CN111126329A - Method for automatically identifying pluripotent stem cell population - Google Patents
Method for automatically identifying pluripotent stem cell population Download PDFInfo
- Publication number
- CN111126329A CN111126329A CN201911396844.1A CN201911396844A CN111126329A CN 111126329 A CN111126329 A CN 111126329A CN 201911396844 A CN201911396844 A CN 201911396844A CN 111126329 A CN111126329 A CN 111126329A
- Authority
- CN
- China
- Prior art keywords
- pluripotent stem
- stem cells
- image
- cells
- picture
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
Abstract
The present invention provides a method for automatically identifying a pluripotent stem cell population, the method comprising: taking a picture of the stem cell sample in culture to obtain an electronic image picture, wherein the taken image picture is subjected to binarization processing firstly and then computer image analysis, so that pluripotent stem cells and non-pluripotent stem cells are divided in the picture; by this method, the area and extent of cells distinguishing pluripotent and non-pluripotent stem cells can be significantly improved.
Description
Technical Field
The invention belongs to the field of cell culture, and particularly relates to a method and a system for automatically identifying pluripotent stem cells.
Background
Fetal stem cells are a class of cells isolated from early embryos or primitive gonads that have the characteristics of immortalization, self-renewal, and multidirectional differentiation in vitro culture. Embryonic stem cells can be induced to differentiate into almost all cell types in the body, both in vitro and in vivo environments.
Embryonic stem cells are pluripotent and are characterized by the ability to differentiate into a variety of tissues, including germ line cells, by cells, but cannot develop into an individual by themselves. It can develop into cell tissues of ectoderm, mesoderm and endoderm. The embryonic stem cells can still maintain universal differentiation after multiple cell divisions.
The embryonic stem cell can be passed and proliferated infinitely theoretically, and does not lose its normal diploid genotype and phenotype, and after it is used as nuclear donor to make nuclear transplantation, a large number of individuals whose genotypes and phenotypes are identical can be obtained in a short period. And as an embryonic stem cell called "seed cell", it can provide a large amount of material for clinical tissue organ transplantation. Thus, the problem of transplant rejection among allotypic individuals which always plague the immunological and medical communities can be solved.
In the process of culturing the embryonic stem cells for multiple generations in daily life, the possibility of differentiation among single cells also exists, cells with various biological functions are always expected to be obtained, the pluripotent stem cells always exist in the form of colonies, the cell colonies are very compact in appearance, the intervals among the cells are small, and the outline boundary is not clear. Differentiated and undifferentiated cells taste mixed together, and in particular, the boundary between undifferentiated cells and cells that have differentiated or have the potential to differentiate is not apparent. Conventionally, it is common practice to sort out those undifferentiated cells in a culture vessel by the human eye through personal experience. The colonies of the selected undifferentiated cells are cultured for the next generation, but the manual selection has the defects of low efficiency, impossibility of realizing the unification of standards and incapability of realizing large-scale industrial production.
There is a need to provide a more efficient and reliable means or method of picking colonies.
Disclosure of Invention
In order to overcome the defects of the conventional technology of clients, the invention provides a method for automatically identifying pluripotent stem cell colonies. Pluripotent stem cell colonies as used herein generally refer to undifferentiated stem cells.
In one aspect, therefore, the present invention provides a method for automatically distinguishing between different cells, the method comprising imaging or photographing a population of cells and processing the photograph or images obtained thereby to distinguish between cells of different functions.
In some preferred modes, the present invention photographs a population of stem cells in which a large number of stem cells are aggregated, and then optically processes the photographs. The cell population is cultured in a vessel, such as a petri dish, which typically contains a culture medium in which the cells reside, and the division and differentiation and growth of the cells, typically adherent growth. Therefore, pictures of the adherent cell population of the vessel are taken, and the pictures or images are processed.
In some methods, the captured electronic image (generally, the electronic image) is subjected to image binarization processing, so that the image is in two different states, and different cells show different states under different shapes, so that different cell groups have outlines, and thus can be automatically distinguished. The display of the outlines also indicates the position of the different cells in the image, which may also correspond to the position of the different cells in the container, thus providing the possibility of achieving automatic sorting or screening.
