WO2006095896A1 - Cultured cell monitoring system - Google Patents

Abstract

It is intended to provide a system of monitoring cultured cells which comprises a means of photographing an image of the cultured cells, and a means of texture-analyzing the image of the cultured cells thus photographed by using a parameter relating to the culture time of the cultured cells.

Description

 Description Cultured cell monitoring system Technical field

 The present invention relates to a cultured cell monitoring system that can monitor the state of a cultured cell (cell population) over time. Background art

 When culturing cells for a desired purpose, it is monitored over time (at appropriate time intervals) to determine whether the culture state is normal and whether there are any abnormalities due to problems such as contamination. There is a need to. Conventionally, in order to confirm the state of culture, for example, whether cells other than the intended purpose are not mixed or whether the cells have the desired differentiation, in addition to microscopic observation, some of the cells in culture are sampled and disrupted. Equally, it was generally done by performing biochemical analysis.

 However, with this monitoring method, it takes a long time to check the culture state each time, and the checking method is not easy, so it is difficult to find an abnormality quickly. In addition, if pathogenic cells are mixed in the sample, there is a risk of infection during the confirmation process. Furthermore, there is a risk of causing problems such as contamination during sampling.

 Therefore, development of a method capable of easily monitoring cultured cells has been desired.

As methods for easily discriminating the state of cells, the “sarcoma classification method by texture analysis” and the “pattern classification method of pigmented nevus by texture analysis” are already publicly known (Yuriko Murase, Toshiyuki Tanaka, Texture Sarcoma classification by analysis, Proceedings of 1999 IEICE General Conference, No. D-16-6, (2000-03); Toshiyuki Tanaka et al "Pattern Classincation of Nevus with Texture Analysis, The 26th International Conference of the IEEE BMES) These These methods are intended for tumor cells. Then, we focused on the uniformity of the size and shape of the three known moonri ¾r tables (homogeneous pattern, globular pattern, reticular pattern), and used them as specific parameters. It is automatically classified using texture analysis, which is a processing method, and is used for simple discrimination as a pre-stage to distinguish between benign and malignant tumors. In addition, these methods are methods for discrimination based on the photographed cell images, so there is no risk of infection or contamination during the confirmation work as described above, and confirmation can be done in a short time. It is excellent in that it can be done.

 However, none of the above methods is applied to patient samples and is a method for discriminating cell types, so it cannot be said that it is a method for monitoring the state of cultured cells over time. . In addition, the target cells are limited to tumor cells, and it cannot be applied to other types of cells. Disclosure of the invention

 The problem to be solved by the present invention is that confirmation of the state of culture of target cells over time can be performed easily and in a short time, and there is no risk of infection or contamination of cultured cells at the time of confirmation. It is to provide a culture cell monitoring system.

 The present inventor has intensively studied to solve the above problems. As a result, if the texture analysis, which is the image processing method described above, is used to monitor the cultured cells over time, and this analysis is performed using the parameters related to the elapsed culture time of the cultured cells as an index, We have found that the above-mentioned problems can be solved at once, and have completed the present invention.

 That is, the present invention is as follows.

(1) Culture cell monitoring equipped with means for taking images of cultured cells and means for analyzing the texture of the taken cells using the parameters related to the culture time of the cultured cells as an index system. In the above system, the cultured cells may be human-derived cells, for example, cells for regenerative medicine are preferable.

 (2) A method for monitoring a cultured cell, comprising: a step of taking an image of the cultured cell; and a step of performing a texture analysis on the taken image of the cell using a parameter related to the culture elapsed time of the cultured cell as an index. Brief Description of Drawings

 FIG. 1 is a schematic diagram of the cultured cell monitoring system of the present invention.

 FIG. 2 is a graph showing a photograph taken in preliminary experiment 1 in Example 1 and an analysis result.

 FIG. 3 is a view showing a photograph taken in preliminary experiment 2 in Example 1 and an analysis result.

