Method for evaluating cell function and cell analysis device

Abstract

An embodiment of the present invention is a method for evaluating a cell function of a target cell that is a stem cell, the method comprising: an image acquisition step (S1, S2) for acquiring a photographed cell image of a target cell; a feature value extraction step (S5-S8) for extracting from the cell image, as a feature value of the target cell, information concerning the gradient of the contour part of the target cell and / or information concerning the orientation of the target cell; and an evaluation step (S9, S10) for evaluating, on the basis of the feature value of the target cell, a cell function of the target cell. This enables a quantitative evaluation of a cell function, e.g., retention of undifferentiated character, migration capability, tissue repair capability, and so forth, of stem cells.

Description

[Technical field]

[0001] The present invention relates to a method and an apparatus for evaluating cell function, and more particularly to a method and an apparatus for non-invasively evaluating stem cell function. [Background technology]

[0002] In recent years, research and industrialization of regenerative medicine and cell therapy have progressed rapidly, and as a result, there is a demand for the establishment of evaluation methods to accurately ensure the quality, safety, and efficacy of cell preparations. Since cell preparations are derived from animals, particularly humans, there is a wide variety of origins of tissue from which cells are collected, types of stem cells, and manufacturing methods. Therefore, there is a demand for the establishment of flexible, widely applicable, reproducible, and rational evaluation methods based on the characteristics of cell preparations.

[0003] Conventionally, well-known methods for evaluating cell functions include analysis of specific genes and proteins expressed in cells, analysis of cell migration and proliferation, analysis of factors secreted by cells, and analysis of cell aging. These cell traits and functions are thought to correlate with the therapeutic effects of cell preparations. However, when considering that the cells used for evaluation will be used for another purpose as is, for example, administered to patients as a regenerative medicine product, evaluation methods that require invasive or destructive treatment of the cells cannot be used.

[0004] As a method for non-invasively evaluating cells, there is a method of extracting various features from still images or videos of cells and evaluating the state of the cells from the features. For example, Patent Document 1 describes a cell evaluation device that judges the quality of the state of cells by comparing various features obtained from cell images with predetermined thresholds. In this device, a "good state" for cells is, for example, a "state in which cells grow well," a "state in which cells differentiate well," a "state in which there are many undifferentiated cells," a "state in which differentiation is difficult to occur," or a "state in which cells are easy to select (easily clone)," depending on the purpose.

[0005] Meanwhile, as another method for non-invasively evaluating cell images, a method for evaluating the differentiation state of cells by analyzing secretory factors derived from cells contained in the supernatant of cultured cells is also known. For example, Patent Document 2 discloses a method for evaluating the differentiation state of cells from the abundance of a specific component contained in the culture supernatant. According to this method, by comparing the abundance of a specific component in a reference cell and a test cell, the state of the cell can be evaluated without tracking changes over time. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent No. 6448129 [Patent Document 2] Patent No. 6332440 [Patent Document 3] International Patent Publication No. 2016 / 084420 [Patent Document 4] Patent No. 3471556 [Non-patent literature]

[0007] [Non-Patent Document 1] Nagaishi, et al., "Umbilical cord extracts improve diabetic abnormalities in bone marrow-derived mesenchymal stem cells and increase their therapeutic effects on diabetic nephropathy," Scientific Reports, 2017, 7(1) Summary of the Invention [Problem to be solved by the invention]

[0008] However, while known non-invasive cell evaluation methods can quantitatively evaluate the state of cells in culture, they cannot evaluate the specific functions of cells. Therefore, conventional evaluation of general cell functions was limited to the level of intuitive judgment by personnel with experience and knowledge in analysis based on cell observation images and evaluation results of cell states. In such evaluation methods, the reliability and reproducibility of the evaluation results are heavily dependent on the experience and knowledge of the personnel. In addition, it is difficult to verify whether such evaluations were appropriate.