In some embodiments, a binarization process is performed to differentiate pluripotent stem cells from non-pluripotent stem cells in a cell population, so that the pluripotent stem cells and the non-pluripotent stem cells have distinct outlines or boundaries, which divide cell populations having different functions, thereby facilitating automatic screening or selecting of target cell populations in a container. For example, non-pluripotent stem cells are removed, the pluripotent stem cells are retained for further culturing, or the retained pluripotent stem cells are divided into smaller aliquots and then cultured again or next generation pluripotent stem cells. This ensures the purity and continuity of the cells in each culture.
In some preferred modes, the step of image binarization processing includes a process of setting the gray value of a pixel point on the image to be 0 or 255, so that the image obviously presents a black-and-white effect. Thereby dividing pluripotent stem cells from non-pluripotent stem cells.
It is understood that different cell populations have different optical representations in different images or photographs taken, and therefore, when the cells are divided by different values in the images taken, cells with the same optical properties can be classified. For example, pluripotent stem cells are a group of cells that function as differentiated cells, similar to the concept of colonies on microorganisms, having the same optical properties, and when treated with different values, behave in the same manner between each value. At each value, however, functionally identical cells or cell populations are differentiated from other functional cells, but not to the same extent; when two proper values are selected, the degree of distinguishing between the cells or cell populations with consistent functions and the cells with other functions is different, and once the values are proper, the maximum degree of distinguishing can be achieved, so that the cell populations with different functions show a generally obvious outline on the image, and thus the cells are distinguished. Once identified, the cell population in the culture vessel can be localized or labeled, providing convenience for subsequent screening.
It can be understood that, for distinguishing the pluripotent stem cells from the non-pluripotent dry-cleaning cells, when the image or the photo is subjected to binarization processing, the gray value is preferably 0 or 255, so that the two cells can be easily distinguished from each other, and then the subsequent purposes of automatic positioning, boundary division and automatic screening can be realized.
In some preferred modes, the image is subjected to binarization processing by using an Otsu binarization method to obtain a processed image. The tsu binarization method is a commonly used image processing method, and is used for automatically binarizing clustering-based images or degrading a gray level image into a binary image in computer vision and image processing. The algorithm is named after Dajin Zhan. The algorithm assumes that the image will contain two classes of pixels according to a bi-modal histogram (foreground pixels and background pixels), and it then calculates the best threshold to separate the two classes so that their intra-class variance is minimal; since the squared distance of two is constant, their inter-class variance is the largest. Therefore, the Dajin binarization method is roughly the discretization simulation of the one-dimensional Fisher discriminant analysis.
In the Otsu algorithm, we exhaustively search for a threshold that minimizes the intra-class variance, defined as the weighted sum of the variances of two classes:
wherein the weight ω isiAre the probabilities of two classes separated by a threshold t, and
is the variance of these two classes.
Otsu demonstrated that minimizing the intra-class variance and maximizing the inter-class variance are the same:
using class probability omegaiSum mean μiTo indicate. Class probability omega1(t) calculating with a histogram with threshold t:
and the mean value mu of1(t) is:
where x (i) is the value in the center of the ith histogram bin. Similarly, ω of the right histogram can be found for bins greater than t2(t)，μ2. The class probabilities and class means may be calculated iteratively. This idea would result in an efficient algorithm.
The Otsu algorithm derives a threshold over the 0:1 range. This threshold is used for the dynamic range of pixel intensities present in the image. For example, if the image contains only between 155 and 255 pixel intensities, the Otsu threshold of 0.75 would map to the grayscale threshold of 230 (rather than 192, since the image contains pixels that are not the full range of 0-255).
The invention also surprisingly discovers that after the image is subjected to binarization processing by an extra-large binarization method, the undifferentiated stem cell group and the differentiated cells can be accurately distinguished, so that the position and the range of the differentiated stem cell group in the cell group can be accurately positioned, and a foundation is laid for automatically selecting the undifferentiated stem cell group.