 FIG. 4 is a graph showing a photograph taken in preliminary experiment 3 in Example 1 and an analysis result.

 FIG. 5 is a graph showing the results of preliminary experiment 4 in Example 1.

 FIG. 6 is a graph showing the results of Preliminary Experiment 5 in Example 1.

 FIG. 7 shows photographs and analysis results taken in preliminary experiment 6 in Example 1. FIG.

 FIG. 8 is a graph showing the analysis results in Monitoring Example 1 in Example 1.

 FIG. 9 is a photograph showing the results in Example 2 and a graph showing the analysis results. BEST MODE FOR CARRYING OUT THE INVENTION

 Hereinafter, the detection method of the present invention will be described in detail. However, the scope of the present invention is not limited to these descriptions, and modifications other than the following examples may be made as appropriate without departing from the scope of the present invention. Can be implemented.

This specification includes the entire specification of Japanese Patent Application No. 2005-63709, which is the basis for claiming priority of the present application. In addition, all prior art documents cited in the present specification, as well as published gazettes, patent gazettes and other patent documents are incorporated herein by reference. Be incorporated

1. Summary of the present invention

 The present invention is characterized in that various parameters in texture analysis, which is an image processing technique, are used for monitoring cultured cells over time. In monitoring by texture analysis, it is sufficient to take an image of the cultured cells, so that the culture state can be easily confirmed each time in a short time. As a result, even when collective monitoring of a large number of cultured cell systems is necessary, for example, it becomes much easier than before. Moreover, since it does not come into direct contact with the cultured cells as in the case of sampling or the like, infection cannot cause contamination.

 In general, it is known that texture analysis includes image analysis methods using a co-occurrence matrix, difference statistics, and density histograms. The spatial density level-dependent method (SGLDM), density level, It is called the difference method (GLDM) or the concentration histogram method (GLHM). These are calculated for a given parameter that is the feature value of the texture for each method.

 In the present invention, it is further important that the texture analysis is performed using a parameter related to the culturing time of the cultured cells as an index. As such parameters, an optimum one can be selected and adopted as appropriate from among the predetermined parameters as the texture feature values described above. By using parameters related to the elapsed culture time as an index, it is easy to make a reasonable judgment and confirmation according to the actual elapsed time as to whether or not the state of the cultured cells is normal. Further, in the present invention, the parameters used as the above-mentioned indices can be generally adopted for each type of cultured cell (one type or two or more types). Without being limited thereto, it can be said that the usefulness of the present invention is very high in this respect.

Furthermore, the present invention shows extremely high utility especially in the field of regenerative medicine. In the field of regenerative medicine, several tens to several thousand equivalent to the number of patients It is considered inevitable that there is a need to manage a large number of cultured cell lines at the same time, and there are changes in the number and morphology of cells, the state of proliferation and differentiation, and the presence or absence of defective cells compared to normal cell culture. It is highly necessary to manage many factors such as these simultaneously and with a certain level of reliability. Under these circumstances, the present invention can be said to satisfy these needs all at once, and is extremely useful and exhibits a remarkable effect.

2. Cultured cell monitoring system

 As described above, the cultured cell monitoring system of the present invention has a means for capturing an image of a cultured cell (hereinafter referred to as “imaging means”) and a means for analyzing a captured cell image (hereinafter referred to as “imaging means”). Analysis system ”), and the texture analysis is performed using a parameter related to the culture elapsed time of the cultured cell as an index. (1) Cultured cells

 In the cultured cell monitoring system of the present invention, the cultured cells to be monitored are not limited. For example, animal cells, plant cells, insect cells and the like can be used as monitoring targets in the present invention. preferable.

 Examples of animal cells include, but are not limited to, human-derived cells, monkey-derived cells (such as monkey kidney-derived cells (Cos cells (Cosl cells))), mouse-derived cells (mouse vascular endothelial cells (mouse vascular endothelial cells ( Pro5 cells), etc.), and pig-derived cells.