[0009] The present invention has been made to solve these conventional problems, and its main objective is to provide a method for evaluating cell function and a cell analysis device that can non-invasively and quantitatively evaluate the function and capacity of stem cells based on image information obtained from the stem cells. [Means for solving the problem]

[0010] One aspect of the method for evaluating a cell function according to the present invention, which has been made to solve the above problems, is a method for evaluating a cell function of a test cell that is a stem cell, comprising the steps of: The test cells were photographed , including information reflecting the thickness of the test cell. An image acquisition step of acquiring a cell image; From the cell image, a contour of the test cell is In the thickness direction Information about the gradient of News a feature extraction step of extracting a feature of the test cell; an evaluation step of evaluating a cellular function of the test cell based on the feature amount of the test cell; It has the following.

[0011] One aspect of the cell analysis device according to the present invention, which has been made to solve the above problems, is a cell analysis device for evaluating a cellular function of a test cell which is a stem cell, comprising: The test cells were photographed , including information reflecting the thickness of the test cell. an image acquisition unit for acquiring a cell image; From the cell image, a contour of the test cell is In the thickness direction Information about the gradient of News a feature extraction unit that extracts a feature of the test cell; an evaluation processing unit that relatively evaluates a cellular function of the test cell based on the feature amount of the test cell; It is equipped with the following.

[0012] The "cell function" referred to here does not refer to the state of the cell itself at that time, but rather to the abilities, performance, or properties that the cell possesses, particularly functions, performance, and properties that are useful or adverse when the cell is used in regenerative medicine or cell therapy, or when the cell is cultured and proliferated for such use. Specifically, the cell function can include at least one of the cell functions, such as maintenance of undifferentiated state, migration ability, tissue repair ability, angiogenesis ability, and immune control ability. Effect of the Invention

[0013] The inventors have repeatedly conducted various experiments and analyses, and have found that the results of analysis of gene expression levels related to various cell functions, such as maintenance of undifferentiated state of stem cells, migration, tissue repair ability, angiogenesis, and immune control, or the results of measurement of cell migration ability, are highly correlated with the gradient in the thickness direction at the outline (periphery) of stem cells and the orientation (alignment of the direction) of stem cells, which are obtained based on observation images of stem cells obtained non-invasively. Based on this finding, in the feature extraction step of the cell function evaluation method, which is one aspect of the present invention, from the observation images of stem cells, either one or both of an index value indicating the degree of gradient in the thickness direction at the outline of the test cell and an index value indicating the degree of orientation of the test cell are extracted as the feature of the test cell. Then, in the evaluation step, the cell function of the test cell is evaluated based on the index value, which is the feature of the test cell. Specifically, for example, the numerical values ​​of the above index values ​​are compared for two types of test cells of different origins, and a judgment is made as to which cell has a higher cell function. In addition, in the cell analysis device according to one aspect of the present invention, the processes in the feature extraction step and evaluation step can be performed on a computer including a CPU, a RAM, and the like.

[0014] According to the cell function evaluation method and cell analysis device of one embodiment of the present invention, the function of stem cells can be quantitatively evaluated based on image information obtained non-invasively and non-destructively from the stem cells. Since such functional evaluation can be performed using numerical values, even a person with little experience or knowledge in cell analysis or evaluation can perform an accurate and reliable evaluation. In addition, the variation in results between individuals, which is common in evaluations that rely on intuition, can be avoided, and the verification of the reasons for the evaluation, etc. can be facilitated. In addition, the efficiency of the evaluation of cell function can be improved. [Brief description of the drawings]

