corresponding microscopic image.
  • the feature point coordinates are the pixel coordinates of the feature points of the target area on the corresponding microscopic image, which can be extracted based on the contour information of the target area.
  • the feature point may be one of the center of the target area, the center of mass, and the center of gravity.
  • the feature point may be the center of the target area.
  • the distance threshold is the determination limit of coincidence or not in the initial coincidence determination. Ideally, if different microscopic images of the target object are collected in the same fixed area, the coordinate distance of the feature points of the corresponding target area of the same target in different microscopic images can be zero.
  • the wavelength of the light source (bright-field light source and/or excitation light source), imaging channel (bright-field light source), and imaging channel (bright-field light source) for acquiring each microscopic image Due to the influence of factors such as different exposure times of field channels and/or fluorescence channels, the corresponding target area of the same target on different microscopic images may have a position shift of several pixels to more than a dozen pixels.
  • the distance threshold can be set as The initial coincidence determination based on the feature point coordinate distance calculation provides a tolerance space.
  • the distance threshold may be set based on the selected light source wavelength of the image acquisition device and/or the selected imaging channel exposure time. In some alternative embodiments, the distance threshold may be set based on user input. The setting method of the distance threshold is not limited here.
  • the step of secondary coincidence determination further includes:
  • the target area intersection ratio is the ratio of the area of the intersection between the two target areas to the area of the union, which can be used to evaluate the degree of overlap between the two target areas.
  • intersection and union ratio threshold is the judgment limit of coincidence or not in the secondary coincidence judgment.
  • the cross-union ratio threshold may be set based on user input.
  • Step 330 is a step of determining the cellular properties of the target.
  • the cellular properties of the target can be determined based only on the coincidence determination results.
  • the above method for determining the properties of the target cells is suitable for the superimposed synthesis analysis of multiple microscopic images of co-culture samples with two types and three types of fluorescent labels.
  • the three fluorescent labels are respectively preset fluorescent label, total cell fluorescent label, and dead cell fluorescent label.
  • Multiple microscopic images required for image overlay synthesis analysis include brightfield microscopic image, first fluorescent microscopic image matching preset fluorescent markers, second fluorescent microscopic image matching total cell fluorescent markers, and matching dead cell fluorescence Labeled third fluorescence microscopy image.
  • the coincidence judgment result of the target object coincidence judgment includes the set of feature information displayed on multiple microscopic images of the fluorescent marker feature of each target object, that is, the coincidence judgment result includes the set of fluorescent marker feature information of each target object.
  • the judgment results can be used to determine the cellular properties of the target.
  • the step of determining the cell property of the target further includes:
  • the properties of cells are living target cells
  • the properties of cells are live effector cells
  • Cell properties are dead effector cells
  • the cellular properties are cellular debris or impurities.
  • the two fluorescent labels are a preset fluorescent label and a dead cell fluorescent label, respectively.
  • the multiple microscopic images required to perform the image overlay synthesis analysis include a brightfield microscopic image, a first fluorescent microscopic image matching the preset fluorescent markers, and a third fluorescent microscopic image matching the dead cell fluorescent markers.
  • the coincidence judgment result of the target object coincidence judgment includes the set of feature information displayed on multiple microscopic images of the fluorescent marker feature of each target object, that is, the coincidence judgment result includes the set of fluorescent marker feature information of each target object.
  • the judgment results can be used to determine the cellular properties of the target. Based on the coincidence determination result, the step of determining the cell property of the target further includes:
  • step 330 includes:
  • the cell property of the corresponding target is determined.
  • the cell diameter determination further includes: comparing the diameter information of the target object with a preset minimum diameter of the target cell and a preset maximum diameter of the effector cell to obtain a diameter determination result; wherein, the diameter of the target object is greater than or equal to the minimum diameter of the target cell , then it is determined that the target is a target cell; if the diameter of the target is smaller than the minimum diameter of the effector cell, it is determined that the target is an effector cell.
  • Carrying out diameter determination based on the contour information of the target area, and obtaining the diameter determination result further includes: comparing the diameter of the target area with the preset minimum diameter of target cells and the preset maximum diameter of effector cells, and determining the target corresponding to the target area as Target cells or effector cells to obtain diameter determination results.
  • each fluorescent label is a dead cell fluorescent label.
  • the multiple microscopic images required to perform the image overlay synthesis analysis include a brightfield microscopic image and a third fluorescent microscopic image that matches the fluorescent markers of dead cells.
  • the coincidence judgment result of the target coincidence judgment includes the collection of feature information displayed on multiple microscopic images of the fluorescent marker feature and cell diameter feature of each target, that is, the coincidence judgment result includes the fluorescent marker feature information of each target. And the collection of cell diameter feature information, the coincidence determination result can be used to determine the cell properties of the target. Its step of determining the cellular properties of the target further comprises:
  • the target object that overlaps the target area in the brightfield microscopic image and the third fluorescence microscopic image, and the target object is determined as a dead cell by the target coincidence judgment; extract the target whose cell attribute is a dead cell in the coincidence judgment result Compare the diameter information of the target with the preset minimum diameter of the target cell and the preset maximum diameter of the effector cell, and further determine that the target whose diameter is greater than or equal to the minimum diameter of the target cell and whose cell attribute is a dead cell is a dead target cell. It is further determined that the target whose diameter is smaller than the maximum diameter of the effector cell and whose cell property is a dead cell is a dead effector cell;
  • the target object is determined as a living cell by the target object coincidence judgment;
  • the cell attribute is the diameter information of the target object of living cells; compare the diameter information of the target object with the preset minimum diameter of the target cell and the preset maximum diameter of the effector cell, and further determine that the diameter is greater than or equal to the minimum diameter of the target cell and the cell attribute is that of the living cell.
  • the target is a living target cell, and it is further determined that the target whose diameter is smaller than the maximum diameter of the effector cell and whose cell attribute is a living cell is a living effector cell.
  • Step 340 is the step of acquiring cell parameters.
  • a plurality of targets are classified, counted and counted based on cell attributes to obtain cell parameters. Specifically, according to the different cell attributes determined in step 330, all targets are classified, counted and counted to obtain cell parameters.
  • the cell parameters For the specific description of the cell parameters, reference may be made to step 220 and related descriptions thereof, which will not be repeated here.
  • Process 300 may also include the step of generating an overlay image based on the plurality of microscopy images.
  • the image analysis module may perform image fusion on multiple microscopy images to generate an overlay image.
  • the process 300 further includes the step of identifying a plurality of objects in the overlay image based on cell properties. For a specific description of image fusion for multiple microscopic images, reference may be made to FIG. 5 and related descriptions, which will not be repeated here.
  • the image analysis module can be connected to an output device (display screen) to output the generated unmarked overlay image or the marked overlay image.
  • the target object properties obtained by the image overlay synthesis analysis are directly marked on the overlay image, and output to the user, so that the way of obtaining the analysis results is faster, more intuitive and more efficient.
  • FIG. 4 is an exemplary flowchart for extracting object contours in microscopic images according to some embodiments of the present specification.
  • process 400 may be performed by an image analysis module. As shown in FIG. 4 , the process 400 includes steps 410 to 430 .
  • Step 410 is a step of denoising the microscopic image.
  • filtering is performed based on each microscopic image to obtain a denoised microscopic image.
  • the filtering process includes median filtering and/or Gaussian filtering.
  • Step 420 is a step of binarizing the microscopic image.
  • binarization processing is performed based on each denoised microscopic image to obtain a binarized microscopic image.
  • Binarization is the process of making microscopic images appear black and white.
  • Step 430 is the step of image segmentation and contour extraction.
  • target object segmentation is performed based on each binarized microscopic image, and multiple target object regions and corresponding contour information are extracted.
  • Object segmentation is the process of dividing the binarized microscopic image into several object regions containing a single object and extracting the information of the object of interest.
  • the object segmentation is one of threshold segmentation, region growing method, watershed segmentation and statistical segmentation.
  • FIG. 5 is an exemplary flowchart of acquiring a fused image according to some embodiments of the present specification.
  • process 500 may be performed by an image analysis module. As shown in FIG. 5 , the process 500 includes steps 510 to 530 .
  • step 510 the fusion feature points of each microscopic image are extracted.
  • the fusion feature points are the pixel points of the same spatial point on the co-culture sample on different multiple microscopic images (bright field microscopic image and at least one fluorescence microscopic image).
  • the fusion feature points may correspond to the same target feature on the co-culture sample.
  • the same object features may include color features, texture features, shape features, and the like.
  • the image analysis module can search for fusion feature points by manual search, automatic search, and semi-automatic search.
  • the image fusion module may also select the found fusion feature points through similarity measurement.
  • the similarity measure may include any combination of one or more of mutual information based measures, Fourier analysis based measures, and the like.
  • the fused feature points of multiple microscopic images may also correspond to the same location coordinates on the co-culture sample.
  • the fusion feature point may be a central location point on the bright-field microscopic image and the at least one fluorescence microscopic image.
  • step 520 the plurality of microscopic images are registered based on the corresponding fused feature points of the plurality of microscopic images.
  • Registration is the determination of the correspondence between multiple spatial points on a co-culture sample and pixel points on multiple microscopic images.
  • the image analysis module can find the correspondence through a registration algorithm.
  • the image analysis module can use a registration algorithm based on at least part of the pixels on a certain arc on a certain object in the brightfield microscope image and the arc on the cell in at least one fluorescence microscope image. Correspondences between at least part of the pixel points on the line are found to find the correspondence between the brightfield microscopic image and the at least one fluorescence microscopic image.
  • registration algorithms may include point-based registration algorithms (eg, signature-based registration algorithms), curve-based registration algorithms, surface-based registration algorithms (eg, surface contour-based registration algorithms) registration algorithm), spatial alignment registration algorithm, cross-correlation configuration registration algorithm, mutual information-based registration algorithm, sequential similarity detection algorithm (SSDA), nonlinear transformation registration algorithm, B-spline registration algorithm, etc., or any combination thereof.
  • point-based registration algorithms eg, signature-based registration algorithms
  • curve-based registration algorithms eg, surface-based registration algorithms (eg, surface contour-based registration algorithms) registration algorithm)
  • spatial alignment registration algorithm eg, cross-correlation configuration registration algorithm
  • SSDA sequential similarity detection algorithm
  • nonlinear transformation registration algorithm e.g., B-spline registration algorithm, etc., or any combination thereof.
  • step 530 the registered multiple microscopic images are fused based on transparency and/or chromaticity to obtain a fused image.
  • Fusion is the synthesis of information from multiple microscopic images into one microscopic image.
  • the image analysis module can Micro-images are fused.
  • the image fusion module may fuse the registered multiple microscopic images based on transparency. Fusion based on transparency is to overlap microscopic images with different transparency, and use the overlapped multiple microscopic images as a fusion image. In some embodiments, blending based on transparency may include alpha blending.
  • the image fusion module may perform fusion based on chroma. Fusion based on chromaticity is to perform a specific operation on the chromaticity of different microscopic images to obtain the chromaticity of the fused image, thereby obtaining the fused image.
  • the pixel point A' in the registered bright field microscopic image has a corresponding relationship with the pixel point A" in the fluorescence microscopic image, based on the chromaticity RGB (220, 200, 100) of the pixel point A' and the The average value of each color component of chroma RGB (0, 200, 200) can obtain the chroma RGB (110, 200, 150) of corresponding pixel A in the fusion image.
  • the image analysis module may further fuse the fused image obtained based on transparency and the fused image obtained based on chrominance to obtain a final fused image.
  • Some embodiments of the present specification fuse brightfield and fluorescence microscopy images based on transparency and/or chromaticity, which can fuse the color of objects in different fluorescence microscopy images while preserving the morphological characteristics of the objects in the different images feature, so that the fused image contains more information, thereby improving the accuracy of cell killing efficacy and/or immune activity.
  • the fusion may also include, but is not limited to, a combination of one or more of the Poisson fusion algorithm, the linear fusion algorithm, and the Collage algorithm.
  • FIG. 6 is an exemplary flowchart of analyzing a fused image according to some embodiments of the present specification.
  • process 600 may be performed by an image analysis module. As shown in FIG. 6, the process 600 includes the following steps.
  • step 610 based on the fused image, a plurality of target object image blocks are acquired.
  • An object image block is an image block containing a single object.
  • the image analysis module may obtain target image patches from the fused image through a detection algorithm.
  • the detection algorithm may segment the fused image and detect a single object according to the characteristics of the segmented image blocks. Specifically, the detection algorithm can first extract multiple image blocks from the fused image through a multi-scale sliding window (Sliding-window), selective search (Selective Search), neural network or other methods, and then extract multiple image blocks from the fused image. The initial features of each image block, and finally determine whether the image block is the target image block based on the initial features of the image block. where the initial features are the shallow features of the image patch. For example, the initial features can only reflect whether the image block contains closed lines, but cannot reflect the specific shape of the lines.
  • step 620 the color feature and shape feature of the target image block are extracted.
  • the color feature is related information representing the color of the target image block, and can reflect the color of the target in the target image block.
  • the color feature may be based on the chromaticity representation of each pixel in the target image block in different color components.
  • the color feature can be represented by the chromaticity of each pixel in the target image block on the red component R, the green component G, and the blue component B, respectively.
  • color features may be represented by other means (eg, color histograms, color moments, color sets, etc.).
  • histogram statistics are performed on the chromaticity of each pixel in the color component of the target image block to generate a histogram representing color features.
  • a specific operation eg, mean, squared difference, etc.
  • the result of the specific operation represents the color feature of the target image block.
  • the image analysis module can extract the color features of the target image block through a color feature extraction algorithm.
  • Color feature extraction algorithms include: color histogram, color moment, color set, color aggregation vector and color correlation graph.
  • the image analysis module can count the gradient histogram based on the chromaticity of each pixel in each color component in the target image block, so as to obtain the color histogram.
  • the image analysis module can divide the target image block into multiple regions, and use the set of binary indices of the multiple regions established by the chromaticity of each color component of each pixel in the target image block to determine the target image block. Describe the color set of the target image patch.
  • the shape feature is related information representing the contour and area of the target image block, and can reflect the shape of the target in the target image block.
  • the image analysis module can detect parallel lines through boundary feature method, Hough transform method, boundary direction histogram method, Fourier shape descriptor method, shape parameter method (shape factor), finite Element method (Finite Element Method or FEM), rotation function (Turning) and wavelet descriptor (Wavelet Deor), etc. to obtain shape features.