In practice, the method does not identify single cells but identifies whether blocks of cell clusters maintain a pluripotent state or not, and has the function of pluripotent stem cells. Pluripotent stem cell (ES) morphological characteristics: ES cells have a morphological structure similar to that of early embryonic cells, have a large nucleus with one or several nucleoli, mostly euchromatin in the nucleus, less cytoplasmic cytoplasm and a simple structure. When cultured in vitro, the cells are closely arranged and grow in colony shape. When stained with alkaline phosphatase, E cells were reddish brown, while surrounding fibroblasts were yellowish; there is a clear boundary between the cell clone and the surrounding, the formed clone cells are not clear of each other, and the cell surface has lipid droplets with stronger refraction. The cell clone has various shapes, and most of the cell clones are island-shaped or nest-shaped.
When the staining is not carried out, the boundary of ES cells and non-ES cells is not clear and obvious, although the ES cells and the non-ES cells can be selected by naked eyes, the manual operation is complicated after all, but the accuracy is not high, the experience of each person is different, the limit of the determined pluripotent stem cells and the limit of the non-pluripotent stem cells are different, if the industrial-scale production is realized, cell groups with basically consistent functions are not easy to obtain, and the occurrence of mixed cells affects the continuous generation culture of the pluripotent stem cells.
In another aspect, the present invention provides a method of automatically identifying a pluripotent stem cell population, the method comprising: and (3) taking a picture of the stem cell sample in culture to obtain an electronic image picture, wherein the taken image picture is subjected to binarization processing firstly and then computer image analysis, so that the pluripotent stem cells and the non-pluripotent stem cells are divided in the picture. In some embodiments, the binarized grayscale values are 0 and 255.
In some approaches, the computer image analysis method employs the madzu binarization algorithm.
In some embodiments, the method further comprises demarcating the pluripotent stem cells and the non-pluripotent stem cells in the image after processing the image.
In some approaches, after the demarcation of the limits, it is localized to a specific sample of stem cells corresponding to the photograph taken.
In another aspect, the present invention provides a use of image binarization processing for distinguishing pluripotent stem cells from non-pluripotent stem cells, wherein in the use, a stem cell sample in culture is photographed to obtain an electronic image photograph, wherein the photographed image photograph is firstly subjected to binarization processing and then subjected to computer image analysis, so as to distinguish the pluripotent stem cells from the non-pluripotent stem cells in the photograph.
In some embodiments, the binarized grayscale values are 0 and 255.
In some approaches, computer image analysis methods employ the great zu binarization algorithm.
In some embodiments, the method further comprises demarcating the pluripotent stem cells and the non-pluripotent stem cells in the image after processing the image.
In some embodiments, wherein the demarcation of the limits is followed by a location on a particular sample of stem cells corresponding to the photograph taken.
Advantageous effects
The invention carries out binarization processing on the image, particularly adopts an Otsu algorithm to carry out binarization processing on the image, and can effectively distinguish the pluripotent stem cell group from other non-pluripotent stem cells, thereby realizing the subsequent automatic identification and screening process of automatic positioning and automatic range division, and greatly improving the accuracy.
Drawings
FIG. 1 is an image of a cell population in a culture vessel, wherein the image has pluripotent stem cells and non-pluripotent stem cells.
FIG. 2 is a set of images of a plurality of pluripotent stem cells divided by manual selection.
FIG. 3 is a set of images of a plurality of non-pluripotent stem cells divided by manual selection.
Fig. 4 is a schematic diagram of regions of multi-energy and non-multi-energy partitioned by the image binarization processing of the present invention.
Detailed Description
It is now described by way of example how the invention may be carried into effect and not constitute a limitation on the invention, the scope of which shall be governed by the claims.
Example 1: method for automatically identifying pluripotent stem cells (image processed by pretreatment)
ESC cultures (embryonic Stem cells) (ESCs actually have pluripotent stem cells (ES) in culture, with non-ES cells present)
Initial culture of ESC
1.1.1 before the cells are thawed, a switch of a water bath kettle is turned on, and the temperature is adjusted to 37 ℃.