 The cells derived from humans are not limited, and for example, various tissue cells (oral mucosal cells, adipocytes, chondrocytes, etc.), and tumor cells (benign and malignant cancer cells (derived from human digestive organ cancer) Cells (HCT15 cells) and the like), etc., but cells for regenerative medicine are preferred in that the effect of the present invention can be sufficiently exerted. For animal cells derived from other than humans, cells for regenerative medicine can be preferably used for the same reason.

Examples of cells for regenerative medicine include various pluripotent stem cells (specifically, all Preferred stem cells and tissue stem cells), but not limited thereto. Totipotent stem cells are cells that can differentiate into all somatic cells, and examples include, but are not limited to, ES cells (embryonic stem cells / embryonic stem cells) and EG cells.

 A tissue stem cell is a cell that can differentiate into a plurality of cell types constituting a specific organ or tissue, and examples include neural stem cells, hematopoietic stem cells (bone marrow cells, etc.), and fat cells. There is no limitation.

 The cultured cells to be monitored are not limited as long as they are cultured in a well-known culture container as appropriate according to the type of the cultured cells. Is preferably transparent (preferably colorless and transparent). Examples of the culture container include a plastic petri dish, a plastic flask, a glass petri dish, and the like. In addition, the culture vessel containing cells is generally placed in a known incubator (incubator) that is temperature-controlled and cultured.

(2) Photography method

 The means for taking an image of the cultured cell is not limited, but at least a device that can magnify the cultured cell (device (a)) and a device that can take a still image of the cultured cell (device (b) ), And equipment (equipment (c)) that can construct image data (electronic data) based on a photographed still image that can be used in the analysis means described later. These devices (devices (a) to (c)) may be devices in which all or any of the two forces are integrated (devices having substantially the functions of each device), or at least The two may be separate devices and is not limited.

 Examples of the device (a) include, but are not limited to, various microscopes such as an inverted microscope, an optical microscope, and a laser microscope.

The device (b) is not limited, and examples thereof include various cameras. The device (C) is not limited, and examples thereof include a scanner.

 In addition, the device having the functions of the devices (b) and (C) is not limited, and examples thereof include a digital camera (including digital video with a camera function).

 The photographing means may further include other devices such as an autofocus device and an automatic exposure time measurement device.

 The photographing means (part or all) may be arranged inside or outside the incubator (incubator), and is not limited. In addition, a plurality of imaging means (part or all of them, in particular, the device (a)) may be provided in combination with the number of culture vessels (number of samples), or a plurality of imaging means may be moved to The culture vessel may be monitored and is not limited.

 The imaging interval of the cultured cells by the imaging means (especially the device (b)) may be set as appropriate depending on the type of the cultured cells to be monitored, and is not limited. For example, the interval is 1 day from the first day of the culture start. It may be 1 hour interval or 1 minute interval. (3) Analysis method

 The means for texture analysis of the photographed cultured cell image is not limited, and examples thereof include a computer that can perform texture analysis on the image data obtained by the photographing means. Existing software can be used for texture analysis, for example, Poplmaging (Digital 'Being. Kids').

Note that the computer may also control other means such as, for example, control of photographing conditions and position of an electronic device (photographing means) such as a digital camera. Existing software can be used to control the digital camera. Examples include Aquacosmos (manufactured by Hamamatsu Photonics), NIH Image (manufactured by NIH), and Scion Image (manufactured by Scion Corporation). As described above, there are three methods for texture analysis (spatial density level-dependent method, density level difference method, and density histogram method). These methods are described in detail below.