  • step 630 based on the color features and shape features of the image blocks of the multiple objects, the cell attributes of the multiple objects are acquired, and the cell parameters associated with the cell attributes are counted.
  • the image analysis module may determine the color and shape of multiple objects corresponding to the multiple object image blocks in the fused image based on the color features and shape features of the multiple object image blocks, and then based on each The color and shape of the target object, the cell properties of the target object are obtained, and the cell parameters associated with the cell properties are counted.
  • Some embodiments of the present specification determine the cell attributes of each target in the fusion image based on the color features and shape features of the target image blocks, and then statistically analyze the cell parameters based on the corresponding cell attributes of each target, which can improve the cell killing efficiency and/or the accuracy of immune activity assays.
  • the image analysis module can process the fused image based on the image recognition model, obtain the cell attributes of multiple objects, and count the cell parameters associated with the cell attributes.
  • the image recognition model can be a machine learning model with preset parameters.
  • Machine learning models that can be used as image recognition models include, but are not limited to, object detection models, semantic segmentation models, instance segmentation models, and the like.
  • Preset parameters refer to the model parameters learned during the training of the machine learning model. Taking a neural network as an example, the model parameters include weight and bias.
  • FIG. 7 is an exemplary schematic diagram of an image recognition model according to some embodiments of the present specification.
  • the image recognition model may include a target image block acquisition layer 710 , a feature extraction layer 720 and an analysis layer 730 .
  • the image analysis module can implement steps 610-630 by using an image recognition model, acquire cell attributes of multiple targets, and count cell parameters associated with the cell attributes.
  • step 610 may be implemented based on the target object image block acquisition layer 710
  • step 620 may be implemented based on the feature extraction layer 720
  • step 630 may be implemented based on the analysis layer 730 .
  • the input of the object image block acquisition layer 710 is the fused image 740 and the output is a plurality of object image blocks 750 .
  • the type of the target image patch acquisition layer may include, but is not limited to, a Visual Geometry Group Network model, an Inception NET model, a Fully Convolutional Network model, and a segmentation network model. and Mask-Region Convolutional Neural Network models, etc.
  • the input of the feature extraction layer 720 is a plurality of object image blocks 750, and the output is the color feature 760 and the shape feature 770 corresponding to each object image block.
  • the type of feature extraction layer may include, but is not limited to, a Convolutional Neural Networks (CNN) model such as ResNet, ResNeXt, SE-Net, DenseNet, MobileNet, ShuffleNet, RegNet, EfficientNet, or Inception, or Recurrent Neural Network Model.
  • CNN Convolutional Neural Networks
  • the input of the analysis layer 730 is the color feature 760 and the shape feature 770 corresponding to each target image block, and the output is the cell attributes 780 of the plurality of targets, and the statistical cell parameters associated with the cell attributes.
  • the types of analysis layers may include, but are not limited to, fully connected layers, deep neural networks (DNNs), and the like.
  • the preset parameters of the image recognition model are generated through a training process.
  • the model acquisition module can train an initial image recognition model in an end-to-end manner based on multiple training samples with labels to obtain an image recognition model.
  • the training samples consist of labeled sample fusion images.
  • the labels of the training samples are the cell attributes of the objects in the sample fusion image and the cell parameters associated with the cell attributes.
  • the labels of the training samples can be obtained by manual labeling.
  • the image recognition model may be pre-trained by the processing device or a third party and stored in the storage device, and the processing device may directly call the image recognition model from the storage device.
  • Some embodiments of the present specification analyze fused images based on an image recognition model to obtain cell attributes of the target and cell parameters associated with the cell attributes, which can improve the efficiency of cell killing and/or immune activity detection; and based on the difference in labels of training samples , an image recognition model for acquiring different cell properties and cell parameters can be obtained, and the applicability and pertinence of cell killing efficacy and/or immune activity detection can be improved.
  • One of the embodiments of the present specification provides the application of a detection method or detection system for cell killing potency and/or immune activity in detecting cell killing potency, effector cell immune activity, preparing immune products, quality control of immune products or evaluating individual immune function.
  • ADCC antibody-dependent cytotoxicity
  • CDC complement-dependent cytotoxicity
  • ADCP Antibody-dependent cell-mediated phagocytosis
  • cell therapy drugs represented by CAR-T cell therapy also need to evaluate the biological function and quality control of the prepared CAR-T cell drugs by testing the cell killing efficacy, so as to ensure the effectiveness of CAR-T cell drugs. and security.
  • the above-mentioned applications of the detection method or detection system for cell killing efficacy and/or immune activity provided by some embodiments of the present specification all have positive significance.
  • test materials in the following examples are conventional methods unless otherwise specified.
  • the test materials used in the following examples were purchased from conventional biochemical reagent companies unless otherwise specified.
  • Tumor cells (K562 cell line) were prepared into a suspension with a cell concentration of 1 ⁇ 10 5 cells/mL, and 1 ⁇ L of 20 ⁇ M CFSE was added to 1 mL of the suspension for labeling, and incubated at 37°C for 30 min in the dark;
  • Experimental group 100 ⁇ L of CFSE-labeled K562 cells in step 2.1 and 100 ⁇ L of natural killer cells in step 2.2 were simultaneously added to the sample wells of 96-well plate as experimental group, the effect-target ratio was set to 10:1, 37 degrees 5% CO 2 incubators for a total of 4 hours. After the co-culture, Hoechst33342 (purchased from Thermofisher, USA) and PI dye (purchased from Sigma, USA) were added for staining to obtain co-culture samples.
  • Hoechst33342 purchased from Thermofisher, USA
  • PI dye purchased from Sigma, USA
  • Control group Add only 100 ⁇ L of the CFSE-labeled K562 cells in step 2.1 and 100 ⁇ L of culture medium to the sample wells of the 96-well plate as control target cells. Add only 100 ⁇ L of natural killer cells from step 2.2 and 100 ⁇ L of medium to the sample wells of a 96-well plate as a control group of natural killer cells.
  • the target cells of the control group and the natural killer cells of the control group were cultured and stained to obtain a target cell sample of the control group and a natural killer cell sample of the control group.
  • Experimental group Place the hemocytometer plate with the co-cultured samples on the sample stage of the detection instrument, and perform microscopic imaging in the fixed area of the co-cultured samples with brightfield and microfluorescence channels matching three fluorescent labels, respectively. Field microscopy images and three fluorescence microscopy images.
  • the information and order of the three fluorescence channels are as follows:
  • FL1 Ex 375nm, Em 460nm
  • FL2 Ex 480nm, Em 535nm
  • FL3 Ex 525nm, Em 600LP;
  • Control group For the hemocytometer plates added to the target cell samples of the control group and the natural killer cell samples of the control group, respectively, microscopic imaging was performed according to the method of the experimental group, and the bright-field microscopic images and fluorescence microscopic images of the control group were obtained. .
  • Target cell death rate (%) d/(b+d) ⁇ 100
  • Figures 8 to 10 are the fluorescence microscopic images under the FL1 channel, the FL2 channel, and the FL3 channel, respectively, and Figure 11 is the superimposed image of the fluorescence microscopic images under the FL1, FL2, and FL3 channels.
  • the area w, area x, area y, and area z in FIG. 11 all contain objects.
  • For the target in the region w it has FL1 signal at the overlapping regions w-1, w-2, and w-3 in Fig. 8 to Fig. 10 , no FL2 and FL3, and its cellular properties are live natural killer cells.
  • the target in the region x it has FL1 signal and FL2 signal at the overlapping regions x-1, x-2, and x-3 of Figure 8 to Figure 10, respectively, but has no FL3 signal, and its cellular properties are living target cells.
  • the target in the region y it has FL1 signal and FL3 signal at the overlapping regions y1, y2, and y3 in Fig. 8 to Fig. 10, respectively, but has no FL2 signal, and its cell property is a dead natural killer cell.
  • the target in the area z it has FL1 signal, FL2 signal and FL3 signal at the overlapping areas z1, z2 and z3 in Fig. 8 to Fig. 10 , and its cell attribute is a dead target cell.
  • the total number of live natural killer cells was 9
  • the total number of live target cells was 15
  • the total number of dead natural killer cells was 2, and the total number of dead target cells was 28.
  • the target cell death rate was 65.11%
  • the natural killer cell death rate was 18.18%.
  • Control group Referring to the detection method of the experimental group, image overlay synthesis analysis was performed on the bright-field microscopic images and fluorescence microscopic images of the control group to obtain the target cell death rate of the control group and the natural killer cell death rate of the control group, so as to calculate the cell death rate. Specific killing rate and natural killer cell self-injury rate.
  • the cell-specific killing rate (%) target cell death rate-target cell death rate in the control group
  • Natural killer cell self-injury rate (%) effector cell death rate - control group natural killer cell death rate.
  • Tumor cells (K562 cell line) were collected and prepared into a suspension with a cell concentration of 1 ⁇ 10 5 cells/mL. To 1 mL of the suspension, 1 ⁇ L of 20 ⁇ M CFSE was added for labeling, and incubated at 37°C for 30 min in the dark.
  • Experimental group Take 100 ⁇ L of the natural killer cell suspension from step 3.1 and 100 ⁇ L of CFSE - labeled K562 cells from step 3.2 and add them to the sample wells of a 96-well plate. Incubate the co-culture for 4 hours in an incubator. After the co-culture, 2 ⁇ L of dead cell dye PI was added and incubated at room temperature for 10 min to obtain a co-culture sample.
  • Control group Add only 100 ⁇ L of the CFSE-labeled K562 cells in step 3.2 and 100 ⁇ L of medium to the sample wells of the 96-well plate as the control group target cells. Add only 100 ⁇ L of natural killer cells from step 3.1 and 100 ⁇ L of medium to the sample wells of a 96-well plate as a control group of natural killer cells.
  • the target cells of the control group and the natural killer cells of the control group were cultured and stained to obtain a target cell sample of the control group and a natural killer cell sample of the control group.
  • Experimental group Take 20 ⁇ L of co-culture sample and add it to the hemocytometer, place it on the sample stage of the detection instrument, and use the brightfield channel, FL1 channel (matching fluorescent dye CFSE), and FL2 channel (matching fluorescent dye) in the fixed area of the co-culture sample.
  • PI for microscopic imaging, resulting in a bright-field microscopic image and two fluorescence microscopic images.
  • the information and order of the two fluorescence channels are as follows:
  • FL1 Ex 480nm, Em 535nm
  • FL2 Ex 525nm, Em 600LP.
  • the FL1 channel excites and collects CFSE fluorescence
  • the FL2 channel excites and collects PI fluorescence.
  • Control group For the hemocytometer with 20 ⁇ L of the target cell sample of the control group and 20 ⁇ L of the natural killer cell sample of the control group, respectively, the microscopic imaging was carried out according to the method of the experimental group, and the bright-field microscopic image and fluorescence display of the control group were obtained. Micro image.
  • the fluorescent microscopic images in bright field, FL1 channel and FL2 channel were subjected to image overlay synthesis analysis.
  • the target of image recognition in brightfield channel is total cells (including target cells and natural killer cells);
  • the target of image recognition in FL1 channel is target cells (including live target cells and dead target cells);
  • the target of image recognition is total dead cells (including dead target cells and dead natural killer cells).
  • the detection system marks the same location of the cell:
  • Control group Referring to the detection method of the experimental group, image overlay synthesis analysis was performed on the bright-field microscopic images and fluorescence microscopic images of the control group to obtain the target cell death rate of the control group and the natural killer cell death rate of the control group, so as to calculate the cell death rate.
  • image overlay synthesis analysis was performed on the bright-field microscopic images and fluorescence microscopic images of the control group to obtain the target cell death rate of the control group and the natural killer cell death rate of the control group, so as to calculate the cell death rate.
  • the specific killing rate and the natural killer cell self-injuring rate refer to step 2.5 of Example 2 for the formula.
  • the target cell death rate is compared with the death rate threshold.
  • the death rate threshold is set according to different situations. In this embodiment, the upper limit of the death rate threshold is set to 40%, and the lower limit is set to 20%.
  • Tumor cells (K562 cell line) were collected and prepared into a suspension with a cell concentration of 1 ⁇ 10 5 cells/mL, and then 100 ⁇ L of K562 cells were added to a 96-well plate, and the minimum diameter of the K562 cells was 9 ⁇ m.
  • Experimental group Take 100 ⁇ L of the natural killer cell suspension in step 4.1 and add it to a 96-well plate with K562 cells, set the effect-target ratio to 10:1, and co-culture for 4 hours in a 37-degree 5% CO 2 incubator.
  • Control group only 100 ⁇ L of medium was added to 100 ⁇ L of target cells (as the target cell control group), and only 100 ⁇ L of medium was added to 100 ⁇ L of natural killer cells (as the natural killer cell control group), and the experiments were carried out under the same conditions as the experimental group. Cultivation, at the time of detection, also used the same detection and analysis methods as the experimental group.
  • step 4.5 After mixing the cell suspension in step 4.4 with the 0.2% trypan blue solution at a volume ratio of 1:1, pipette 20 ⁇ L into the counting plate and place it on the detection instrument (the instrument used in this example is the Countstar automatic cell. counter) on the sample stage.
  • the detection instrument the instrument used in this example is the Countstar automatic cell. counter
  • the detection system of cell killing efficacy and/or immune activity performs image overlay synthesis analysis on the image, and then counts the dead and living cells of different diameters in the image:
  • the unstained cells with a diameter greater than or equal to 9 ⁇ m are viable target cells, and the number of viable target cells is counted as a;
  • the stained target cells with a diameter of 9 ⁇ m or more are dead, and the number of dead target cells is counted as b;
  • the unstained cells with a diameter of less than 9 ⁇ m are live effector cells, and the number of live effector cells is counted as c;
  • Dyed effector cells with a diameter of less than 9 ⁇ m were counted, and the number of dead effector cells was counted as d;
  • effector cell control group those with a diameter of less than 9 ⁇ m and stained are dead effector cells, and the number of dead effector cells is counted as M.
  • the target cell death rate can be calculated by the following formula:
  • the specific killing rate of target cells and the self-injury rate of effector cells can also be calculated:
  • Target cell specific killing rate (%) target cell death rate in experimental group - target cell death rate in control group;
  • the possible beneficial effects of the embodiments of the present specification include, but are not limited to: (1)
  • the detection methods of some embodiments of the present specification can respectively use three-stained, double-stained, and single-stained samples collected from the co-culture of target cells and effector cells.
  • the micro-image is subjected to image overlay synthesis analysis to accurately and efficiently distinguish live target cells, dead target cells, live effector cells and dead effector cells, and the cell killing rate can be calculated based on the number of cells with corresponding attributes according to the actual detection requirements.
  • the detection methods provided in some embodiments of this specification can simultaneously obtain direct-reading image information and data processing results of cell samples to be tested, compared with flow cytometry Compared with the detection results provided by other commonly used detection methods, the obtained results are more intuitive. Cluster analysis and high-content analysis can be simultaneously performed on one instrument, and multiple data such as cell death rate, cell self-injury rate, and cell-specific killing rate can be obtained. , reducing the detection steps and improving the detection efficiency; and the detection method is simple and has a wide range of applications; (3) the detection methods of some embodiments of this specification can respectively use triple staining, double staining, single staining of target cells and effector cells.
  • Image fusion analysis is performed on the microscopic images collected from the co-culture samples. Based on the shape and color characteristics of the target in the target image block contained in the fusion image, the cell properties of the target and the cell parameters associated with the cell properties can be quickly obtained, reducing the need for The processing flow improves the efficiency of detection and analysis; (4)
  • the detection methods provided in some embodiments of this specification can detect the image identifiable features of the co-culture samples in various combinations, and the detection is applicable to a wide range.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Immunology (AREA)
  • Biomedical Technology (AREA)
  • Chemical & Material Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Hematology (AREA)
  • Urology & Nephrology (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biochemistry (AREA)
  • Pathology (AREA)
  • Analytical Chemistry (AREA)
  • Cell Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Toxicology (AREA)
  • Biotechnology (AREA)
  • Microbiology (AREA)
  • Medicinal Chemistry (AREA)
  • Food Science & Technology (AREA)
  • Tropical Medicine & Parasitology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Medical Informatics (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Dispersion Chemistry (AREA)
  • Signal Processing (AREA)
  • Investigating Or Analysing Biological Materials (AREA)
  • Apparatus Associated With Microorganisms And Enzymes (AREA)
  • Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)