1.1.2 bringing 6-well plates coated with Matrigel (Matrigel) into a biosafety cabinet, aspirating the coated DMEM-HG medium (purchased from Stem cell technology, Canada Inc.) and then adding freshly prepared 500uL pluripotent Stem cell medium (mTeSR1) to the wells
1.1.3 about 3mL of mTeSR1 was added to a 15mL tube and left to stand.
1.1.4 taking out ESC to be thawed from liquid nitrogen, putting the ESC into a gradient cooling box with-20 ℃ precooling, and transferring the ESC into a cell room.
1.1.5 taking out the cryopreservation tube from the gradient cooling box with the help of tweezers, putting the cryopreservation tube into a water bath kettle, shaking until cells in the cryopreservation tube are thawed, and paying attention to prevent water in the water bath kettle from soaking the cover of the cryopreservation tube.
1.1.6 carry some alcohol into the biosafety cabinet after spraying it against the cryovial, aspirate 1mL mTeSR1 from the 15mL tube with a 5mL pipette, and then slowly add it to the cryovial.
1.1.7 remove all liquid from the vial, add slowly to the tube containing the remaining 2ml of mTeSR1, and then add all to a 15ml centrifuge tube.
1.1.8 remove all media from the 15ml centrifuge tube and then add slowly to a well containing 500uL of mTeSR 1.
1.1.9 addition of media cells, inhibitors (Rock, https:// www.biomart.cn/infosuply/68159453. htm from ═ search _1) were added to the wells to a final concentration of 10 uM.
1.1.10 the plates were gently brought under a microscope to see cell density.
1.1.11 the 6 well plate was shaken back and forth 3-4 times, then the plate was placed in the incubator and moved back and forth 3-4 times without moving the plate within 24 hours.
Maintenance of hESC cells
4.3.1 passages or resuscitations after 24 hours of inoculation, plates were examined under a microscope and observed for adherent ESCs.
4.3.2A 50ml centrifuge tube containing pluripotent Stem cell culture medium (mTeSR1-Stem cell technology) was removed from the 4 ℃ freezer, placed on a centrifuge tube rack at room temperature and allowed to stand for 30 minutes or more to allow the culture to warm to room temperature. The broth and an empty 50ml centrifuge tube were placed into a biosafety cabinet.
4.3.3 bring the plate into a biosafety cabinet, carefully open the lid of the cover plate, put it aside, then tilt the plate forward, gently suck off the old media, add the old media to an empty 50ml centrifuge tube and recover, the old media is named embryonic stem cell conditioned medium (ESC-CM).
4.3.4 2ml of room temperature mTeSR1 was gently added from both sides of the wall to 1 well of a 6-well plate (the amount of culture depends on the type of well) so that the cells were not disturbed. The uncapped 6-well plate is not exposed to air for a long time, which may cause contamination.
4.3.5 the plate is then taken to a microscope to observe the cells.
4.3.6 the plate is then gently placed into an incubator. Ensuring that the incubator door is correctly closed.
4.3.7 liquid change is carried out every day.
Observation of
Starting the day after passage, the cells are observed on a microscope and photographed to obtain a picture (typically 2 pictures per photograph) or an electron picture. The images are labeled with detailed information such as cell line number, number of generations of cells, magnification of objective lens, and number of days after passage. Thus, a plurality of photographs were taken of cells under different cell culture conditions and generations.
Pretreatment of
First, in order to verify the reliability of the method of the present invention, cells of different generations of different cell lines (photographed on a culture dish) are photographed respectively to obtain 100 photographs (corresponding to 100 photographs of cells cultured in different culture vessels, wherein there are different generations of cell photographs), the photographed cells are simultaneously subjected to manual division of the pluripotent and non-pluripotent ranges (at the back of the culture dish), and then the accuracy analysis of the manual division of the ranges is verified by cell staining, which is performed by the conventional method.