 In general, the spatial density level-dependent method is the probability P (i, j) (i, j = 0, l, In this method, a co-occurrence matrix with elements 2, ',', and n-1) is obtained, and the texture features are obtained from the matrix. In this method, texture parameters are calculated for five parameters: “energy, entropy, correlation, local uniformity, and inertia”. As for the calculation method (calculation formula) of each parameter, a known method is adopted. In general, the density level difference method is based on the probability P (k) (k = 0, l, 2, “”, nl) that the density difference of a pixel separated by a distance d in the 0 direction from a pixel is k. In this method, the texture features are “contrast, second moment by angle, entropy, average, inverse difference moment”. Calculated for two parameters. For the calculation method (calculation formula) of each parameter, a known method is adopted. Natural logarithm (base = e) is used for log in entropy calculation.

 In general, the density histogram method is a method for obtaining texture features from a density histogram P (i) normalized so that the whole becomes 1. In this method, the texture features are calculated for the four parameters “average, variance, skewness, and kurtosis”. Note that the skewness is a parameter indicating the symmetry shape characteristic of the density histogram, that is, how much the density histogram has deviated from the symmetrical shape. The kurtosis is the distribution shape characteristic of the density histogram, In other words, it is a parameter that represents how much the concentration histogram distribution is concentrated around the average value. As a calculation method (calculation formula) of each parameter, a known method is adopted.

In texture analysis, the image data obtained by the photographing means is generally subdivided into equal areas (divided into squares), and the parameter values as described above are measured for each square. , Not limited to this, The entire image data may be measured.

 In the texture analysis according to the present invention, it is necessary to use parameters that are effective for monitoring cultured cells over time as an index. Specifically, as described above, parameters related to the elapsed time of cultured cells. It is important to be carried out using as an index. Here, “to be performed using parameters related to the elapsed time of cultured cells as an index” generally means that measurement of parameters in texture analysis is performed only for parameters that can be used as the above indexes from the beginning (this However, the present invention is not limited to this. For example, various parameters can be measured over time, and the parameters that can be used as the index from the measurement results. Is also included to select appropriately.

 In the present invention, the parameter related to the culture elapsed time of the cultured cell is substantially equal to the parameter value measured over time (Y axis) plotted against the culture elapsed time (X axis). However, there is no limitation, for example, the measured value (Y axis) is substantially constant with respect to the culture elapsed time (X axis). Preferably, a parameter showing an increasing or decreasing tendency (proportional relationship) at a ratio of. In this case (for example, when a proportional relationship is shown), a parameter in which the increase and decrease in measured values are observed alternately with respect to the culture elapsed time, a parameter in which the increase or decrease is not constant, or a parameter in which no substantial change is observed Etc. will not be included in the parameters related to the culture elapsed time of the cultured cells. In general, the parameter is unique depending on the type of cultured cells, and is not limited to a specific type. Further, it may be one type or two or more types and is not limited. (4) Other means

The cultured cell monitoring system of the present invention may include other means in addition to the imaging means and analysis means, and is not limited. For example, there is a means that can notify when there is any abnormality such as abnormal culture state. These may be provided alone or in combination of two or more, and are not limited.

 The means for notifying the abnormality is not limited, but examples thereof include a device such as an alarm device. Specifically, it is preferable to operate by sending a signal from a computer used for the analyzing means.

(5) Usage of monitoring system

 For the arrangement of each means in the cultured cell monitoring system of the present invention, the schematic diagram shown in FIG. 1 can be referred to.

 The usage mode of the cultured cell monitoring system of the present invention is not limited, but for example, the following modes ω and (ϋ) are preferable.

(i) A mode in which parameters to be used as an index for texture analysis are selected in advance In this usage mode, prior to the use of the system of the present invention, a preliminary parameter for selecting parameters that are effective for monitoring the target cultured cells. do an experiment. In the preliminary experiment, target cells are cultured in a normal culture state (a desired culture progress state to be actually realized), and the values of various parameters are measured over time. After plotting the measured value (Y axis) for each parameter against the culture elapsed time (X axis), a graph (line graph) was created. Judge whether it is appropriate as a parameter related to time, and select one or more parameters specific to the cell.