Abstract

本说明书实施例提供一种细胞杀伤效力和/或免疫活性的检测方法、系统及其应用。所述方法包括获取共培养样品的固定区域的多个显微图像，其中，所述共培养样品为靶细胞与效应细胞共培养得到的细胞样品，所述固定区域包含多个目标物，多个所述目标物为包含多种不同属性细胞的细胞组，每个所述目标物均带有图像可识别特征；对多个所述显微图像进行图像叠加合成分析或图像融合分析，获取多个所述目标物的所述细胞属性，并统计关联所述细胞属性的细胞参数；基于所述细胞参数，评价所述效应细胞的杀伤效力和/或免疫活性。

Description

细胞杀伤效力和/或免疫活性的检测方法、系统及其应用
交叉引用
本申请请求2020年10月28日提交的中国申请号CN202011173496.4、CN202011173508.3、CN202011176203.8、CN202011173498.3、CN202011173457.4的优先权，请求2021年1月29日提交的中国申请号CN202110127301.0的优先权，以及请求2021年5月28日提交的中国申请号CN202110595121.5的优先权，全部内容通过引用并入本文。
技术领域
本说明书涉及图像处理领域，特别涉及细胞杀伤效力和/或免疫活性检测的方法、系统及其应用。
背景技术
细胞杀伤效力的检测对于免疫细胞治疗产品的质量控制具有重要的意义。由于免疫细胞治疗产品具有起始个体差异大、制备工艺规模化程度低、制剂多为活细胞产品、作用机制不是很明确等，使得这类产品的一致性差、批量有限、效期短、同一类产品的可比性差等特点，所以免疫细胞治疗产品的质控研究较为复杂，其中杀伤效力质量控制便是难点之一。
常用的细胞杀伤检测方法，如镉51释放实验、乳酸脱氢酶(LDH)释放法、BATDA法、CAM法、CytoTox-Glo法、PKH法、流式细胞术(Flow Cytometry，FCM)等，均存在多方面的应用局限，无法兼顾直观性、准确性及高效性。因此，开发一种直观、准确、高效的细胞杀伤效力和/或免疫活性的检测方法，在研究、制备免疫细胞治疗产品等方面有积极意义。
发明内容
本说明书实施例之一提供细胞杀伤效力和/或免疫活性的检测方法，其特征在于，包括：获取共培养样品的固定区域的多个显微图像，其中，所述共培养样品为靶细胞与效应细胞共培养得到的细胞样品，所述固定区域包含多个目标物，多个所述目标物为包含多种不同属性细胞的细胞组，每个所述目标物均带有图像可识别特征，每个所述目标物的细胞属性通过该所述目标物的所述图像可识别特征在多个所述显微图像上显示的特征信息的集合来表征；对多个所述显微图像进行图像叠加合成分析或图像融合分析，获取多个所述目标物的所述细胞属性，并统计关联所述细胞属性的细胞参数；基于所述细胞参数，评价所述效应细胞的杀伤效力和/或免疫活性。
本说明书实施例之一提供一种细胞杀伤效力和/或免疫活性的检测系统，其特征在于，所述检测系统包括如下模块：显微成像模块，用于获取共培养样品的固定区域的多个显微图像，其中，所述共培养样品为靶细胞与效应细胞共培养得到的细胞样品，所述共培养样品的所述固定区域包含多个目标物，多个所述目标物为包含多种不同属性细胞的细胞组，每个所述目标物均带有图像可识别特征，每个所述目标物的细胞属性通过该所述目标物的所述图像可识别特征在多个所述显微图像上显示的特征信息的集合来表征；图像分析模块，用于基于多个所述显微图像进行图像叠加合成分析或图像融合分析，获取多个所述目标物的所述细胞属性，并统计关联所述细胞属性的细胞参数；评价模块，用于基于所述细胞参数评价所述效应细胞的杀伤效力和/或免疫活性。
本说明书实施例之一提供一种细胞杀伤效力和/或免疫活性的检测装置，包括处理器，所述处理器用于执行细胞杀伤效力和/或免疫活性的检测方法。
本说明书实施例之一提供一种计算机可读存储介质，所述存储介质存储计算机指令，当计算机读取存储介质中的计算机指令后，计算机执行细胞杀伤效力和/或免疫活性的检测方法。
本说明书实施例之一提供细胞杀伤效力和/或免疫活性的检测方法或检测系统在检测细胞杀伤力、效应细胞免疫活性、制备免疫制品、免疫制品质量控制或个体免疫功能评价中的应用。
附图说明
本说明书将以示例性实施例的方式进一步说明，这些示例性实施例将通过附图进行详细描述。这些实施例并非限制性的，在这些实施例中，相同的编号表示相同的结构，其中：
图1是根据本说明书一些实施例所示的细胞杀伤效力和/或免疫活性的检测系统的应用场景示意图；
图2是根据本说明书一些实施例所示的细胞杀伤效力和/或免疫活性的检测方法的示例性流程图；
图3是根据本说明书一些实施例所示的对多个显微图像进行图像叠加合成分析的示例性流程图；
图4是根据本说明书一些实施例所示的提取显微图像中目标物轮廓的示例性流程图；
图5是根据本说明书一些实施例所示的获取融合图像的示例性流程图；
图6是根据本说明书一些实施例所示的分析融合图像的示例性流程图；
图7是根据本说明书一些实施例所示的图像识别模型的示例性示意图；
图8是本说明书实施例1中FL1通道采集的荧光显微图像；
图9是本说明书实施例1中FL2通道采集的荧光显微图像；
图10是本说明书实施例1中FL3通道采集的荧光显微图像；
图11是本说明书实施例1中FL1、FL2、FL3通道采集的荧光显微图像的叠加图像。
具体实施方式
为了更清楚地说明本说明书实施例的技术方案，下面将对实施例描述中所需要使用的附图作简单的介绍。显而易见地，下面描述中的附图仅仅是本说明书的一些示例或实施例，对于本领域的普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图将本说明书应用于其它类似情景。除非从语言环境中显而易见或另做说明，图中相同标号代表相同结构或操作。
应当理解，本文使用的“系统”、“装置”、“单元”和/或“模块”是用于区分不同级别的不同组件、元件、部件、部分或装配的一种方法。然而，如果其他词语可实现相同的目的，则可通过其他表达来替换所述词语。
如本说明书和权利要求书中所示，除非上下文明确提示例外情形，“一”、“一个”、“一种”和/或“该”等词并非特指单数，也可包括复数。一般说来，术语“包括”与“包含”仅提示包括已明确标识的步骤和元素，而这些步骤和元素不构成一个排它性的罗列，方法或者设备也可能包含其它的步骤或元素。
本说明书中使用了流程图用来说明根据本说明书的实施例的系统所执行的操作。应当理解的是，前面或后面操作不一定按照顺序来精确地执行。相反，可以按照倒序或同时处理各个步骤。同时，也可以将其他操作添加到这些过程中，或从这些过程移除某一步或数步操作。
图1是根据本说明书一些实施例所示的细胞杀伤效力和/或免疫活性的检测系统的应用场景示意图。
如图1所示，检测系统100可以包括服务器110、网络120、存储设备130和图像采集装置140。
服务器110可以用于管理资源以及处理来自检测系统100至少一个组件或外部数据源(例如，云数据中心)的数据和/或信息。例如，对多个显微图像(明场显微图像和至少一个荧光显微图像)图像叠加合成分析。又例如，对多个显微图像进行图像融合分析。在处理过程中，服务器110可以从存储设备140获取数据(如多个显微图像中的一个或多个)或将数据(例如，目标物的细胞属性、细胞参数)保存到存储设备140，也可以通过网络120从图像采集装置140等其他来源读取数据(例如，明场显微图像和/或至少一个荧光显微图像)。
在一些实施例中，服务器110可以是单一服务器或服务器组。该服务器组可以是集中式或分布式的(例如，服务器110可以是分布式系统)，可以是专用的也可以由其他设备或 系统同时提供服务。在一些实施例中，服务器110可以是区域的或者远程的。在一些实施例中，服务器110可以在云平台上实施，或者以虚拟方式提供。仅作为示例，所述云平台可以包括私有云、公共云、混合云、社区云、分布云、内部云、多层云等或其任意组合。
在一些实施例中，服务器110可包含处理设备。处理设备可以处理从其他设备或系统组成部分中获得的数据和/或信息。处理器可以基于这些数据、信息和/或处理结果执行程序指令，以执行一个或多个本申请中描述的功能。在一些实施例中，处理设备可以包含一个或多个子处理设备(例如，单核处理设备或多核多芯处理设备)。仅作为示例，处理设备可以包括中央处理器(CPU)、专用集成电路(ASIC)、专用指令处理器(ASIP)、图形处理器(GPU)、物理处理器(PPU)、数字信号处理器(DSP)、现场可编程门阵列(FPGA)、可编辑逻辑电路(PLD)、控制器、微控制器单元、精简指令集电脑(RISC)、微处理器等或以上任意组合。
网络120可以连接检测系统100的各组成部分和/或连接系统与外部资源部分。网络120使得各组成部分之间，以及与系统之外其他部分之间可以进行通讯，促进数据和/或信息的交换。在一些实施例中，网络120可以是有线网络或无线网络中的任意一种或多种。例如，网络120可以包括电缆网络、光纤网络、电信网络、互联网、局域网络(LAN)、广域网络(WAN)、无线局域网络(WLAN)、城域网(MAN)、公共交换电话网络(PSTN)、蓝牙网络、紫蜂网络(ZigBee)、近场通信(NFC)、设备内总线、设备内线路、线缆连接等或其任意组合。各部分之间的网络连接可以是采用上述一种方式，也可以是采取多种方式。在一些实施例中，网络可以是点对点的、共享的、中心式的等各种拓扑结构或者多种拓扑结构的组合。在一些实施例中，网络120可以包括一个或以上网络接入点。例如，网络120可以包括有线或无线网络接入点，例如基站和/或网络交换点，通过这些进出点系统100的一个或多个组件可连接到网络120上以交换数据和/或信息。
存储设备130可以用于存储数据(如明场显微图像和至少一个荧光显微图像)和/或指令。存储设备130在单个中央服务器、通过通信链路连接的多个服务器或多个个人设备中实现。在一些实施例中，存储设备130可包括大容量存储器、可移动存储器、挥发性读写存储器(例如，随机存取存储器RAM)、只读存储器(ROM)等或以上任意组合。示例性的，大容量储存器可以包括磁盘、光盘、固态磁盘等。在一些实施例中，存储设备130可在云平台上实现。
图像采集装置140可以用于获取共培养样品的固定区域的多个显微图像(明场显微图像和至少一个荧光显微图像)。在一些实施例中，获取不同显微图像(明场显微图像和至少一个荧光显微图像)的图像采集装置可以相同。例如，图像采集装置140可以是金相显微镜。在一些实施例中，获取不同显微图像的图像采集装置可以不同。例如，图像采集装置140可以包括用于获取明场显微图像的明视场显微镜和用于获取至少一幅荧光显微图像的荧光显微镜。
在一些实施例中，检测系统100还可以包括一个终端设备(未示出)。终端设备可以包括输入设备(如键盘、鼠标)和/或输出设备(如显示屏、扬声器)。用户可以通过终端设备与处理设备110、图像采集装置140等设备进行互动。例如，用户可以通过终端设备查看图像采集装置140获取的显微图像。再例如，用户可以通过终端设备直接观察处理设备处理的图像分析结果。
在一些实施例中，检测系统100可以包括显微成像模块、图像分析模块和评价模块。
显微成像模块可以用于获取细胞样品的显微图像。显微图像可以包括明场显微图像和至少一个荧光显微图像。在一些实施例中，细胞样品可以包括共培养样品和对照组样品。共培养样品为靶细胞与效应细胞共培养得到的细胞样品。对照组样品为靶细胞和/或效应细胞单独培养得到的细胞样品。
关于显微成像模块的更多描述可以参见步骤210，在此不再赘述。
图像分析模块可以用于基于多个显微图像进行图像叠加合成分析，获取多个目标物的细胞属性，以及关联多个目标物的细胞属性的细胞参数。进一步的，在一些实施例中，图像分析模块可以提取每个显微图像中的多个目标物区域及对应的轮廓信息。目标物区域为包含单个目标物的图像区域，单个目标物具有封闭的轮廓。在一些实施例中，图像分析模块可以基于多个显微图像的目标物区域及对应的轮廓信息进行目标物重合判定，获取重合判定结果。在一些实施例中，图像分析模块可以基于重合判定结果，获取对应目标物的细胞属性。
在一些实施例中，图像分析模块可以基于细胞属性对多个目标物进行分类计数及统计，获取细胞参数。
图像分析模块还可以用于基于多个显微图像进行图像融合分析，获取多个目标物的细胞属性，以及关联多个目标物的细胞属性的细胞参数。
在一些实施例中，图像分析模块可以基于多个显微图像，获取融合图像。进一步的，在一些实施例中，图像分析模块可以提取每个所述显微图像的特征点。在一些实施例中，图像分析模块可以基于多个显微图像的对应特征点，配准多个显微图像。在一些实施例中，图像分析模块可以基于透明度和/或色度融合配准后的多个显微图像，获取融合图像。
在一些实施例中，图像分析模块可以分析融合图像，获取多个目标物的细胞属性，以及关联多个目标物的细胞属性的细胞参数。进一步的，在一些实施例中，图像分析模块可以基于融合图像，获取多个目标物图像块。目标物图像块为包含单个目标物的图像块。在一些实施例中，图像分析模块可以提取多个目标物图像块的颜色特征和形状特征。在一些实施例中，图像分析模块可以基于多个目标物图像块的颜色特征和形状特征，获取多个目标物的细胞属性，以及关联多个目标物的细胞属性的细胞参数。
在一些实施例中，图像分析模块还可以基于图像识别模型处理融合图像，获取多个目标物的细胞属性，以及关联多个目标物的细胞属性的细胞参数。在一些实施例中，图像识别模型可为机器学习模型。
评价模块可以用于基于细胞参数评价效应细胞的杀伤效力和/或免疫活性。在一些实施例中，评价模块可以基于细胞参数中的靶细胞死亡率、细胞特异杀伤率和效应细胞自伤率中的一种或多种，评价效应细胞的杀伤效力和/或免疫活性。
在一些实施例中，检测系统100还可以包括样品台模块和样品自动更换模块。
样品台模块可以用于承载细胞培养板。其中，细胞培养板有多个样品孔，用于承载至少两种效靶比的共培养样品，显微成像模块对细胞培养板上各孔中的共培养样品的固定区域分别进行成像，以获取共培养样品的固定区域的多个显微图像。
样品自动更换模块可以用于更换细胞培养板。其中，显微成像模块对更换后的细胞培养板上的共培养样品的固定区域分别进行成像，得到更换后的细胞培养板上固定区域的多个显微图像。
需要注意的是，以上对于检测系统及其模块的描述，仅为描述方便，并不能把本说明书限制在所举实施例范围之内。可以理解，对于本领域的技术人员来说，在了解该系统的原理后，可能在不背离这一原理的情况下，对各个模块进行任意组合，或者构成子系统与其他模块连接。在一些实施例中，图1中披露的显微成像模块与图像分析模块、评价模块可以是一个系统中的不同模块，也可以是一个模块实现上述的两个或两个以上模块的功能。例如，显微成像模块和荧光显微成像模块可以为同一个模块，既可以获取共培养样品的明场显微图像，也可以获取共培养样品的荧光显微图像。例如，各个模块可以共用一个存储模块，各个模块也可以分别具有各自的存储模块。诸如此类的变形，均在本说明书的保护范围之内。
图2是根据本说明书一些实施例所示的细胞杀伤效力和/或免疫活性的检测方法的示例性流程图。如图2所示，流程200包括步骤210至步骤230。
步骤210为获取显微图像的步骤。在步骤210中，获取共培养样品的固定区域的多个显微图像。在一些实施例中，显微成像模块可以执行步骤210。在一些实施例中，显微成像模 块可以通过图像采集装置140获取多个显微图像。在一些实施例中，显微成像模块可以通过网络120从存储设备140获取预采集的多个显微图像。
共培养样品为靶细胞与效应细胞共培养得到的细胞样品。共培养样品的固定区域包含多个目标物，多个所述目标物为包含多种不同属性细胞的细胞组，每个目标物均带有图像可识别特征，每个目标物的细胞属性通过该目标物的图像可识别特征在多个显微图像上显示的特征信息的集合来表征。在一些实施例中，固定区域可以为共培养样品的包含所有目标物的全部图像采集区域。在一些实施例中，固定区域可以为共培养样品的包含部分目标物的部分图像采集区域。
靶细胞是指免疫细胞相对应的各种肿瘤细胞或病毒感染的细胞。在一些实施例中，靶细胞为病毒感染的细胞和/或肿瘤细胞。可作为靶细胞的肿瘤细胞和病毒感染的细胞包括但不限于K562细胞、Daudi细胞、Jurkat细胞、MCF-7细胞、A549细胞、HepG2细胞等。
效应细胞是指在免疫应答中参与清除异物抗原和行使效应功能的免疫细胞或工程化细胞。在一些实施例中，效应细胞为免疫细胞和/或工程化细胞。可作为效应细胞的免疫细胞和工程化细胞包括但不限于PBMC细胞、NK细胞、T细胞、CTL细胞、LAK细胞、CIK细胞、TIL细胞、DC细胞、CAR-T细胞、CAR-NK细胞、NK92MI-CD16a细胞等。