Taking a picture, 100 pictures of 1600X1200 (pixels), wherein each picture is decomposed into a plurality of pictures of 80X 80; not only is: decomposing each image into 80 × 80 (pixel) images, performing extra-large binarization processing on the decomposed images, selecting gray values for processing, imaging at 0 and 255 (i.e. forming images with black and white effects), so as to form image results under different gray values, and performing image fitting on the image results under two gray values to form final region division, so as to display the boundary between the pluripotent stem cell group and the non-pluripotent stem cell group in the image, for example, as shown in fig. 4, in the image, the region divided by a black line is the boundary for distinguishing the pluripotent stem cells from the non-pluripotent stem cells, the region in the black line is the non-pluripotent stem cell region, and the region outside the black line is the non-pluripotent stem cells. The multiple images of the same photo are divided and then subjected to binarization processing, and then fitting is carried out again to form an image which is finally consistent with the original photo, so that 100 original area images are formed, and a multi-energy range area and a non-multi-energy range area are divided in the photographed area. The division of the image area into multi-functional and non-multi-functional areas can be reflected in the photographed area.
Image binarization is to select a gray level image with 256 brightness levels by an appropriate threshold value to obtain a binarized image which can still reflect the overall and local characteristics of the image. In digital image processing, a binary image plays a very important role, and firstly, the binarization of the image is beneficial to further processing of the image, so that the image is simple, the data volume is reduced, and the outline of an interested target can be highlighted. Secondly, the processing and analysis of the binary image are carried out, firstly, the gray level image is binarized to obtain a binarized image. All pixels with the gray levels larger than or equal to the threshold are judged to belong to the specific object, the gray level of the pixels is 255 for representation, otherwise the pixels are excluded from the object area, the gray level is 0, and the pixels represent the background or the exceptional object area. The invention discovers that the boundary between the pluripotent stem cells and the non-pluripotent stem cells in the cell culture can be obviously distinguished by adopting the binarization treatment. Any processing method in the prior art can be adopted for the specific analysis scheme after the binarization processing. In a preferred embodiment, the process is performed by the Dajin binarization method.
The specific method of the binarization processing is the Dajin binarization method. The tsu binarization method is a commonly used image processing method, and is used for automatically binarizing clustering-based images or degrading a gray level image into a binary image in computer vision and image processing. The algorithm is named after Dajin Zhan. The algorithm assumes that the image will contain two classes of pixels according to a bi-modal histogram (foreground pixels and background pixels), and it then calculates the best threshold to separate the two classes so that their intra-class variance is minimal; since the squared distance of two is constant, their inter-class variance is the largest. Therefore, the Dajin binarization method is roughly the discretization simulation of the one-dimensional Fisher discriminant analysis.
In the Otsu algorithm, we exhaustively search for a threshold that minimizes the intra-class variance, defined as the weighted sum of the variances of two classes:
wherein the weight ω isiAre the probabilities of two classes separated by a threshold t, and
is the variance of these two classes.
Otsu demonstrated that minimizing the intra-class variance and maximizing the inter-class variance are the same:
using class probability omegaiSum mean μiTo indicate. Class probability omega1(t) calculating with a histogram with threshold t:
and the mean value mu of1(t) is:
where x (i) is the value in the center of the ith histogram bin. Similarly, ω of the right histogram can be found for bins greater than t2(t)，μ2. The class probabilities and class means may be calculated iteratively. This algorithm will result in an efficient algorithm.
The Otsu algorithm derives a threshold over the 0:1 range. This threshold is used for the dynamic range of pixel intensities present in the image. For example, if the image contains only between 155 and 255 pixel intensities, the Otsu threshold of 0.75 would map to the grayscale threshold of 230 (rather than 192, since the image contains pixels that are not the full range of 0-255).
In this way, the captured image is binarized by the algorithm of Otsu, imaged at gray values of 0 and 255, respectively, and then processed or the final image.
The photographed area of a percentage of the samples photographed was simultaneously manually divided into areas, and then the areas divided on the manually processed samples were compared with the binarized photographs of the images photographed by the samples, and the consistency of the areas divided was compared, and it was found that the accuracy of the areas divided by the photographing process and the consistency of the areas divided by the manual areas were maintained at 98% conformity, the time taken manually for a percentage of the samples was about 1000 minutes (10 minutes per sample on average), and the time taken for the areas divided by the photographing process was about 10 seconds, and the time taken for each picture was 0.1 second.