After that, culture of cells to be monitored (1 sample or 2 samples or more) is started, and using the system of the present invention, the value of each parameter selected above is measured along the course of culture for each sample ( Measure over time). Regarding the measured values, when plotted in the same manner as in the preliminary experiment, samples that showed a value significantly deviating from the graph created in the preliminary experimental stage for at least one parameter were measured in the normal culture state. It can be determined that the deviation from the above (the desired culture progress state was not obtained). (ii) A mode in which parameters that serve as indices for texture analysis are selected while performing the analysis

 In this mode of use, parameters that are effective for monitoring the target culture cells are selected while using the system of the present invention, so that it is not necessary to conduct preliminary experiments as in the mode of (i) above. Highly useful in terms of time and cost. It can also be used for research and development of new parameters. In this embodiment, the number of samples of cells to be monitored needs to be at least 3 or more, and is preferably 10 or more, more preferably, in order to improve reliability. 20 or more, particularly preferably 100 or more.

 Culture of cells to be monitored (three or more samples) is started under conditions that allow the desired culture progress, and various parameters in each sample are measured over time using the system of the present invention. On the other hand, along with the progress of the culture, plotting the measured value (Y axis) for each parameter with respect to the culture elapsed time (X axis) to create a graph (line graph) in parallel. As a result of comparing the above graphs between each sample at each time measurement, the behavior of the measured value with respect to the culture elapsed time is more than half (for example, 1/2 or more, preferably 2/3 or more, more preferably 19/20 or more More preferably 99/100 or more), a common parameter is recognized as one of the above-mentioned “parameters related to the culturing time of cultured cells”, Two or more parameters can be selected. After selection, measure various parameters and create a graph in the same way. (You may limit to the selected parameters.) For at least one of the selected parameters, the other samples in the majority sample above Samples that show a value significantly different from this graph can be judged to have deviated from the normal culture state (the desired culture progress state was not obtained).

(6) Influence of shooting conditions In general, the optimal imaging conditions differ depending on the type of cultured cells, but automatic analysis of many types of cells becomes difficult if the analysis results vary greatly depending on the imaging conditions. Therefore, when using the cultured cell monitoring system of the present invention, it is preferable to select in advance parameters that are not influenced by the imaging conditions or have little influence. The imaging conditions that should be considered are not limited, but examples include exposure time and focus.

3. Cultured cell monitoring method

 As described above, the cultured cell monitoring method of the present invention includes a step of capturing an image of a cultured cell (hereinafter referred to as “imaging step”) and a step of analyzing a texture of the captured cell image (hereinafter referred to as “analysis step”). And the texture analysis is performed using a parameter related to the culture elapsed time of the cultured cells as an index. (1) Cultured cells

 The cultured cells to be monitored in the cultured cell monitoring method of the present invention are not limited, but the same explanation as the system of the present invention can be preferably applied. (2) Shooting process

The process of taking an image of the cultured cell is not limited, but at least a device capable of magnifying the cultured cell (device (a)), a device capable of photographing a still image of the cultured cell (device (b)), and This can be done using equipment (equipment (c)) that can construct image data (electronic data) based on the captured still image that can be used in the analysis means described later. These devices (devices (a) to (c)) may be devices in which all or any two of them are integrated (devices having substantially the functions of each device), or at least Both can be devices that are separate and are not limited. For examples of the above devices and the imaging interval of the cultured cells, The same explanation as the system can be preferably applied.

(3) Analysis process

 The process for texture analysis of captured cultured cell images is not limited. For example, texture analysis can be performed on the image data obtained by the imaging process (with software for texture analysis). Can be carried out using the data.