细胞属性可以为细胞的一系列生命现象(如生长、发育、增殖、分化、遗传、代谢、应激、运动、衰老和死亡)中一种或多种的体现。在一些实施例中，细胞属性可以包括细胞类型和细胞存活状态。例如，细胞类型可以包括靶细胞和效应细胞；细胞存活状态可以包括活细胞和死细胞。其中，活细胞是可进行新陈代谢、繁殖及复制的细胞，主要表现为细胞膜完整和具有选择透过性。死细胞是无法正常发挥生物学功能、进行新陈代谢、繁殖及复制的细胞，主要表现为细胞膜破损和选择透过性的丧失。根据细胞不同死亡方式，可包括由细胞凋亡和细胞坏死、细胞铁死亡、细胞焦亡、细胞自噬等细胞死亡过程而产生的死细胞。又例如，细胞类型可以包括靶细胞、效应细胞、细胞碎片和杂质；细胞存活状态可以包括活细胞、凋亡细胞和坏死细胞。其中，凋亡细胞是为维持内环境稳定，由基因控制的自主性、有序性的死亡的细胞。坏死细胞是指经过极端的物理、化学因素或严重的病理性刺激引起的损伤和死亡的细胞。在一些实施例中，固定区域内的多个目标物为活靶细胞、死靶细胞、活效应细胞和死效应细胞的细胞组。
不同细胞属性的目标物带有不同的图像可识别特征，以使图像可识别特征表征目标物的细胞属性。目标物的图像可识别特征可以有多种不同的具体类型，采用不同的显微图像组合对不同类型的图像可识别特征进行图像分析。
在一些实施例中，图像可识别特征包括荧光标记特征。当确定固定区域内每个目标物的细胞属性分别为活靶细胞、死靶细胞、活效应细胞和死效应细胞中的一种时，可目标物的细胞属性可采用不同的荧光标记组合来表征。
在一些实施例中，目标物的细胞属性通过三种荧光标记的组合来表征。具体的，共培养样品通过使用三种荧光标记进行标记获得，共培养样品的获取步骤包括：基于携带预置荧光标记的靶细胞和无荧光标记的效应细胞进行共培养，获得共培养物；共培养预定时间后分别使用总细胞荧光标记和死细胞荧光标记标记共培养物，获得共培养样品。其中，表征目标物的细胞属性的图像可识别特征具体为：共培养样品的固定区域内，携带预置荧光标记和总细胞荧光标记的目标物为活靶细胞，携带预置荧光标记、总细胞荧光标记和死细胞荧光标记的目标物为死靶细胞，仅携带总细胞荧光标记的目标物为活效应细胞，携带总细胞荧光标记和死细胞荧光标记的目标物为死效应细胞。
预置荧光标记用于对共培养前的靶细胞进行标记。在一些实施例中，预置荧光标记可以为荧光蛋白或细胞染料。在一些实施例中，优选的，可作为预置荧光标记的荧光蛋白为绿色荧光蛋白(GFP)或红色荧光蛋白(RFP)。在一些实施例中，优选的，可作为预置荧光标记的细胞染料为羧基荧光素二醋酸盐琥珀酰亚胺酯(CFSE)或钙黄绿素AM(Calcein-AM)。需要注 意的是，对于使用容易产生背景荧光的细胞染料进行标记的靶细胞，在与效应细胞共培养前，靶细胞需要进行洗涤的操作。
总细胞荧光标记可以对共培养样品中的所有细胞进行标记。在一些实施例中，总细胞荧光标记可以为核染料。可作为总细胞荧光标记的核染料包括但不限于Hoechst33342、DAPI等。
死细胞荧光标记可以仅对共培养样品中的死细胞进行标记。死细胞荧光标记可以为任意死细胞标记染料。可作为死细胞荧光标记的死细胞标记染料包括但不限于膜联蛋白-V(Annexin-V)、绿菁SYTOX(SYTOX Green)、溴化丙啶(PI)和7-氨基放线菌素D(7-AAD)等。
在一些实施例中，目标物的细胞属性通过两种荧光标记的组合来表征。具体的，共培养样品通过使用两种荧光标记进行标记获得，共培养样品的获取步骤包括：基于携带预置荧光标记的靶细胞和无荧光标记的效应细胞进行共培养，获得共培养物；共培养预定时间后使用死细胞荧光标记标记所述共培养物，获得所述共培养样品。其中，表征目标物的细胞属性的图像可识别特征具体为：共培养样品的固定区域内，仅携带预置荧光标记的目标物为活靶细胞，携带预置荧光标记和死细胞荧光标记的目标物为死靶细胞，无荧光标记的目标物为活效应细胞，仅携带死细胞荧光标记的目标物为死效应细胞。
在一些实施例中，图像可识别特征包括荧光标记特征和细胞直径特征。当确定固定区域内每个目标物的细胞属性分别为活靶细胞、死靶细胞、活效应细胞和死效应细胞中的一种时，目标物的细胞属性可采用荧光标记特征与细胞直径特征的不同组合来表征。
在一些实施例中，目标物的细胞属性通过一种荧光标记与效应/靶细胞不同细胞直径的组合来表征。具体的，共培养样品通过使用一种荧光标记进行标记获得，共培养样品的获取步骤包括：基于无荧光标记的靶细胞和无荧光标记的效应细胞进行共培养，获得共培养物；共培养预定时间后使用死细胞荧光标记标记所述共培养物，获得所述共培养样品。其中，表征目标物的细胞属性的图像可识别特征具体为：共培养样品的固定区域内，无荧光标记且直径大于等于靶细胞最小直径的目标物为活靶细胞，携带死细胞荧光标记且直径大于等于靶细胞最小直径的目标物为死靶细胞，无荧光标记且直径小于效应细胞最大直径的目标物为活效应细胞，携带死细胞荧光标记且直径小于效应细胞最大直径的目标物为死效应细胞。
在一些实施例中，步骤210进一步包括：
步骤211，获取共培养样品的固定区域的明场显微图像；
步骤212，确定表征共培养样品中多个目标物的细胞属性的图像可识别特征；
步骤213，获取共培养样品的固定区域的至少一个荧光显微图像，至少一个荧光显微图像的成像参数基于多个目标物的图像可识别特征确定。
明场显微图像是图像采集装置140以明场光源照射细胞样品采集的图像。明场显微图像中的视野背景明亮，而细胞样品中的细胞边缘是黑暗的。
荧光显微图像是图像采集装置140以激发光光源照射细胞样品，使之发出荧光后采集的图像。荧光显微图像可反映细胞样品中细胞的形状及其所在位置。
在一些实施例中，显微图像的格式可以包括Joint Photographic Experts Group(JPEG)图像格式、Tagged Image File Format(TIFF)图像格式、Graphics Interchange Format(GIF)图像格式、Kodak Flash PiX(FPX)图像格式和Digital Imaging and Communications in Medicine(DICOM)图像格式等。
在一些实施例中，至少一个荧光显微图像的成像参数包括荧光通道种类和激发光波长。检测系统100进行图像分析所使用的荧光显微图像的荧光通道种类及对应激发光波长根据共培养样品中多个目标物的细胞属性的图像可识别特征确定。
共培养样品通过使用三种荧光标记进行标记获得的情况下，至少一个荧光显微图像的成像参数根据共培养样品中多个目标物的细胞属性的具体图像可识别特征确定。在一些实施例中，确定表征目标物的细胞属性的图像可识别特征为预置荧光标记、总细胞荧光标记和死 细胞荧光标记的组合，则至少一个荧光显微图像包括第一荧光显微图像、第二荧光显微图像和第三荧光显微图像。其中，采集第一荧光显微图像的荧光通道及该荧光通道的激发光波长匹配预置荧光标记，采集第二荧光显微图像的荧光通道及该荧光通道的激发光波长匹配总细胞荧光标记，采集第三荧光显微图像的荧光通道及该荧光通道的激发光波长匹配死细胞荧光标记。
共培养样品通过使用两种荧光标记进行标记获得的情况下，至少一个荧光显微图像的成像参数根据共培养样品中多个目标物的细胞属性的具体图像可识别特征确定。在一些实施例中，确定表征目标物的细胞属性的图像可识别特征为预置荧光标记和死细胞荧光标记的组合，则至少一个荧光显微图像包括第一荧光显微图像和第三荧光显微图像。其中，采集第一荧光显微图像的荧光通道及该荧光通道的激发光波长匹配预置荧光标记，采集第三荧光显微图像的荧光通道及该荧光通道的激发光波长匹配死细胞荧光标记。
共培养样品通过使用一种荧光标记进行标记获得的情况下，至少一个荧光显微图像的成像参数根据共培养样品中多个目标物的细胞属性的具体图像可识别特征确定。在一些实施例中，确定表征目标物的细胞属性的图像可识别特征为死细胞荧光标记与不同类型细胞直径的组合，则至少一个荧光显微图像为第三荧光显微图像。其中，采集第三荧光显微图像的荧光通道及该荧光通道的激发光波长匹配死细胞荧光标记。
为获取更多与评价细胞杀伤效力和/或免疫活性相关的细胞参数，在一些实施例中，步骤210还包括获取对照组效应细胞样品的固定区域的多个对照组显微图像的步骤。其中，对照组效应细胞样品为效应细胞单独培养得到的细胞样品，对照组靶细胞样品的固定区域包含多个带图像可识别特征的第一对照目标物。表征第二对照目标物的细胞属性的图像可识别特征与表征目标物的细胞属性的图像可识别特征保持一致。
为获取更多与评价细胞杀伤效力和/或免疫活性相关的细胞参数，在一些实施例中，步骤210还包括获取对照组靶细胞样品的固定区域的多个对照组显微图像的步骤。其中，对照组靶细胞样品为靶细胞单独培养得到的细胞样品，对照组靶细胞样品的固定区域包含多个带图像可识别特征的第二对照目标物。表征第二对照目标物的细胞属性的图像可识别特征与表征目标物的细胞属性的图像可识别特征保持一致。
步骤220为进行图像分析的步骤。在步骤220中，对多个显微图像进行图像叠加合成分析或图像融合分析，获取多个目标物的细胞属性，并统计关联细胞属性的细胞参数。在一些实施例中，图像分析模块可以执行步骤220。
在一些实施例中，对多个显微图像进行图像叠加合成分析，获取多个目标物的细胞属性，并统计关联细胞属性的细胞参数。关于对多个显微图像进行图像叠加合成分析的具体描述可以参见图3及其相关描述，此处不再赘述。
在一些实施例中，对多个显微图像进行图像融合分析，获取多个目标物的细胞属性，并统计关联细胞属性的细胞参数。在一些实施例中，对多个显微图像进行图像融合分析，获取多个目标物的细胞属性，并统计关联细胞属性的细胞参数进一步包括：
基于多个显微图像，获取融合图像；
分析融合图像，获取多个目标物的细胞属性，并统计关联细胞属性的细胞参数。
关于获取融合图像的具体描述可以参见图5及其相关描述，此处不再赘述。关于分析融合图像的具体描述可以参见图6及其相关描述，此处不再赘述。
细胞参数是可用于评价细胞杀伤效力和/或免疫活性的统计数据。在一些实施例中，细胞参数可包括关联共培养样品的多个目标物的细胞属性的第一细胞参数。第一细胞参数可通过基于细胞属性对多个目标物进行分类计数及统计而获取。在一些实施例中，第一细胞参数可以包括靶细胞和效应细胞总数、靶细胞总数、活靶细胞总数、死靶细胞总数、靶细胞死亡率、效应细胞总数、活效应细胞总数、死效应细胞总数和效应细胞死亡率中的一种或多种。具体的，靶细胞死亡率和效应细胞死亡率采用如下公式计算：
靶细胞死亡率＝死靶细胞总数/靶细胞总数×100％；
效应细胞死亡率＝死效应细胞总数/效应细胞总数×100％。
为获取更多与评价细胞杀伤效力和/或免疫活性相关的细胞参数，在一些实施例中，步骤220还进一步包括基于对照组靶细胞样品的多个对照组显微图像进行叠加合成分析，获取多个第一对照目标物的细胞属性，并统计关联细胞属性的细胞参数的步骤。
在一些实施例中，细胞参数还可包括关联对照组靶细胞样品的多个第一对照目标物的细胞属性的第二细胞参数。第二细胞参数可通过基于细胞属性对多个第一对照目标物进行分类计数及统计而获取。在一些实施例中，第二细胞参数可以包括对照组靶细胞总数、对照组活靶细胞总数、对照组死靶细胞总数、对照组靶细胞死亡率和细胞特异杀伤率中的一种或多种。具体的，对照组靶细胞死亡率和细胞特异杀伤率采用如下公式计算：
对照组靶细胞死亡率＝对照组死靶细胞总数/对照组靶细胞总数×100％；
细胞特异杀伤率＝靶细胞死亡率-对照组靶细胞死亡率。
为获取更多与评价细胞杀伤效力和/或免疫活性相关的细胞参数，在一些实施例中，步骤220还进一步包括基于对照组效应细胞样品的多个对照组显微图像进行叠加合成分析，获取多个第二对照目标物的细胞属性，并统计关联细胞属性的细胞参数的步骤。
在一些实施例中，细胞参数还可包括关联对照组效应细胞样品的多个第二对照目标物的细胞属性的第三细胞参数。第三细胞参数可通过基于细胞属性对多个第二对照目标物进行分类计数及统计而获取。在一些实施例中，第三细胞参数可以包括对照组效应细胞总数、对照组活效应细胞总数、对照组死效应细胞总数、对照组效应细胞死亡率和效应细胞自伤率中的一种或多种。具体的，对照组效应细胞死亡率和效应细胞自伤率采用如下公式计算：
对照组效应细胞死亡率＝对照组死效应细胞总数/对照组效应细胞总数×100％；
效应细胞自伤率＝效应细胞死亡率-对照组效应细胞死亡率。
步骤230为评价细胞杀伤效力和/或免疫活性的步骤。在步骤230中，基于细胞参数，评价效应细胞的杀伤效力和/或免疫活性。在一些实施例中，评价模块可以执行步骤230。
在一些实施例中，细胞参数中的靶细胞死亡率、效应细胞死亡率、细胞特异杀伤率和效应细胞自伤率中的一个或多个的组合可用于表征效应细胞的杀伤效力和/或免疫活性。具体的，一个细胞参数或多个细胞参数的组合通过与参数阈值进行对比，可直观地反映效应细胞杀伤效力水平和/或免疫活性水平。例如，以选取的一个对照样品的一个细胞参数或多个细胞参数的组合为参数阈值，通过待测样品的对应细胞参数或细胞参数的组合与参数阈值的对比，可评价待测样品的效应细胞杀伤效力和/或免疫活性相对于对照样品的高低情况，对比样品按评价目的的不同来选择。又例如，以选取的包含多个对照样品的对照组的一个细胞参数或多个细胞参数的组合的总体均值的区间估计值作为参数阈值，通过待测样品的对应细胞参数或细胞参数的组合与参数阈值的对比，可评价待测样品的效应细胞杀伤效力和/或免疫活性相对于对照组的高低情况，对照组按评价目的的不同来选择。
在一些实施例中，基于细胞参数，评价效应细胞的杀伤效力和/或免疫活性可进一步包括：对比靶细胞死亡率与死亡率阈值，其中，死亡率阈值包括上限和下限，根据对比结果评价效应细胞的杀伤效力和/或免疫活性。
示例性的，在药物筛选的应用场景中，取肿瘤细胞(靶细胞)和空白对照组小鼠(定期灌胃给蒸馏水)的自然杀伤细胞(效应细胞)共培养，采用上述检测方法或检测系统检测空白对照组小鼠的共培养样品的靶细胞死亡率。通过BootStrap方法，基于空白对照组小鼠的靶细胞死亡率数据求取靶细胞死亡率的均值范围作为死亡率阈值，用以代表正常小鼠的靶细胞死亡率总体均值。取肿瘤细胞和实验组小鼠(定期灌胃给待测药物)的自然杀伤细胞进行共培养，采用上述检测方法或检测系统检测实验组小鼠的共培养样品的靶细胞死亡率。对比分析实验组小鼠的靶细胞死亡率与死亡率阈值：若实验组小鼠的靶细胞死亡率高于死亡率阈值的上限，则实验组小鼠的自然杀伤细胞杀伤效力和/或免疫活性高于正常水平，表明待测药 物可提高小鼠的自然杀伤细胞杀伤效力和/或免疫活性；若实验组小鼠的靶细胞死亡率处于死亡率阈值的上限与下限之间，则实验组小鼠的自然杀伤细胞杀伤效力和/或免疫活性处于正常水平，表明待测药物未提高小鼠的自然杀伤细胞杀伤效力和/或免疫活性；若实验组小鼠的靶细胞死亡率低于死亡率阈值的下限，则实验组小鼠的自然杀伤细胞杀伤效力和/或免疫活性低于正常水平，表明待测药物可降低小鼠的自然杀伤细胞杀伤效力和/或免疫活性。
图3是根据本说明书一些实施例所示的对多个显微图像进行图像叠加合成分析的示例性流程图。在一些实施例中，流程300可以由图像分析模块执行。如图3所示，流程300包括步骤310至步骤340。
步骤310为提取显微图像中目标物轮廓的步骤。在一些实施例中，提取每个显微图像中的多个目标物区域及对应的轮廓信息。
目标物区域是指显微图像中包含单个目标物的图像区域，目标物区域的边缘为对应目标物的外轮廓。
轮廓信息可以为表征目标物轮廓特征的相关信息。在一些实施例中，轮廓信息可以包括目标物位置信息、目标物尺寸信息和目标物荧光信息中的一种或多种。目标物位置信息包括但不限于目标物外轮廓像素点在显微图像上的坐标信息、目标物轮廓特征点(如中心和质心)坐标信息等。目标物尺寸信息包括但不限于直径信息、轮廓面积信息等。目标物荧光信息包括但不限于荧光强度信息、颜色信息等。
关于提取显微图像中目标物轮廓的具体描述可参见图4及其相关描述，在此不再赘述。
步骤320为多个显微图像进行目标物重合判定的步骤。在一些实施例中，基于多个显微图像的目标物区域及对应的所述轮廓信息进行目标物重合判定，获取重合判定结果。
在一些实施例中，目标物重合判定包括基于特征点坐标距离计算进行的初次重合判定和基于交并比计算进行的二次重合判定。进一步的，步骤320包括：
对每个显微图像的每个目标物区域，遍历其余显微图像的每个目标物区域进行目标物重合判定，获取重合判定结果；
其中，在目标物重合判定过程中：
若对比的两个目标物区域在所述初次重合判定中判定为重合，则初次重合判定的判定结果即为本轮目标物重合判定的判定结果；
若对比的两个目标物区域在初次重合判定中判定为不重合，则基于对比的两个目标物区域进行二次重合判定，二次重合判定的判定结果即为本轮目标物重合判定的判定结果。
在一些实施例中，目标物重合判定包括基于特征点坐标距离计算进行的初次重合判定和基于交并比计算进行的二次重合判定。
在一些实施例中，初次重合判定的步骤进一步包括：
基于待判定的两个目标物区域对应的轮廓信息，确定两个目标物区域的特征点及特征点坐标，其中，两个目标物区域分别位于不同的显微图像上；
计算两个目标物区域的特征点坐标距离；
对比特征点坐标距离与预设的距离阈值，判定两个目标物区域是否重合，其中，当两个目标物区域的特征点坐标距离小于等于距离阈值时，判定两个目标物区域重合，反之则判定不重合。
特征点是目标物区域在对应显微图像上的特征像素点。特征点坐标是目标物区域的特征点在对应显微图像上的像素点坐标，可基于目标物区域的轮廓信息提取。在一些实施例中，特征点可以为目标物区域的中心、质心和重心中的一种。优选的，特征点可以为目标物区域的中心。