The calculation rule of the conformity is: if the coincidence degree of the range of the image processing division and the range of the manual processing division is 100%, the image processing division is completely coincided, and if the coincidence degree of the range of the image processing division and the range of the manual processing division is 95%, the image processing division is completely coincided with the manual division region, and only 5 regions are incompletely coincided with the manual division region. In incomplete registration, some parts are registered and some parts are misaligned.
The staining treatment was performed on the cell population artificially divided into regions, and the degree of coincidence between the stained regions and the regions divided after the image treatment was 95%. This may be due to errors caused by the dyeing process.
Study of
We photographed photographs of manually divided regions and extracted colony images of pluripotent stem cells (fig. 2) and non-pluripotent stem cells (fig. 3), and then let the computer perform CNN network learning. After 100 cycles of learning we stopped training. So-called Convolutional Neural Networks (CNN) are a class of feed-forward Neural Networks (fed-forward Neural Networks) that contain convolution computations and have a deep structure.
The convolutional neural network uses the BP framework for learning in supervised learning, and the calculation process of the convolutional neural network is determined in LeCun (1989), and is one of the deep algorithms which learn in the BP framework at the earliest time. BP in convolutional neural networks is divided into three parts, namely a back propagation of the full connectivity layer with the convolution kernel and a back pass of the pooling layer (backward pass). The BP computation of the fully-connected layer is the same as that of the conventional feedforward neural network, and the backward propagation of the convolutional layer is a cross-correlation computation similar to the forward propagation:
where E is the error calculated by the cost function (cost function), f' is the derivative of the excitation function, α
Is the learning rate (learning rate), and if the forward propagation of the convolution kernel uses convolution calculation, the backward propagation also inverts the convolution kernel to perform convolution operation. The error function of the convolutional neural network can be selected from various options, and commonly includes Softmax loss function (Softmax loss), hinge loss function (hinge loss), triple loss function (tripletloss), and the like.
The parameter of the pooling layer is not updated in the reverse propagation, so that the error is only distributed to a proper position of the characteristic diagram according to a pooling method, and for the large pooling, all errors are endowed to the position of the maximum value; for mean pooling, the error is evenly distributed throughout the pooled region.
Convolutional neural networks typically use random gradient descent (SGD) within the BP framework and variants thereof, such as the Adam algorithm (Adaptive motion estimation). The SGD randomly selects a sample to calculate the gradient in each iteration, is favorable for information screening under the condition of a large number of learning samples, can quickly converge at the initial stage of the iteration, and has smaller calculation complexity.
Through CNN learning, the characteristics and features of pluripotent stem cells and non-pluripotent stem cells are stored in a computer, automatic analysis and learning are realized, and then detection and verification of other pluripotent stem cell regions are carried out.
And (3) verifying the test result:
the image which is subjected to binarization is used for enabling the CNN network to automatically identify, the result reaches 95% accuracy, and the image which is subjected to binarization processing is divided by the model which is learned by the CNN, so that the approximate region between the pluripotent stem cells and the non-pluripotent stem cells can be automatically and accurately divided, and the result is close to the actual manual division result. The coincidence or the accuracy rate of the pluripotent stem cell area obtained by the picture of binarization processing and the pluripotent stem cell identified by the CNN network is 95%.
Example 2 of implementation: method for automatic identification of pluripotent stem cells (images not processed in advance)
Compared with embodiment 1 described above, the obtained image was not subjected to binarization processing, but was directly subjected to processing with the result of CNN learning, and it was found that the degree of matching was only 60%. Although it is shown that the pluripotent stem cell population and the non-pluripotent stem cell population can be distinguished without binarization, the division region is not accurate, and some of the non-pluripotent stem cells are treated as pluripotent stem cells, and some of the non-pluripotent stem cells are treated as non-pluripotent stem cells, resulting in many results called false positives or false negatives.