 As described above, there are three methods for texture analysis (spatial density level dependent method, density level difference method, and density histogram method). The same description as the system of the present invention can be preferably applied to the details of these methods and the explanation regarding the segmentation of image data. In addition, “texture analysis is performed using the parameters effective for monitoring the cultured cells over time (specifically, parameters related to the elapsed culture time of the cultured cells) as an index, which is a feature of the method of the present invention. The description similar to that of the system of the present invention can also be preferably applied to the explanation regarding the parameters.

(4) Other processes

 The cultured cell monitoring method of the present invention may include other steps in addition to the imaging step and the analysis step, and is not limited.

(5) Usage of monitoring method

 The usage mode of the cultured cell monitoring method of the present invention is not limited, but the same examples and explanations as those of the system of the present invention can be preferably applied. (6) Influence of shooting conditions

As for the influence of imaging conditions when performing the cultured cell monitoring method of the present invention, the same illustrations and explanations as those of the system of the present invention can be preferably applied. Hereinafter, the present invention will be described more specifically with reference to examples and the like. However, the present invention is not limited to these examples.

Example 1

Embodiment in which effective parameters are selected in advance as an index for texture analysis Preliminary experiment 1: Influence of shooting conditions>

 Of the parameters in texture analysis, parameters that were not affected by the shooting conditions (exposure time) were selected.

 First, a cultured cell monitoring system as shown in the schematic diagram of Fig. 1 was prepared. Specifically, as an imaging method, an inverted microscope (manufactured by Olympus, product name: IX 71), and a digital camera (manufactured by Hamamatsu Photonitas, product name: ORCA-) connected so as to be able to photograph images with the microscope are used. ER), and the analysis method is a computer that can analyze the image data (Epson, product name: Endeavor), and image analysis software (Digital Bi-Fung Kids, product name, : Poplmaging), and the above imaging means can be placed in an incubator where the temperature can be set and adjusted as required. In the preliminary experiments, monitoring examples, and examples shown below, culture and texture analysis were conducted using the above monitoring system, although not particularly mentioned.

 Next, for the cultured cells (Pro5 cells) cultured for 24 hours, using the above monitoring system, images taken with the digital camera exposure time changed to 244ms (milliseconds), 365ms, 547ms, and 730ms (Fig. 2A) Each was texture analyzed. The obtained analysis results (parameter values) were plotted at each exposure time when the value at 244 ms was 1. The graph is shown in Figure 2B.

From the graph above, the parameters of contrast, angular momentum, entropy and average in GLDM, and enthalpy, correlation, and local uniformity in SGLDM are greatly affected by differences in exposure time. Therefore, it was considered to be a suitable parameter for evaluating the state of cultured cells. Preliminary Experiment 2: Cell density monitoring>

 By dividing the photograph of the cultured cells into small areas and analyzing the texture, it is possible to determine how much the cells cover the bottom of the culture vessel. As shown in Fig. 3A, the photograph was divided into 16 equal-area areas, and for each, the GLDM contrast was analyzed as a parameter. As a result, the values shown in Figure 3B were obtained for each area. Based on the magnitude of this value, each area is represented by shading as shown in Figure 3C. As a result, it was shown that texture analysis is as effective as the naked eye in determining the cell density in a culture vessel and its variation.

<Preliminary experiment 3: Detection of culture trouble>

 Pro5 cells were intentionally changed to a cultured state (abnormality: trouble) by removing serum from the culture medium during the culture, and the changes were monitored over time by texture analysis and evaluated.

 Cells at a certain point in the culture (this time is designated as “0 hour”) when the culture medium was replaced with a serum-free medium (non-control group), and cells cultured in a culture medium containing normal serum from beginning to end For the control group, texture analysis was performed over time using the GLDM contrast as a parameter, and the parameter values were graphed for comparison. This examined the effect of switching to a serum-free medium.