距离阈值为初次重合判定中重合与否的判定界限。理想情况下，在同一固定区域内采集目标物的不同显微图像，则同一目标物在不同显微图像上的对应目标物区域的特征点坐标 距离可以为零。在实际情况下，由于多个显微图像包括明场显微图像和至少一个荧光显微图像，受采集各显微图像的光源(明场光源和/或激发光光源)波长、成像通道(明场通道和/或荧光通道)曝光时间不同等因素的影响，同一目标物在不同显微图像上的对应目标物区域可能产生几个像素至十几个像素的位置偏移，设置距离阈值可以为基于特征点坐标距离计算进行的初次重合判定提供容差空间。在一些实施例中，距离阈值可基于选取的图像采集装置的光源波长和/或选取的成像通道曝光时间设定。在一些替代性实施例中，距离阈值可基于用户输入设定。距离阈值的设定方式在此不做限制。
在一些实施例中，二次重合判定的步骤进一步包括：
基于待判定的两个目标物区域计算目标物区域交并比；
对比目标物区域交并比与预设的交并比阈值，判定两个目标物区域是否重合，其中，当两个目标物区域的目标物区域交并比大于等于交并比阈值时，判定两个目标物区域重合，反之则判定不重合。
目标物区域交并比是两个目标物区域之间相交的面积与并集的面积的比值，可用于评价两个目标物区域的重叠度。
交并比阈值为二次重合判定中重合与否的判定界限。在一些实施例中，交并比阈值可基于用户输入设定。
步骤330为确定目标物的细胞属性的步骤。
在一些实施例中，仅基于重合判定结果，即可确定目标物的细胞属性。具体的，上述确定目标物细胞属性的方式适用于带两种及带三种荧光标记的共培养样品的多个显微图像的叠加合成分析。
示例性的，共培养样品通过使用三种荧光标记进行标记获得的情况下，三种荧光标记分别为预置荧光标记、总细胞荧光标记、死细胞荧光标记。进行图像叠加合成分析所需的多个显微图像包括明场显微图像、匹配预置荧光标记的第一荧光显微图像、匹配总细胞荧光标记的第二荧光显微图像和匹配死细胞荧光标记的第三荧光显微图像。其目标物重合判定的重合判定结果包括每个目标物的荧光标记特征在多个显微图像上显示的特征信息的集合，即重合判定结果包括每个目标物的荧光标记特征信息的集合，重合判定结果可用于确定目标物的细胞属性。其基于重合判定结果，确定目标物的细胞属性的步骤进一步包括：
标记在明场显微图像、第一荧光显微图像和第二荧光显微图像上存在目标物区域重合，且在第三荧光显微图像上无目标物区域重合的目标物，确定该目标物的细胞属性为活靶细胞；
标记在明场显微图像、第一荧光显微图像、第二荧光显微图像和第三荧光显微图像均存在目标物区域重合的目标物，确定该目标物的细胞属性为死靶细胞；
标记在明场显微图像和第二荧光显微图像上存在目标物区域重合，且在第一荧光显微图像和第三荧光显微图像上无目标物区域重合的目标物，确定该目标物的细胞属性为活效应细胞；
标记在明场显微图像、第二荧光显微图像和第三荧光显微图像上目标物区域重合，且在第一荧光显微图像上无目标物区域重合的目标物，确定该目标物的细胞属性为死效应细胞；
标记仅在明场显微图像上存在目标物区域，且在第一荧光显微图像、第二荧光显微图像和第三荧光显微图像均无目标物区域重合的目标物，确定该目标物的细胞属性为细胞碎片或杂质。
示例性的，共培养样品通过使用两种荧光标记进行标记获得的情况下，两种荧光标记分别为预置荧光标记和死细胞荧光标记。进行图像叠加合成分析所需的多个显微图像包括明场显微图像、匹配预置荧光标记的第一荧光显微图像和匹配死细胞荧光标记的第三荧光显微图像。其目标物重合判定的重合判定结果包括每个目标物的荧光标记特征在多个显微图像上显示的特征信息的集合，即重合判定结果包括每个目标物的荧光标记特征信息的集合，重合判定结果可用于确定目标物的细胞属性。其基于重合判定结果，确定目标物的细胞属性的步 骤进一步包括：
标记在明场显微图像和第一荧光显微图像存在目标物区域重合，且在第三荧光显微图像无目标物区域重合的目标物，确定该目标物的细胞属性为活靶细胞；
标记在明场显微图像、第一荧光显微图像和第三荧光显微图像均存在目标物区域重合的目标物，确定该目标物的细胞属性为死靶细胞；
标记仅在明场显微图像上存在目标物区域，且在第一荧光显微图像和第三荧光显微图像均无目标物区域重合的目标物，确定该目标物的细胞属性为活效应细胞；
标记在明场显微图像和第三荧光显微图像上存在目标物区域重合，且在第一荧光显微图像上无目标物区域重合的目标物，确定该目标物的细胞属性为死效应细胞。
在一些实施例中，步骤330包括：
提取重合判定结果中目标物的直径信息进行细胞直径判定，获得直径判定结果；
基于重合判定结果和直径判定结果，确定对应目标物的细胞属性。
在一些实施例中，细胞直径判定进一步包括：对比目标物的直径信息与预设的靶细胞最小直径、预设效应细胞最大直径，获得直径判定结果；其中，目标物直径大于等于靶细胞最小直径，则判定该目标物为靶细胞；目标物直径小于效应细胞最小直径，则判定该目标物为效应细胞。
基于目标物区域的轮廓信息进行直径判定，获得直径判定结果进一步包括：对比目标物区域的直径与预设的靶细胞最小直径、预设效应细胞最大直径，以判定目标物区域对应的目标物为靶细胞或效应细胞，获得直径判定结果。
示例性的，共培养样品通过使用一种荧光标记进行标记获得的情况下，一种荧光标记分别为死细胞荧光标记。进行图像叠加合成分析所需的多个显微图像包括明场显微图像和匹配死细胞荧光标记的第三荧光显微图像。其目标物重合判定的重合判定结果包括每个目标物的荧光标记特征和细胞直径特征在多个显微图像上显示的特征信息的集合，即重合判定结果包括每个目标物的荧光标记特征信息和细胞直径特征信息的集合，重合判定结果可用于确定目标物的细胞属性。其确定目标物的细胞属性的步骤进一步包括：
标记在明场显微图像和第三荧光显微图像存在目标物区域重合的目标物，该目标物经目标物重合判断确定细胞属性为死细胞；提取重合判断结果中细胞属性为死细胞的目标物的直径信息；对比目标物的直径信息与预设的靶细胞最小直径、预设效应细胞最大直径，进一步确定直径大于等于靶细胞最小直径且细胞属性为死细胞的目标物为死靶细胞，进一步确定直径小于效应细胞最大直径且细胞属性为死细胞的目标物为死效应细胞；
标记在明场显微图像上存在目标物区域，且在第三荧光显微图像无目标物区域重合的目标物，该目标物经目标物重合判断确定细胞属性为活细胞；提取重合判断结果中细胞属性为活细胞的目标物的直径信息；对比目标物的直径信息与预设的靶细胞最小直径、预设效应细胞最大直径，进一步确定直径大于等于靶细胞最小直径且细胞属性为活细胞的目标物为活靶细胞，进一步确定直径小于效应细胞最大直径且细胞属性为活细胞的目标物为活效应细胞。
步骤340为获取细胞参数的步骤。在步骤340中，基于细胞属性对多个目标物进行分类计数及统计，获取细胞参数。具体的，按照步骤330确定的不同细胞属性对所有目标物进行分类计数及统计，获取细胞参数。关于细胞参数的具体描述可以参见步骤220及其相关描述，在此不再赘述。
流程300还可以包括基于多个显微图像生成叠加图像的步骤。在一些实施例中，图像分析模块可对多个显微图像进行图像融合，以生成叠加图像。在一些实施例中，流程300还包括基于细胞属性标识叠加图像中的多个目标物的步骤。关于对多个显微图像进行图像融合的具体描述可以参见图5及其相关描述，在此不再赘述。
在一些实施例中，图像分析模块可连接输出设备(显示屏)输出生成的无标识的叠加图像或带标识的叠加图像。本说明书的一些实施例通过将图像叠加合成分析获得的目标物属 性直接标识在叠加图像上，并向用户输出，使获得分析结果的方式更快速、直观、高效。
图4是根据本说明书一些实施例所示的提取显微图像中目标物轮廓的示例性流程图。在一些实施例中，流程400可以由图像分析模块执行。如图4所示，流程400包括步骤410至步骤430。
步骤410为显微图像去噪的步骤。在步骤410中，基于每个显微图像进行滤波处理，获取去噪的显微图像。在一些实施例中，滤波处理包括中值滤波和/或高斯滤波。
步骤420为显微图像二值化的步骤。在步骤420中，基于每个去噪的显微图像进行二值化处理，获取二值化的显微图像。二值化处理是使显微图像呈现出黑白效果的处理过程。
步骤430为图像分割及轮廓提取的步骤。在步骤430中，基于每个二值化的显微图像进行目标物分割，提取多个目标物区域及对应的轮廓信息。目标物分割是把二值化的显微图像分成若干个包含单个目标物的目标物区域并提取感兴趣的目标物信息的处理过程。在一些实施例中，目标物分割为阈值分割、区域生长法、分水岭分割和统计学分割中的一种。
图5是根据本说明书一些实施例所示的获取融合图像的示例性流程图。在一些实施例中，流程500可以由图像分析模块执行。如图5所示，流程500包括步骤510至步骤530。
在步骤510中，提取每个显微图像的融合特征点。
融合特征点是共培养样品上的同一空间点分别在不同的多个显微图像(明场显微图像和至少一个荧光显微图像)上的像素点。
在一些实施例中，融合特征点可以对应于共培养样品上的同一目标物特征。在一些实施例中，所述同一目标物特征可以包括颜色特征、纹理特征和形状特征等。例如，共培养样品中某目标物的形状特征为该目标物上的某一段弧线，则融合特征点为该段弧线在多个显微图像上对应的像素点。在一些实施例中，图像分析模块可以通过手动查找、自动查找和半自动查找等方式查找融合特征点。在一些实施例中，图像融合模块还可以通过相似度测量选取查找到的融合特征点。在一些实施例中，相似度测量可以包括基于互信息的测量、基于傅里叶分析的测量等等中的一种或多种的任何组合。
在一些实施例中，多个显微图像的融合特征点也可以对应于共培养样品上的同一位置坐标。例如，当明场显微图像和至少一个荧光显微图像采集于共培养样品的某一固定区域，则融合特征点可以是明场显微图像和至少一个荧光显微图像上的中心位置点。
在步骤520中，基于多个显微图像的对应融合特征点，配准多个显微图像。
配准是确定共培养样品上的多个空间点在多个显微图像上的像素点之间的对应关系。在一些实施例中，图像分析模块可以通过配准算法找出对应关系。
示例性地，图像分析模块可以通过配准算法，基于明场显微图像中某目标物上的某一段弧线上的至少部分像素点和至少一个荧光显微图像中该细胞上的该段弧线上的至少部分像素点之间的对应关系，找到明场显微图像和至少一个荧光显微图像之间的对应关系。
在一些实施例中，配准算法可以包括基于点的配准算法(例如，基于特征标志的配准算法)、基于曲线的配准算法、基于表面的配准算法(例如，基于表面轮廓的配准算法)、空间对齐配准算法、互相关配置配准算法、基于互信息的配准算法、顺序相似度检测算法(SSDA)、非线性变换配准算法、B样条配准算法等，或其任意组合。
在步骤530中，基于透明度和/或色度融合配准后的多个显微图像，获取融合图像。
融合是将多个显微图像中的信息综合到一个显微图像中。为了使得融合图像中同时包含多个显微图像(明场图像和至少一个荧光显微图像)中目标物的形状信息和目标物显示的颜色信息，图像分析模块可以将配准后的多个显微图像进行融合。
在一些实施例中，图像融合模块可以基于透明度将配准后的多个显微图像进行融合。基于透明度进行融合是将具有不同透明度的显微图像进行重叠，将重叠后的多个显微图像作为融合图像。在一些实施例中，基于透明度进行融合可以包括Alpha融合。
在一些实施例中，图像融合模块可以基于色度进行融合。基于色度进行融合是将不同 显微图像的色度进行特定运算，得到融合图像的色度，从而获取融合图像。例如，配准后的明场显微图像中的像素点A’和荧光显微图像中的像素点A”具有对应关系，基于像素点A’的色度RGB(220,200,100)和像素点A”的色度RGB(0,200,200)各颜色分量的平均值，可以获取融合图像中对应的像素点A的色度RGB(110,200,150)。
在一些实施例中，图像分析模块还可以进一步融合基于透明度获取的融合图像和基于色度获取的融合图像，获取最终的融合图像。
本说明书的一些实施例基于透明度和/或色度融合明场显微图像和荧光显微图像，可以在保留不同图像中目标物的形态特征的同时，融合不同荧光显微图像中目标物的颜色特征，使得融合图像包含更多信息，从而提高细胞杀伤效力和/或免疫活性的准确性。
在一些实施例中，融合还可以包括但不限于泊松融合算法、线性融合算法和Collage算法等中的一种或多种的组合。
图6是根据本说明书一些实施例所示的分析融合图像的示例性流程图。在一些实施例中，流程600可以由图像分析模块执行。如图6所示，流程600包括下述步骤。
在步骤610中，基于融合图像，获取多个目标物图像块。
目标物图像块为包含单个目标物的图像块。在一些实施例中，图像分析模块可以通过检测算法，从融合图像中获取目标物图像块。
在一些实施例中，检测算法可以通过对融合图像进行分割，根据分割得到的图像块的特征来检测单个目标物。具体地，检测算法可以先通过多尺度(multi-scale)的滑动窗口(Sliding-window)、选择性搜索(Selective Search)、神经网络或其他方法从融合图像中提取多个图像块，再提取多个图像块的初始特征，最后基于图像块的初始特征判断图像块是否为目标物图像块。其中，初始特征是图像块的浅层特征。例如，初始特征可以仅反映图像块中是否包含封闭线条，而不能反映线条的具体形状。
在步骤620中，提取目标物图像块的颜色特征和形状特征。
颜色特征是表征目标物图像块颜色的相关信息，可以反映所述目标物图像块中的目标物的颜色。在一些实施例中，颜色特征可以基于目标物图像块中各像素点在不同颜色分量中的色度表示。例如，颜色特征可以用目标物图像块中各像素点分别在红色分量R、绿色分量G和蓝色分量B上的色度表示。在一些实施例中，颜色特征可以通过其他方式表示(如，颜色直方图、颜色矩、颜色集等)。例如，对目标物图像块中各像素点在颜色分量中的色度进行直方图统计，生成表示颜色特征的直方图。又例如，对目标物图像块中各像素点在颜色分量中的色度进行特定运算(如，均值、平方差等)，将该特定运算的结果表示该目标物图像块的颜色特征。
在一些实施例中，图像分析模块可以通过颜色特征提取算法提取目标物图像块的颜色特征。颜色特征提取算法包括：颜色直方图、颜色矩、颜色集、颜色聚合向量和颜色相关图等。例如，图像分析模块可以基于目标物图像块中各像素点分别在每个颜色分量的色度，统计梯度直方图，从而获取颜色直方图。又例如，图像分析模块可以将目标物图像块分割为多个区域，用目标物图像块中各像素点分别在每个颜色分量的色度建立的多个区域的二进制索引的集合，以确定所述目标物图像块的颜色集。
形状特征是表征目标物图像块轮廓和区域的相关信息，可以反映所述目标物图像块中的目标物的形状。
在一些实施例中，图像分析模块可以通过边界特征法、Hough变换检测平行直线法、边界方向直方图法、傅里叶形状描述符法(Fourier shape deors)、形状参数法(shape factor)、有限元法(Finite Element Method或FEM)、旋转函数(Turning)和小波描述符(Wavelet Deor)等获取形状特征。
在步骤630中，基于多个目标物图像块的颜色特征和形状特征，获取多个目标物的细胞属性，并统计关联细胞属性的细胞参数。
在一些实施例中，图像分析模块可以分别基于多个目标物图像块的颜色特征和形状特征，确定融合图像中多个目标物图像块对应的多个目标物的颜色和形状，再基于每个目标物的颜色和形状，获取目标物的细胞属性，并统计关联细胞属性的细胞参数。
关于更多获取目标物的细胞属性的相关描述可以参见图7及其相关描述，在此不再赘述。
本说明书的一些实施例基于目标物图像块的颜色特征和形状特征确定融合图像中每个目标物的细胞属性，再基于每个目标物对应的细胞属性，统计分析细胞参数，可以提高细胞杀伤效力和/或免疫活性检测的准确性。
在一些实施例中，图像分析模块可以基于图像识别模型处理融合图像，获取多个目标物的细胞属性，并统计关联细胞属性的细胞参数。图像识别模型可以是预置参数的机器学习模型。可作为图像识别模型的机器学习模型包括但不限于目标检测模型、语义分割模型、实例分割模型等。预置参数是指机器学习模型训练过程中，学习到的模型参数。以神经网络为例，模型参数包括权重(Weight)和偏置(bias)等。
图7是根据本说明书一些实施例所示的图像识别模型的示例性示意图。
如图7所示，图像识别模型可以包括目标物图像块获取层710、特征提取层720及分析层730。例如，图像分析模块可以利用图像识别模型实现步骤610-630，获取多个目标物的细胞属性，并统计关联细胞属性的细胞参数。具体的，可以基于目标物图像块获取层710实现步骤610，基于特征提取层720实现步骤620，基于分析层730实现步骤630。
在一些实施例中，目标物图像块获取层710的输入为融合图像740，输出为多个目标物图像块750。在一些实施例中，目标物图像块获取层的类型可以包括但不限于视觉几何群网络(Visual Geometry Group Network)模型、Inception NET模型、全卷积神经网络(Fully Convolutional Network)模型、分割网络模型和掩模-卷积神经网络(Mask-Region Convolutional Neural Network)模型等。