Meanwhile, compared with the manual division of the regions, the conformity of the manual division of the regions is only 50%, so that when the automatic image is adopted to identify the pluripotent stem cell regions, the image binarization processing can obviously improve the accuracy of the division of the pluripotent stem cell regions and the non-pluripotent stem cell regions, simultaneously improve the processing efficiency and carry out subsequent standardized production.
The invention shown and described herein may be practiced in the absence of any element or elements, limitation or limitations, which is specifically disclosed herein. The terms and expressions which have been employed are used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, and it is recognized that various modifications are possible within the scope of the invention. It should therefore be understood that although the present invention has been specifically disclosed by various embodiments and optional features, modification and variation of the concepts herein described may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention as defined by the appended claims.
The contents of the articles, patents, patent applications, and all other documents and electronically available information described or cited herein are hereby incorporated by reference in their entirety to the same extent as if each individual publication was specifically and individually indicated to be incorporated by reference. Applicants reserve the right to incorporate into this application any and all materials and information from any such articles, patents, patent applications, or other documents.
Claims (10)
1. A method of automatically identifying a population of pluripotent stem cells, the method comprising: and (3) taking a picture of the stem cell sample in culture to obtain an electronic image picture, wherein the taken image picture is subjected to binarization processing firstly and then computer image analysis, so that the pluripotent stem cells and the non-pluripotent stem cells are divided in the picture.
2. The method as claimed in claim 1, wherein the binarized gray scale values are 0 and 255.
3. The method of claim 2, wherein the computer image analysis method adopts the Dajin binarization algorithm.
4. The method of claim 1, further comprising demarcating the pluripotent stem cells from the non-pluripotent stem cells in the image after processing the image.
5. The method of claim 4, wherein after the partition of the boundary, locating onto a specific sample of stem cells corresponding to the taken picture.
6. The application of image binarization processing to distinguishing pluripotent stem cells from non-pluripotent stem cells is that in the application, stem cell samples in culture are photographed to obtain an electronic image photo, wherein the photographed image photo is firstly subjected to binarization processing and then subjected to computer image analysis, so that the pluripotent stem cells and the non-pluripotent stem cells are divided in the photo.
7. Use according to claim 6, the binarized gray values being 0 and 255.
8. The use according to claim 7, the computer image analysis method adopts the Dajin binarization algorithm.
9. The use of claim 8, the method further comprising demarcating pluripotent stem cells from non-pluripotent stem cells in the image after processing the image.
10. Use according to claim 9, wherein after the division of the limits, the specific sample corresponding to the stem cells of the photographed picture is located.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911396844.1A CN111126329A (en) | 2019-12-30 | 2019-12-30 | Method for automatically identifying pluripotent stem cell population |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911396844.1A CN111126329A (en) | 2019-12-30 | 2019-12-30 | Method for automatically identifying pluripotent stem cell population |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111126329A true CN111126329A (en) | 2020-05-08 |
Family
ID=70505260
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911396844.1A Pending CN111126329A (en) | 2019-12-30 | 2019-12-30 | Method for automatically identifying pluripotent stem cell population |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111126329A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112102232A (en) * | 2020-07-31 | 2020-12-18 | 深圳大学 | Method and device for automatically evaluating colony quality of induced pluripotent stem cells |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102737232A (en) * | 2012-06-01 | 2012-10-17 | 天津大学 | Cleavage cell recognition method |