As a result, after 1 hour, there was already a large difference in parameter values between the control group and the non-control group (Fig. 4A). Here, as seen in the photograph in Fig. 4B, the number of cells in the serum-free medium (non-control group) was significantly different between the time point of 0 hours (before serum removal) and the time point of 1 hour. However, it was recognized that the cell morphology had changed. Therefore, it can be said that this morphological change was detected by texture analysis. Note that the difference in parameter values after 6 hours is considered to be due to the difference in the number of cells. Preliminary Experiment 4: Selection of Effective Parameters as an Indicator for Monitoring Cultured Cells> (1) Pro5 cells

 Pro5 cells were seeded, and photographs were taken every day from the next day, and SGLDM entropy, SGLDM correlation, SGLDM local uniformity, GLDM contrast, GLDM angular secondary moment, GLDM entropy, and GLDM average were evaluated. The results (graph) are shown in Fig. 5A.

 From this graph, the parameters related to the elapsed time of cultured cells (those that serve as an index for monitoring the culture state) include SGLDM entropy, SGLDM local uniformity, GLDM contrast, GLDM second moment, and GLDM. You can select the pea pea and the GLDM average.

(2) COS cells

 COS cells were seeded, and photographs were taken every day from the next day, and SGLDM entropy, SGLDM correlation, SGLDM local uniformity, GLDM contrast, GLDM angular momentum, GLDM entropy, and GLDM average were evaluated. The results (graph) are shown in Fig. 5B.

 From this graph, the parameters related to the elapsed time of the cultured cells (those that serve as an index for monitoring the culture state) are: SGLDM inlet mouthpiece, SGLDM local uniformity, GLDM contrast, GLDM second moment by angle , GLDM entry, and GLDM average can be selected.

(3) HCT15 cells

 HCT15 cells were seeded, and photographs were taken every day from the next day, and SGLDM entropy, SGLDM correlation, SGLDM local uniformity, GLDM contrast, GLDM angular second moment, GLDM entropy, and GLDM average were evaluated. The results (graph) are shown in Fig. 5C.

From the graph, the parameters related to the elapsed time of cultured cells (those that serve as an index for monitoring the culture state) include SGLDM entropy, SGLDM local uniformity, GLDM contrast, and GLDM second moment by angle. , GLDM Mouth pea, GLDM average can be selected

<Preliminary experiment 5: Differentiation of cell types>

 Based on the results of Preliminary Experiment 4 (5th day of culture), each parameter of SGLDM method (entropy, local uniformity, GLDM contrast, angle 2) for Pro5 cells, COS cells, and HCT15 cells. Next moment, entropy and average) were measured, and the values in Pro5 cells were taken as standard values (1), and the characteristics of each cell type were compared. The result is shown in FIG.

 Specifically, for each parameter, the converted value when the value in Pro5 cells is 1 is shown. COS cells have a distinctly different profile from the other two cells, and it was found that cell types can be distinguished to some extent. Preliminary Experiment 6: Identification of mixed cell types>

 When multiple types of cells coexist in the culture vessel, we experimented with what the texture analysis would show. COS cells and Pro5 cells were lightly seeded in the culture vessel, and photographs were taken when the colonies were formed 5 days later (Fig. 7A). The photograph is divided into 16 areas for the time being, and texture analysis is performed for each area. (Parameters include contrast in GLDM, second moments by angle, entropy and average, and entropy in SGLDM. Local uniformity) and the obtained parameter values are shown (Fig. 7B).

 The 16 numbers corresponding to each area were color-coded by larger and smaller numbers using the numerical value that can be divided into 8 as the boundary value (Figure 7B). As shown in the photograph in Fig. 7A, the upper right part is occupied by Pro5 cell colonies and the lower left part is occupied by COS cell colonies. At the boundary, an area where both cells coexist diagonally from the upper left to the lower right. The results of texture analysis were divided into the following three patterns.

(1) The contrast of GLDM method and the average of GLDM method showed different values depending on the cell type. Both indicators can detect the presence of two cell types. It is thought.