在一些实施例中，特征提取层720的输入为多个目标物图像块750，输出为每个目标物图像块对应的颜色特征760和形状特征770。在一些实施例中，特征提取层的类型可以包括但不限于ResNet、ResNeXt、SE-Net、DenseNet、MobileNet、ShuffleNet、RegNet、EfficientNet或Inception等卷积神经网络(Convolutional Neural Networks,CNN)模型，或循环神经网络模型。
在一些实施例中，分析层730的输入为每个目标物图像块对应的颜色特征760和形状特征770，输出为多个目标物的细胞属性780，以及统计的关联细胞属性的细胞参数。在一些实施例中，分析层的类型可以包括但不限于全连接层、深度神经网络(DNN)等。
在一些实施例中，图像识别模型的所述预置参数通过训练过程生成。例如，模型获取模块可以基于带有标签的多个训练样本，通过端到端的方式训练初始图像识别模型，以得到图像识别模型。训练样本包括带有标签的样本融合图像。训练样本的标签为样本融合图像中目标物的细胞属性以及关联细胞属性的细胞参数。在一些实施例中，训练样本的标签可以通过人工标注获取。
在一些实施例中，图像识别模型可以由处理设备或第三方预先训练后保存在存储设备中，处理设备可以从存储设备中直接调用图像识别模型。
本说明书的一些实施例基于图像识别模型分析融合图像，获取目标物的细胞属性以及关联细胞属性的细胞参数，可以提高细胞杀伤效力和/或免疫活性检测的效率；并且基于训练样本的标签的不同，可以得到获取不同细胞属性及细胞参数的图像识别模型，提高细胞杀伤效力和/或免疫活性检测的适用性和针对性。
本说明书实施例之一提供细胞杀伤效力和/或免疫活性的检测方法或检测系统在检测细胞杀伤力、效应细胞免疫活性、制备免疫制品、免疫制品质量控制或个体免疫功能评价中的应用。
细胞杀伤已成为抗体药物研发和质量控制过程中至关重要的一步。在抗体药物研发生产过程中，需要对得到的抗体药进行生物学功能鉴定，其中就包括抗体药介导的ADCC(抗体依赖的细胞毒作用)、CDC(补体依赖的细胞毒作用)和ADCP(抗体依赖细胞介导的吞噬作用)效应的检测。本说明书的一些实施例提供的细胞杀伤效力和/或免疫活性的检测方法或检测系统可通过检测细胞杀伤效力来直接评价上述抗体药生物学活性。此外，以CAR-T细胞治疗为代表的细胞治疗药物，也需要通过检测细胞杀伤效力来对制备的CAR-T细胞药物进行生物学功能评价和质量控制，从而保障CAR-T细胞药物的有效性和安全性。本说明书的一些实施例提供的细胞杀伤效力和/或免疫活性的检测方法或检测系统的上述应用均具有积极意义。
下述实施例中的实验方法，如无特殊说明，均为常规方法。下述实施例中所用的试验材料，如无特殊说明，均为自常规生化试剂公司购买得到的。
实施例1-磁珠分离人自然杀伤细胞
1.1、取人外周抗凝血50ml，用Ficoll-Hypaque密度梯度法离心分离出人PBMC。
1.2、用15mL PBS(含2mmol/L EDTA)以300g离心5min，洗涤细胞2次，以PBS(含1％血清，2mmol/L EDTA)悬浮细胞至1×10 8个/mL，置于1.5mL离心管中4℃备用。
1.3、将抗CD16抗体加入细胞中(终浓度为10μg/mL),4℃孵育30min。
1.4、用冷PBS(含2mmol/L EDTA)以700g，30s洗涤细胞2次，用0.8mL PBS(含1％血清，2mmol/L EDTA)重悬1×10 8个细胞，加入0.2mL羊抗小鼠IgG包被的磁珠，4℃孵育30min，并每5min振荡一次。
1.5、用PBS(含2mmol/L EDTA)以700g，30s洗涤细胞2次，用1mL PBS(含1％血清，2mmol/L EDTA)重悬细胞，室温备用。
1.6、MS柱安装于磁铁架上，用1mL PBS(含1％血清，2mmol/L EDTA)预洗柱子3次，然后将上述细胞悬液加入柱子中，收集流出液，再加入柱子中，以1mL PBS(含2mmol/L EDTA)洗涤柱子10次。
1.7、取下分离柱以脱离磁场，用3mL PBS(含1％血清，2mmol/L EDTA)冲洗下磁珠结合的细胞，反复3次，300g离心5min，以适量培养基重悬细胞，计数，4℃备用。
实施例2-自然杀伤细胞免疫活性检测
2.1、靶细胞标记
肿瘤细胞(K562细胞系)，制备成细胞浓度为1×10 5个/mL的悬液，向1mL悬液中加入1μL浓度为20μM的CFSE进行标记，37℃避光孵育30min；
孵育完成后，室温下400g离心3min，吸走上清，加入1mL含有血清的培养基，得到CFSE标记的K562细胞。
2.2、制备自然杀伤细胞悬液
取适量实施例1制备的自然杀伤细胞，调节其细胞浓度为1×10 6个/mL，置于4℃备用。
2.3、细胞共培养
实验组：在96孔板的样品孔中同时加入100μL步骤2.1中的CFSE标记的K562细胞和100μL步骤2.2中的自然杀伤细胞作为实验组，效靶比设置为10:1，37度5％CO 2培养箱中共培养4小时。共培养结束后，加入Hoechst33342(购买自Thermofisher，USA)和PI染料(购买自Sigma，USA)进行染色，得到共培养样品。
对照组：在96孔板的样品孔中仅加入100μL步骤2.1中的CFSE标记的K562细胞和100μL培养基作为对照组靶细胞。在96孔板的样品孔中仅加入100μL步骤2.2中的自然杀伤细胞和100μL培养基作为对照组自然杀伤细胞。参照实验组的方法对对照组靶细胞和对照组自然杀伤细胞进行培养和染色，得到对照组靶细胞样品和对照组自然杀伤细胞样品。
2.4、显微成像
将步骤2.3得到的共培养样品、对照组靶细胞样品和对照组自然杀伤细胞样品分别加入血球计数板。
实验组：将加入共培养样品的血球计数板放置到检测仪器的样品台上，在共培养样品的固定区域分别用明场和匹配三种荧光标记的显微荧光通道进行显微成像，得到明场显微图像和三个荧光显微图像。
其中，三个荧光通道信息及顺序分别如下：
FL1：Ex 375nm，Em 460nm；FL2：Ex 480nm，Em 535nm；FL3：Ex 525nm，Em 600LP；
FL1通道激发与采集Hoechst33342荧光，FL2通道激发与采集CFSE荧光，FL3通道激发与采集PI荧光。
对照组：对于分别加入对照组靶细胞样品和对照组自然杀伤细胞样品的血球计数板，参照实验组的方法对其分别进行显微成像，得到对照组的明场显微图像和荧光显微图像。
2.5、图像叠加合成分析
实验组：细胞杀伤效力和/或免疫活性的检测系统将共培养样品在FL1通道、FL2通道、FL3通道下的荧光显微图像以及明场显微图像进行图像叠加合成分析。图像叠加后，检测系统对细胞所在的同一位置进行标示，若此处只有FL1的信号，标示并统计数目为a；此处同时有FL1和FL2信号，标示并统计数目为b；此处同时有FL1和FL3信号，标示并统计数目为c；此处同时有FL1，FL2和FL3信号，标示并统计数目为d。
并自定义编辑公式如下：
活靶细胞总数＝b；死靶细胞总数＝d；
靶细胞死亡率(％)＝d/(b+d)×100；
活自然杀伤细胞总数＝a；死自然杀伤细胞总数＝c；
自然杀伤细胞死亡率(％)＝c/(a+c)×100。
图8至图10分别为FL1通道、FL2通道、FL3通道下的荧光显微图像，图11为FL1、FL2、FL3通道下的荧光显微图像的叠加图像。图11中的区域w、区域x、区域y、区域z内均包含目标物。对于区域w内的目标物，其在图8至图10的重合区域w-1、w-2、w-3处有FL1信号，没有FL2和FL3，其细胞属性为活自然杀伤细胞。对于区域x内的目标物，其在图8至图10的重合区域x-1、x-2、x-3处分别有FL1信号和FL2信号，没有FL3信号，其细胞属性为活靶细胞。对于区域y内的目标物，其在图8至图10的重合区域y1、y2、y3处分别有FL1信号和FL3信号，没有FL2信号，其细胞属性为死自然杀伤细胞。对于区域z内的目标物，其在图8至图10的重合区域z1、z2、z3处分别有FL1信号、FL2信号和FL3信号，其细胞属性为死靶细胞。
在经过检测系统的分析，图8至图10中，活自然杀伤细胞总数为9，活靶细胞总数为15，死自然杀伤细胞总数为2，死靶细胞总数为28。进一步的，靶细胞死亡率65.11％，自然杀伤细胞死亡率为18.18％。
对照组：参照实验组的检测方法，对对照组的明场显微图像和荧光显微图像进行图像叠加合成分析，以获得对照组靶细胞死亡率以及对照组自然杀伤细胞死亡率，从而计算细胞特异杀伤率和自然杀伤细胞自伤率。
其中，细胞特异杀伤率(％)＝靶细胞死亡率-对照组靶细胞死亡率；
自然杀伤细胞自伤率(％)＝效应细胞死亡率-对照组自然杀伤细胞死亡率。
在设置对照的情况下，排除细胞自然死亡等情况对自然杀伤细胞免疫活性评估结果的影响，提高检测精确度。
实施例3-自然杀伤细胞免疫活性检测
3.1、制备自然杀伤细胞悬液
取适量实施例1制备的自然杀伤细胞，调节其细胞浓度为1×10 6个/mL，置于4℃备用。
3.2、标记肿瘤细胞
收集肿瘤细胞(K562细胞系)，制备成细胞浓度为1×10 5个/mL的悬液，向1mL悬液中加入1μL浓度为20μM的CFSE进行标记，37℃避光孵育30min。
孵育完成后，室温下400g离心3min，吸走上清，加入1mL含有血清的培养基，得到CFSE标记的K562细胞。
3.3、细胞共培养
实验组：取100μL步骤3.1中的自然杀伤细胞悬液和100μL步骤3.2中的CFSE标记的K562细胞加入96孔板的样品孔中，效靶比设置为10:1，37度5％CO 2培养箱中孵育共培养4小时。共培养结束后，加入2μL死细胞染料PI，室温孵育10min，得到共培养样品。
对照组：在96孔板的样品孔中仅加入100μL步骤3.2中的CFSE标记的K562细胞和100μL培养基作为对照组靶细胞。在96孔板的样品孔中仅加入100μL步骤3.1中的自然杀伤细胞和100μL培养基作为对照组自然杀伤细胞。参照实验组的方法对对照组靶细胞和对照组自然杀伤细胞进行培养和染色，得到对照组靶细胞样品和对照组自然杀伤细胞样品。
3.4、显微成像
实验组：取20μL共培养样品加入血球计数板，放置到检测仪器的样品台上，在共培养样品的固定区域分别用明场通道、FL1通道(匹配荧光染料CFSE)、FL2通道(匹配荧光染料PI)进行显微成像，得到明场显微图像和两个荧光显微图像。
其中，两个荧光通道信息及顺序分别如下：
FL1：Ex 480nm，Em 535nm；FL2：Ex 525nm，Em 600LP。
FL1通道激发与采集CFSE荧光，FL2通道激发与采集PI荧光。
对照组：对于分别加入20μL对照组靶细胞样品和20μL对照组自然杀伤细胞样品的血球计数板，参照实验组的方法对其分别进行显微成像，得到对照组的明场显微图像和荧光显微图像。
3.5、图像合成分析
实验组：细胞杀伤效力和/或免疫活性的检测系统将明场、FL1通道以及FL2通道下的荧光显微图像进行图像叠加合成分析。其中，明场通道下拍摄图像识别的目标为总细胞(包含靶细胞与自然杀伤细胞)；FL1通道下拍摄图像识别的目标为靶细胞(包含活靶细胞与死靶细胞)；FL2通道下拍摄图像识别的目标为总死细胞(包含死靶标细胞和死自然杀伤细胞)。
三张图像叠加后，检测系统对细胞所在的同一位置进行标示：
无荧光信号，标示并统计数目为a；只有FL1信号，标示并统计数目为b；同时有FL1和FL2信号，标示并统计数目为d；只有FL2信号，标示并统计数目为c。
自定义编辑公式参照实施例2的步骤2.5。
对照组：参照实验组的检测方法，对对照组的明场显微图像和荧光显微图像进行图像叠加合成分析，以获得对照组靶细胞死亡率以及对照组自然杀伤细胞死亡率，从而计算细胞特异杀伤率和自然杀伤细胞自伤率，公式参照实施例2的步骤2.5。
3.6、评价免疫活性
将靶细胞死亡率与死亡率阈值进行比较，所述死亡率阈值根据不同的情况进行设定，本实施例中死亡率阈值的上限设为40％，下限设为20％。
需要注意的是，此处死亡率阈值的上限和下限需要根据实际情况综合评定，此处数值仅作为示例或者参考。
所得检测结果中：靶细胞死亡率大于等于40％的实验组中，其自然杀伤细胞的免疫活性较好；靶细胞死亡率小于等于20％的实验组中，其自然杀伤细胞的免疫活性较差；靶细胞死亡率大于20％小于40％的实验组中，其自然杀伤细胞的免疫活性正常。
实施例4-自然杀伤细胞免疫活性检测
4.1、制备自然杀伤细胞悬液
取适量自然杀伤细胞，调节其细胞浓度为1×10 6个/mL，置于4℃备用。所述自然杀伤细胞的最大直径为9μm。
4.2、制备肿瘤细胞
收集肿瘤细胞(K562细胞系)，制备成细胞浓度为1×10 5个/mL的悬液，再向96孔板中加入100μL K562细胞，所述K562细胞的最小直径为9μm。
4.3、自然杀伤细胞与肿瘤细胞共培养
实验组：取100μL步骤4.1中的自然杀伤细胞悬液加入至已有K562细胞的96孔板中，效靶比设置为10:1，37度5％CO 2培养箱中共培养4小时。
对照组：在100μL靶细胞中仅加入100μL培养基(作为靶细胞对照组)，100μL自然杀伤细胞中仅加入100μL培养基(作为自然杀伤细胞对照组)，并在与实验组相同的条件下进行培养，在检测时，也与实验组使用相同的检测和分析方法。
4.4、共培养4小时后，去掉上清，并用PBS缓冲液清洗后，加入0.25％胰蛋白酶将贴壁生长的细胞消化下来，离心，PBS重悬，制得细胞悬液。
4.5、将步骤4.4中的细胞悬液与0.2％台盼蓝溶液以1:1的体积比混匀后，吸取20μL至计数板，放置到检测仪器(本实施例中所用仪器为Countstar全自动细胞计数仪)的样品台上。
4.6、在检测仪器上设置台盼蓝明场检测程序后，进行检测。
4.7、细胞杀伤效力和/或免疫活性的检测系统对图像进行图像叠加合成分析，然后对图像中不同直径大小的死活细胞进行计数：
具体对应为：
I、实验组中所得信息如下：
直径大于等于9μm的、未染色的为活的靶细胞，统计活靶细胞数记为a；
直径大于等于9μm的、染色的为死的靶细胞，统计死靶细胞数记为b；
直径小于9μm的、未染色的为活的效应细胞，统计活效应细胞数记为c；
直径小于9μm的、染色的为死的效应细胞，统计死效应细胞数记为d；
II、靶细胞对照组中，直径小于9μm的、染色的为死的靶细胞，统计死靶细胞数记为N；
III、效应细胞对照组中，直径小于9μm的、染色的为死的效应细胞，统计死效应细胞数记为M。
因此，所述靶细胞死亡率可以采用如下公式进行计算：
靶细胞死亡率＝死靶细胞数/(活靶细胞数+死靶细胞数)＝(b+d-M)/(a+b+d-M)×100％；
或者，靶细胞死亡率＝死靶细胞数/(活靶细胞数+死靶细胞数)＝(b+N)/(a+b+N)×100％。
所述效应细胞死亡率可以采用如下公式进行计算：
效应细胞死亡率＝死效应细胞数/(活效应细胞数+死效应细胞数)＝(d-N)/(c+d-N)×100％。
同时，还可以计算靶细胞特异杀伤率和效应细胞自伤率：
靶细胞特异杀伤率(％)＝实验组靶细胞死亡率-对照组靶细胞死亡率；
效应细胞自伤率(％)＝实验组效应细死亡率-对照组效应细胞死亡率。
本说明书实施例可能带来的有益效果包括但不限于：(1)本说明书的一些实施例检测方法可分别使用三染、双染、单染的靶细胞与效应细胞的共培养样品采集的显微图像进行图像叠加合成分析，准确且高效地区分活靶细胞、死靶细胞、活效应细胞和死效应细胞，并可按照实际检测需求依据对应属性的细胞个数计算得到细胞杀伤率等用于评价细胞杀伤效力和/或免疫活性的细胞参数；(2)本说明书的一些实施例提供的检测方法能够同时得到待测细胞样品的直读式图像信息和数据处理结果，相比流式细胞仪等常用检测方法提供的检测结果而言，所得结果更加直观，能在一台仪器上同时实现聚类分析和高内涵分析，得到细胞死亡率、细胞自伤率以及细胞特异杀伤率等多项数据，减少了检测步骤，提高了检测效率；且 检测方法简单，适用范围较广；(3)本说明书的一些实施例检测方法可分别使用三染、双染、单染的靶细胞与效应细胞的共培养样品采集的显微图像进行图像融合分析，基于融合图像中包含的目标物图像块中目标物的形状特征和颜色特征，可以快速获取目标物的细胞属性及关联细胞属性的细胞参数，减少处理流程，提高检测分析的效率；(4)本说明书的一些实施例提供的检测方法可检测的共培养样品的图像可识别特征组合方式多样，检测的适用范围广。
需要说明的是，不同实施例可能产生的有益效果不同，在不同的实施例里，可能产生的有益效果可以是以上任意一种或几种的组合，也可以是其他任何可能获得的有益效果。
本领域的技术人员应当理解，以上实施例仅为说明本发明，而不对本发明构成限制。凡在本发明的精神和原则内所作的任何修改、等同替换和变动等，均应包含在本发明的保护范围之内。