| US20120315620A1 (en) * | 2009-10-09 | 2012-12-13 | Kawasaki Jukogyo Kabushiki Kaisha | Method and Apparatus for Discriminating Undifferentiated Pluripotent Stem Cells, and Automated Culture Method and System |
| CN104778711A (en) * | 2015-04-27 | 2015-07-15 | 厦门大学 | Cell counting and positioning method at cleavage stage |
| CN108138113A (en) * | 2015-08-31 | 2018-06-08 | 爱平世股份有限公司 | Multipotent stem cells manufacture system |
| CN110060229A (en) * | 2019-05-29 | 2019-07-26 | 武汉兰丁医学高科技有限公司 | A kind of cell automatic positioning dividing method of myeloplast |
-
2019
- 2019-12-30 CN CN201911396844.1A patent/CN111126329A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120315620A1 (en) * | 2009-10-09 | 2012-12-13 | Kawasaki Jukogyo Kabushiki Kaisha | Method and Apparatus for Discriminating Undifferentiated Pluripotent Stem Cells, and Automated Culture Method and System |
| CN102737232A (en) * | 2012-06-01 | 2012-10-17 | 天津大学 | Cleavage cell recognition method |
| CN104778711A (en) * | 2015-04-27 | 2015-07-15 | 厦门大学 | Cell counting and positioning method at cleavage stage |
| CN108138113A (en) * | 2015-08-31 | 2018-06-08 | 爱平世股份有限公司 | Multipotent stem cells manufacture system |
| CN110060229A (en) * | 2019-05-29 | 2019-07-26 | 武汉兰丁医学高科技有限公司 | A kind of cell automatic positioning dividing method of myeloplast |
Non-Patent Citations (1)
| Title |
|---|
| 庞晋雁: ""干细胞图像分割及生长跟踪研究"", 《中国优秀硕士学位论文全文数据库-基础科学辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112102232A (en) * | 2020-07-31 | 2020-12-18 | 深圳大学 | Method and device for automatically evaluating colony quality of induced pluripotent stem cells |
| CN112102232B (en) * | 2020-07-31 | 2021-11-02 | 深圳大学 | Method and device for automatically evaluating colony quality of induced pluripotent stem cells |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2202291B1 (en) | Culture apparatus, culture information management method, and program | |
| EA025172B1 (en) | Method for determining developmental potential of one or more human embryos to reach blastocyst stage | |
| CN110504032B (en) | Method for predicting tumor mutation load based on image processing of hematoxylin-eosin staining tablet | |
| Piotrowski et al. | Deep-learning-based multi-class segmentation for automated, non-invasive routine assessment of human pluripotent stem cell culture status | |
| JP6015113B2 (en) | Cell evaluation apparatus, cell evaluation method and program | |
| CN112418130A (en) | Banana maturity detection method and device based on BP neural network | |
| Rudolph et al. | Online monitoring of microcarrier based fibroblast cultivations with in situ microscopy | |
| CN114283407A (en) | Self-adaptive automatic leukocyte segmentation and subclass detection method and system | |
| CN111126329A (en) | Method for automatically identifying pluripotent stem cell population | |
| JP6015112B2 (en) | Cell evaluation apparatus, cell evaluation method and program | |
| Ibaraki et al. | Application of image analysis to plant cell suspension cultures | |
| CN104755608A (en) | Automatic surveillance of in vitro incubating embryos | |
| Hradecká et al. | Segmentation and tracking of mammary epithelial organoids in brightfield microscopy | |
| Viswanath et al. | Grading of mammalian cumulus oocyte complexes using machine learning for in vitro embryo culture | |
| JPH07121220B2 (en) | Animal cell culture device, culture method and activity diagnostic device | |
| CN111598029B (en) | Method, system, server and storage medium for screening target cell strain | |
| CN107194319A (en) | The mitotic mapping sorted based on SVMs and knowledge method for distinguishing | |
| Coston et al. | Automated hiPSC culture and sample preparation for 3D live cell microscopy | |
| JP2011229412A (en) | Cell evaluation device, incubator, program, culture method, hand-technical evaluation device and cell evaluation model-building device | |
| JP6376196B2 (en) | Cell evaluation apparatus, cell evaluation method and program | |
| CN109633143B (en) | System and method for detecting bone marrow microcirculation environment of patient after hematopoietic stem cell transplantation | |
| CA3156826A1 (en) | Imaging system and method of use thereof | |
| Resto Irizarry et al. | Machine learning-assisted imaging analysis of a human epiblast model | |
| Wang et al. | OC_Finder: A deep learning-based software for osteoclast segmentation, counting, and classification | |
| JP7235027B2 (en) | Cell evaluation instruments, incubators and programs |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200508 |