 (2) The local uniformity of the SGLDM method and the ENTDM of the GLDM method

 A small value was shown in the mixed area.

 (3) Both the GLDM method and the SGLDM method

 A large value was shown in the mixed cell area.

 From the above analysis, it was shown that the same result as that of the naked eye observation can be obtained for the existence of multiple types of cells, their areas, and their overlap.

<Monitoring example 1: Monitoring the culture state of Pro5 cells (Detection of incubator failure using GLDM contrast)>

 Pro5 cells were seeded in a culture dish, and observation was started after 1 day. Since it is known from the analysis of the normal group that the contrast of the GLDM method is suitable for the evaluation of cell proliferation as described above (see Preliminary Experiment 4 (1)), the culture state is monitored using the contrast as an index. Went. As a sample, after 2 days of culturing, the incubator was turned off to create a situation where the cell culture state was intentionally abnormal.

 In the analysis after 3 days of culture, the sample contrast clearly deviated from the known normal range (Fig. 5A), confirming that the culture state was abnormal. The results (graph) are shown in Fig. 8. Each value in the graph is expressed as “average soil SD J”.

Example 2

Embodiment in which effective parameters are selected as indices while performing texture analysis

Cosl cells were seeded in 6 culture dishes, and the culture was started. After 2 days of observation, 10ng / ml epidermal growth factor was added to 3 cultures (non-control group). The changes were compared with the remaining 3 sheets (control group) to which no epidermal growth factor was added. A photograph is taken and observed every day after the addition, and based on the photograph (image data), “GLDM contrast, GLDM second moment, The culture state was monitored by texture analysis using GLDM entropy, GLDM average, SGLDM entropy, and SGLDM local uniformity as indices.

 The photographs on the second day after the addition (that is, the fourth day of culture) are shown in FIGS. 9A and 9B. Fig. 9A shows cells in the control group, and Fig. 9B shows cells to which epidermal growth factor has been added (non-control group cells). The cells in the non-control group clearly changed in cell shape 2 days after addition, indicating that the widened fibroblast-like shape has increased in cell thickness and differentiated into epithelial-like cells.

 On the other hand, with regard to the results of monitoring over time by texture analysis, when using “GLDM angular second moment, GLDM entropy, SGLDM entropy, SGLDM local uniformity” as an index, the control group or the non-control group Regardless of whether or not it was, there was no noticeable tendency to change after the addition, but when `` GLDM contrast, GLDM average '' was used as an index, the control group showed stable changes over time, In the non-control group, there was a noticeable variation trend. The results when “GLDM contrast” and “GLDM average” are used as indices are shown in FIG. 9C (GLDM contrast) and FIG. 9D (GLDM average). In both graphs of FIG. 9C and FIG. 9D, the solid line graph (3 types) represents the control group, and the dotted line graph (3 types) represents the non-control group. The arrow indicates the time point when epidermal growth factor was added to the cells in the non-control group (after 2 days in culture).

 Based on the above results, it is clear that “GLDM contrast” and “GLDM average” are effective parameters for monitoring Cosl cell culture conditions (particularly cell morphology) over time using texture analysis. At the same time, it was also demonstrated that the above-mentioned parameters that can confirm cell differentiation are also effective. Industrial applicability

 According to the present invention, the culture state of the target cell over time can be confirmed easily and in a short time, and there is no risk of infection or contamination of the cultured cell at the time of confirmation. It is possible to provide a monitoring system.

Claims

 Scope of request Culture with means for taking images of cultured cells and means for texture analysis of the taken images of cells using the parameters related to the culture elapsed time as an index Cell monitoring system. The system according to claim 1, wherein the cultured cells are human-derived cells. The system according to claim 2, wherein the human-derived cell is a cell for regenerative medicine. A cultured cell monitoring method, comprising: a step of photographing an image of a cultured cell; and a step of performing texture analysis of the photographed cell image using a parameter related to the culture cell culturing time as an index.