Claims (27)

  1. 一种细胞杀伤效力和/或免疫活性的检测方法，其特征在于，包括：
    获取共培养样品的固定区域的多个显微图像，其中，
    所述共培养样品为靶细胞与效应细胞共培养得到的细胞样品，所述固定区域包含多个目标物，多个所述目标物为包含多种不同属性细胞的细胞组，每个所述目标物均带有图像可识别特征，每个所述目标物的细胞属性通过该所述目标物的所述图像可识别特征在多个所述显微图像上显示的特征信息的集合来表征；
    对多个所述显微图像进行图像叠加合成分析或图像融合分析，获取多个所述目标物的所述细胞属性，并统计关联所述细胞属性的细胞参数；
    基于所述细胞参数，评价所述效应细胞的杀伤效力和/或免疫活性。
  2. 如权利要求1所述的方法，其特征在于，所述细胞属性包括细胞类型和细胞存活状态；多个所述目标物为包含活靶细胞、死靶细胞、活效应细胞和死效应细胞的细胞组。
  3. 如权利要求2所述的方法，其特征在于，不同所述细胞属性的所述目标物带有不同的所述图像可识别特征，所述图像可识别特征包括荧光标记特征。
  4. 如权利要求3所述的方法，其特征在于，所述共培养样品的获取包括如下步骤：
    基于携带预置荧光标记的靶细胞和无荧光标记的效应细胞进行共培养，获得共培养物；
    共培养预定时间后分别使用总细胞荧光标记和死细胞荧光标记来标记所述共培养物，获得所述共培养样品；其中
    所述共培养样品的所述固定区域内，携带所述预置荧光标记和所述总细胞荧光标记的所述目标物为所述活靶细胞，携带所述预置荧光标记、所述总细胞荧光标记和所述死细胞荧光标记的所述目标物为所述死靶细胞，仅携带所述总细胞荧光标记的所述目标物为所述活效应细胞，携带所述总细胞荧光标记和所述死细胞荧光标记的所述目标物为所述死效应细胞。
  5. 如权利要求3所述的方法，其特征在于，所述共培养样品的获取包括如下步骤：
    基于携带预置荧光标记的靶细胞和无荧光标记的效应细胞进行共培养，获得共培养物；
    共培养预定时间后使用死细胞荧光标记标记所述共培养物，获得所述共培养样品；其中，
    所述共培养样品的所述固定区域内，仅携带所述预置荧光标记的所述目标物为所述活靶细胞，携带所述预置荧光标记和所述死细胞荧光标记的所述目标物为所述死靶细胞，无荧光标记的所述目标物为所述活效应细胞，仅携带所述死细胞荧光标记的所述目标物为所述死效应细胞。
  6. 如权利要求2所述的方法，其特征在于，不同所述细胞属性的所述目标物带有不同的所述图像可识别特征，所述图像可识别特征包括荧光标记特征和细胞直径特征。
  7. 如权利要求6所述的方法，其特征在于，所述共培养样品的获取包括如下步骤：
    基于无荧光标记的靶细胞和无荧光标记的效应细胞进行共培养，获得共培养物；
    共培养预定时间后使用死细胞荧光标记标记所述共培养物，获得所述共培养样品；其中，
    所述共培养样品的所述固定区域内，无荧光标记且直径大于等于靶细胞最小直径的所述目标物为所述活靶细胞，携带所述死细胞荧光标记且直径大于等于靶细胞最小直径的所述目标物为所述死靶细胞，无荧光标记且直径小于效应细胞最大直径的所述目标物为所述活效应细胞，携带所述死细胞荧光标记且直径小于效应细胞最大直径的所述目标物为所述死效应细胞。
  8. 如权利要求2至7中任一项所述的方法，其特征在于，多个所述显微图像包括明场显 微图像和至少一个荧光显微图像，其中，所述至少一个荧光显微图像的成像参数基于多个所述目标物的所述图像可识别特征确定。
  9. 如权利要求8所述的方法，其特征在于，所述成像参数包括荧光通道种类和激发光波长。
  10. 如权利要求2至7中任一项所述的方法，其特征在于，所述细胞参数包括关联多个所述目标物的所述细胞属性的第一细胞参数；所述第一细胞参数包括靶细胞和效应细胞总数、靶细胞总数、活靶细胞总数、死靶细胞总数、靶细胞死亡率、效应细胞总数、活效应细胞总数、死效应细胞总数和效应细胞死亡率中的一种或多种。
  11. 如权利要求10所述的方法，其特征在于，所述基于所述细胞参数，评价所述效应细胞的杀伤效力和/或免疫活性包括：
    对比所述靶细胞死亡率与死亡率阈值，根据对比结果评价所述效应细胞的杀伤效力和/或免疫活性，其中，所述死亡率阈值包括上限和下限。
  12. 如权利要求2至7中任一项所述的方法，其特征在于，所述检测方法还包括：
    获取对照组靶细胞样品的固定区域的多个对照组显微图像，其中，所述对照组靶细胞样品为靶细胞单独培养得到的细胞样品，所述对照组靶细胞样品的所述固定区域包含多个带所述图像可识别特征的第一对照目标物；
    基于多个所述对照组显微图像进行图像叠加合成分析，获取多个所述第一对照目标物的所述细胞属性，并统计关联所述细胞属性的所述细胞参数。
  13. 如权利要求12所述的方法，其特征在于，所述细胞参数还包括关联多个所述第一对照目标物的所述细胞属性的第二细胞参数；所述第二细胞参数包括对照组靶细胞总数、对照组活靶细胞总数、对照组死靶细胞总数、对照组靶细胞死亡率和细胞特异杀伤率中的一种或多种。
  14. 如权利要求2至7中任一项所述的方法，其特征在于，所述检测方法还包括：
    获取对照组效应细胞样品的固定区域的多个对照组显微图像，其中，所述对照组效应细胞样品为效应细胞单独培养得到的细胞样品，所述对照组效应细胞样品的所述固定区域包含多个带所述图像可识别特征的第二对照目标物；
    基于多个所述对照组显微图像进行图像叠加合成分析，获取多个所述第二对照目标物的所述细胞属性，并统计关联所述细胞属性的所述细胞参数。
  15. 如权利要求14所述的方法，其特征在于，所述细胞参数包括关联所述第二对照目标物的所述细胞属性的第三细胞参数；所述第三细胞参数包括对照组效应细胞总数、对照组活效应细胞总数、对照组死效应细胞总数、对照组效应细胞死亡率和效应细胞自伤率中的一种或多种。
  16. 如权利要求1至7中任一项所述的方法，其特征在于，所述基于多个所述显微图像进行图像叠加合成分析或图像融合分析，获取多个所述目标物的所述细胞属性，以及关联所述细胞属性的细胞参数包括：
    提取每个所述显微图像中的多个目标物区域及对应的轮廓信息；
    基于多个所述显微图像的所述目标物区域及对应所述轮廓信息进行目标物重合判定，获取重合判定结果，其中，所述重合判定结果包括每个所述目标物的所述图像可识别特征在多 个所述显微图像上显示的所述特征信息的集合；
    基于所述重合判定结果，确定对应所述目标物的所述细胞属性；
    基于所述细胞属性对多个所述目标物进行分类计数及统计，获取所述细胞参数。
  17. 如权利要求16所述的方法，其特征在于，所述提取每个所述显微图像中的多个目标物轮廓及对应轮廓信息包括：
    基于每个所述显微图像进行滤波处理，获取去噪的所述显微图像；
    基于每个去噪的所述显微图像进行二值化处理，获取二值化的所述显微图像；
    基于每个二值化的所述显微图像进行目标物分割，提取多个所述目标物区域及对应的所述轮廓信息。
  18. 如权利要求16所述的方法，其特征在于，所述目标物重合判定包括基于特征点坐标距离计算进行初次重合判定和基于交并比计算进行的二次重合判定；所述基于多个所述显微图像的所述目标物区域及对应轮廓信息进行目标物重合判定，获取重合判定结果包括：
    对每个所述显微图像的每个所述目标物区域，遍历其余所述显微图像的每个所述目标物区域进行目标物重合判定，获取重合判定结果；
    其中，在所述目标物重合判定过程中：
    若对比的两个所述目标物区域在所述初次重合判定中判定为重合，则所述初次重合判定的判定结果即为本轮所述目标物重合判定的判定结果；
    若对比的两个所述目标物区域在所述初次重合判定中判定为不重合，则基于对比的两个所述目标物区域进行所述二次重合判定，所述二次重合判定的判定结果即为本轮所述目标物重合判定的判定结果。
  19. 如权利要求1至7中任一项所述的方法，其特征在于，所述基于多个所述显微图像进行图像融合分析或图像融合分析，获取多个所述目标物的所述细胞属性，并统计关联所述细胞属性的细胞参数包括：
    基于多个所述显微图像，获取融合图像；
    分析所述融合图像，获取多个所述目标物的所述细胞属性，并统计关联所述细胞属性的所述细胞参数。
  20. 如权利要求19所述的方法，所述基于多个所述显微图像，获取融合图像，包括：
    提取每个所述显微图像的融合特征点；
    基于多个所述显微图像的对应所述融合特征点，配准多个所述显微图像；
    基于透明度和/或色度融合配准后的多个所述显微图像，获取所述融合图像。
  21. 如权利要求19所述的方法，其特征在于，所述分析所述融合图像，获取多个所述目标物的所述细胞属性，并统计关联所述细胞属性的所述细胞参数，包括：
    基于图像识别模型处理所述融合图像，获取多个所述目标物的所述细胞属性，并统计关联所述细胞属性的细胞参数；所述图像识别模型为机器学习模型。
  22. 一种细胞杀伤效力和/或免疫活性的检测系统，其特征在于，所述检测系统包括如下模块：
    显微成像模块，用于获取共培养样品的固定区域的多个显微图像，其中，所述共培养样品为靶细胞与效应细胞共培养得到的细胞样品，所述共培养样品的所述固定区域包含多个目标物，多个所述目标物为包含多种不同属性细胞的细胞组，每个所述目标物均带有图像可识别特征，每个所述目标物的细胞属性通过该所述目标物的所述图像可识别特征在多个所述显 微图像上显示的特征信息的集合来表征；
    图像分析模块，用于基于多个所述显微图像进行图像叠加合成分析或图像融合分析，获取多个所述目标物的所述细胞属性，并统计关联所述细胞属性的细胞参数；
    评价模块，用于基于所述细胞参数评价所述效应细胞的杀伤效力和/或免疫活性。
  23. 如权利要求22所述的系统，其特征在于，所述系统还包括：
    样品台模块，用于承载细胞培养板，所述细胞培养板有多个样品孔，用于承载至少两种效靶比的所述共培养样品，所述显微成像模块对所述细胞培养板上各孔中的所述共培养样品分别进行成像，以获取所述共培养样品的多个所述显微图像。
  24. 如权利要求22所述的系统，其特征在于，所述系统还包括：
    样品自动更换模块，用于更换细胞培养板，再由所述显微成像模块对更换后的所述细胞培养板上的所述共培养样品进行拍摄，得到所述更换后的所述细胞培养板上的所述共培养样品的显微图像。
  25. 如权利要求1至21中任一项所述的检测方法或如权利要求22至24中任一项所述的系统在检测细胞杀伤力、效应细胞免疫活性、制备免疫制品、免疫制品质量控制或个体免疫功能评价中的应用。
  26. 一种细胞杀伤效力和/或免疫活性的检测装置，所述装置包括至少一个处理器以及至少一个存储器；
    所述至少一个存储器用于存储计算机指令；
    所述至少一个处理器用于执行所述计算机指令中的至少部分指令以实现如权利要求1至21中任意一项所述的方法。
  27. 一种计算机可读存储介质，所述存储介质存储计算机指令，当所述计算机指令被处理器执行时实现如权利要求1至21中任意一项所述的方法。
PCT/CN2021/127199 2020-10-28 2021-10-28 细胞杀伤效力和/或免疫活性的检测方法、系统及其应用 WO2022089552A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP21885279.6A EP4212853A4 (en) 2020-10-28 2021-10-28 METHOD AND SYSTEM FOR DETECTING CELL-KILLING EFFICACY AND/OR IMMUNITY ACTIVITY AND USE THEREOF
US18/301,247 US20230273188A1 (en) 2020-10-28 2023-04-16 Method and system for detecting cell-killing efficacy and/or immune activity, and application thereof

Applications Claiming Priority (14)

Application Number Priority Date Filing Date Title
CN202011173457.4A CN112301086A (zh) 2020-10-28 2020-10-28 一种体外自然杀伤细胞免疫活性的评价方法及其应用
CN202011173498.3A CN112285082A (zh) 2020-10-28 2020-10-28 一种基于细胞直径评价细胞杀伤效力的方法及其应用
CN202011173508.3 2020-10-28
CN202011176203.8 2020-10-28
CN202011173508.3A CN112285083B (zh) 2020-10-28 2020-10-28 一种细胞杀伤效力的评价方法
CN202011173496.4A CN112285081B (zh) 2020-10-28 2020-10-28 一种细胞杀伤效力的检测方法及其应用
CN202011173498.3 2020-10-28
CN202011173457.4 2020-10-28
CN202011176203.8A CN112304851B (zh) 2020-10-28 2020-10-28 一种体外自然杀伤细胞免疫活性的评价方法及其应用
CN202011173496.4 2020-10-28
CN202110127301.0 2021-01-29
CN202110127301.0A CN112813133B (zh) 2021-01-29 2021-01-29 一种细胞杀伤效力的检测方法、系统及其应用
CN202110595121.5A CN113237818A (zh) 2021-05-28 2021-05-28 一种细胞分析方法和系统
CN202110595121.5 2021-05-28

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/301,247 Continuation US20230273188A1 (en) 2020-10-28 2023-04-16 Method and system for detecting cell-killing efficacy and/or immune activity, and application thereof

Publications (1)

Publication Number Publication Date
WO2022089552A1 true WO2022089552A1 (zh) 2022-05-05

Family

ID=81383640

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/127199 WO2022089552A1 (zh) 2020-10-28 2021-10-28 细胞杀伤效力和/或免疫活性的检测方法、系统及其应用

Country Status (3)

Country Link
US (1) US20230273188A1 (zh)
EP (1) EP4212853A4 (zh)
WO (1) WO2022089552A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115266664A (zh) * 2022-07-13 2022-11-01 广州沙艾生物科技有限公司 一种免疫细胞治疗制剂体外杀伤效力评价方法及其应用
CN116188462A (zh) * 2023-04-24 2023-05-30 深圳市翠绿贵金属材料科技有限公司 一种基于视觉鉴定的贵金属质量检测方法及系统

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120302462A1 (en) * 2009-11-09 2012-11-29 Bjorn Onfelt System and Method for Detecting and Quantifying Active T-cells or Natural Killer Cells
CN110354142A (zh) * 2019-05-30 2019-10-22 天津市康婷生物工程集团有限公司 人体外扩增自然杀伤细胞对裸鼠卵巢癌实体瘤的治疗方法
CN112285083A (zh) * 2020-10-28 2021-01-29 上海睿钰生物科技有限公司 一种细胞杀伤效力的评价方法
CN112285081A (zh) * 2020-10-28 2021-01-29 上海睿钰生物科技有限公司 一种细胞杀伤效力的检测方法及其应用
CN112285082A (zh) * 2020-10-28 2021-01-29 上海睿钰生物科技有限公司 一种基于细胞直径评价细胞杀伤效力的方法及其应用
CN112301086A (zh) * 2020-10-28 2021-02-02 上海睿钰生物科技有限公司 一种体外自然杀伤细胞免疫活性的评价方法及其应用
CN112304851A (zh) * 2020-10-28 2021-02-02 上海睿钰生物科技有限公司 一种体外自然杀伤细胞免疫活性的评价方法及其应用
CN112813133A (zh) * 2021-01-29 2021-05-18 上海睿钰生物科技有限公司 一种细胞杀伤效力的检测方法、系统及其应用

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108753916A (zh) * 2018-06-04 2018-11-06 杭州启澜生物医学技术有限公司 一种快速评估car-t细胞体外杀伤效率的方法
CN111041064A (zh) * 2019-07-22 2020-04-21 徐州医科大学 一种体外评价car-t杀伤活性的方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120302462A1 (en) * 2009-11-09 2012-11-29 Bjorn Onfelt System and Method for Detecting and Quantifying Active T-cells or Natural Killer Cells
CN110354142A (zh) * 2019-05-30 2019-10-22 天津市康婷生物工程集团有限公司 人体外扩增自然杀伤细胞对裸鼠卵巢癌实体瘤的治疗方法
CN112285083A (zh) * 2020-10-28 2021-01-29 上海睿钰生物科技有限公司 一种细胞杀伤效力的评价方法
CN112285081A (zh) * 2020-10-28 2021-01-29 上海睿钰生物科技有限公司 一种细胞杀伤效力的检测方法及其应用
CN112285082A (zh) * 2020-10-28 2021-01-29 上海睿钰生物科技有限公司 一种基于细胞直径评价细胞杀伤效力的方法及其应用
CN112301086A (zh) * 2020-10-28 2021-02-02 上海睿钰生物科技有限公司 一种体外自然杀伤细胞免疫活性的评价方法及其应用
CN112304851A (zh) * 2020-10-28 2021-02-02 上海睿钰生物科技有限公司 一种体外自然杀伤细胞免疫活性的评价方法及其应用
CN112813133A (zh) * 2021-01-29 2021-05-18 上海睿钰生物科技有限公司 一种细胞杀伤效力的检测方法、系统及其应用

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4212853A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115266664A (zh) * 2022-07-13 2022-11-01 广州沙艾生物科技有限公司 一种免疫细胞治疗制剂体外杀伤效力评价方法及其应用
CN116188462A (zh) * 2023-04-24 2023-05-30 深圳市翠绿贵金属材料科技有限公司 一种基于视觉鉴定的贵金属质量检测方法及系统
CN116188462B (zh) * 2023-04-24 2023-08-11 深圳市翠绿贵金属材料科技有限公司 一种基于视觉鉴定的贵金属质量检测方法及系统

Also Published As

Publication number Publication date
EP4212853A1 (en) 2023-07-19
US20230273188A1 (en) 2023-08-31
EP4212853A4 (en) 2024-04-03

Similar Documents

Publication Publication Date Title
AU2013353154B2 (en) Cell monitoring device, cell monitoring method and program thereof.
US20230273188A1 (en) Method and system for detecting cell-killing efficacy and/or immune activity, and application thereof
CN107748256B (zh) 一种循环肿瘤细胞的液体活检检测方法
He et al. A new method for CTC images recognition based on machine learning
CN105122308B (zh) 用于使用连续染色组织上的单细胞划分的多路生物标志定量的系统和方法
CN103020585B (zh) 一种免疫组织阳性细胞和阴性细胞识别方法
Vink et al. An automatic vision‐based malaria diagnosis system
JP2018525746A (ja) コロニーコントラスト収集
CN103984939B (zh) 一种样本有形成分分类方法及系统
Singh et al. Breast cancer detection and classification of histopathological images
Chakrabortya et al. A combined algorithm for malaria detection from thick smear blood slides
US20200193139A1 (en) Systems and methods for automated cell segmentation and labeling in immunofluorescence microscopy
CN103994964B (zh) 针对荧光显微图像的凋亡细胞形态的量化分析方法
CN110060229A (zh) 一种骨髓白细胞的细胞自动定位分割方法
CN113237818A (zh) 一种细胞分析方法和系统
John et al. A novel approach for detection and delineation of cell nuclei using feature similarity index measure
CN113450317A (zh) 一种针对妇科临床微生物感染的免疫荧光图像检测方法
JP2011027542A (ja) 細胞画像解析装置
Takko et al. ShapeMetrics: A userfriendly pipeline for 3D cell segmentation and spatial tissue analysis
Sunny et al. Oral epithelial cell segmentation from fluorescent multichannel cytology images using deep learning
US20180040120A1 (en) Methods for quantitative assessment of mononuclear cells in muscle tissue sections
Fishman et al. Practical segmentation of nuclei in brightfield cell images with neural networks trained on fluorescently labelled samples
Lojk et al. Automatic cell counter for cell viability estimation
JP2024515512A (ja) 迅速で自動化された画像ベースのウイルスプラークおよびポテンシーアッセイ
WO2022111367A1 (zh) 基于高内涵成像的细胞耐药性检测方法、介质及电子设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21885279

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021885279

Country of ref document: EP

Effective date: 20230410

NENP Non-entry into the national phase

Ref country code: DE