[0015] [Figure 1] 1 is a schematic diagram of an embodiment of a cell analysis device for carrying out a method for evaluating a cell function according to the present invention. [Diagram 2]4 is a flowchart showing the procedure and processing flow when evaluating cell functions using the cell analysis device of the present embodiment. [Diagram 3] 3A and 3B are conceptual diagrams for explaining the imaging operation and image reconstruction processing in the cell analysis device of the present embodiment. [Figure 4] 1 is a schematic flowchart of a process for calculating the orientation score of a cell. [Diagram 5] 13 is a detailed flowchart of a portion of the process for calculating the orientation score of a cell. [Figure 6] 5A and 5B are diagrams showing an example of a differential filter used in a differential process in the cell analysis apparatus of the present embodiment. [Figure 7] 6A and 6B are diagrams showing other examples of differential filters used in the differential processing in the cell analysis device of the present embodiment. [Figure 8] 1A to 1C are diagrams showing an example of an IHM phase image created by the cell analysis device of the present embodiment and a differential image obtained therefrom. [Figure 9] FIG. 4 is an explanatory diagram of an example of a method for calculating a gradient score from a differential value histogram in the cell analysis device of the present embodiment. [Figure 10] 10 is an explanatory diagram of another example of the method for calculating a gradient score from a differential value histogram in the cell analysis device of the present embodiment. FIG. [Figure 11] 10 is an explanatory diagram of another example of the method for calculating a gradient score from a differential value histogram in the cell analysis device of the present embodiment. FIG. [Figure 12] 10 is an explanatory diagram of another example of the method for calculating a gradient score from a differential value histogram in the cell analysis device of the present embodiment. FIG. [Figure 13] FIG. 4 is an explanatory diagram of a method for calculating an orientation score in the cell analysis device of the present embodiment. [Figure 14] 4A to 4C are diagrams showing examples of images for evaluating the orientation of two types of samples in the cell analysis device of the present embodiment. [Figure 15] FIG. 4 is a diagram showing a comparative example of orientation scores for two types of samples in the cell analysis device of the present embodiment. [Figure 16] 13 shows optical microscope images of cell tissue to explain the collection site of umbilical cord-derived mesenchymal stem cells in the experimental example. [Figure 17] FIG. 1 shows the results of an analysis of the expression of cell surface antigens for four types of mesenchymal stem cells derived from one sample A. [Figure 18] FIG. 1 shows the quantitative results of gene expression levels related to various cell functions for four types of mesenchymal stem cells derived from sample A. [Figure 19] FIG. 1 shows the results of measuring the migration ability of four types of mesenchymal stem cells derived from sample A using a scratch assay method. [Figure 20] FIG. 13 is a diagram showing a comparison of the orientation scores and gradient scores calculated by the cell analysis device of the present embodiment for four types of samples derived from specimen A. [Figure 21] FIG. 1 shows the results of analyzing the expression of cell surface antigens for four types of mesenchymal stem cells derived from sample B. [Figure 22] FIG. 1 shows the quantitative results of gene expression levels related to various cell functions for four types of mesenchymal stem cells derived from sample B. [Diagram 23] FIG. 1 shows the results of measuring the migration ability of four types of mesenchymal stem cells derived from sample B using a scratch assay method. [Figure 24] FIG. 13 is a diagram showing a comparison of the orientation scores and gradient scores calculated by the cell analysis device of this embodiment for four types of samples derived from specimen B. [Diagram 25] FIG. 1 shows the results of an analysis of the expression of cell surface antigens for four types of mesenchymal stem cells derived from sample C. [Figure 26] FIG. 1 shows the quantitative results of gene expression levels related to various cell functions for four types of mesenchymal stem cells derived from sample C. [Figure 27] FIG. 1 shows the results of measuring the migration ability of four types of mesenchymal stem cells derived from sample C using a scratch assay method. [Figure 28] FIG. 13 is a diagram showing a comparison of the orientation scores and gradient scores calculated by the cell analysis device of the present embodiment for four types of samples derived from specimen C. [Figure 29] FIG. 1 shows the results of analyzing the expression of cell surface antigens for four types of mesenchymal stem cells derived from sample D. [Diagram 30]FIG. 1 shows the quantitative results of gene expression levels related to various cell functions for four types of mesenchymal stem cells derived from sample D. [Diagram 31] FIG. 13 shows the results of measuring the migration ability of four types of mesenchymal stem cells derived from sample D using a scratch assay method. [Diagram 32] FIG. 13 is a diagram showing a comparison of the orientation scores and gradient scores calculated by the cell analysis device of this embodiment for four types of samples derived from specimen D. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Hereinafter, a method for evaluating a cell function according to one embodiment of the present invention will be described with reference to the accompanying drawings.

[0017] [Configuration of the cell analysis device according to this embodiment] FIG. 1 is a schematic diagram of one embodiment of a cell analysis device for evaluating cell functions according to the present invention.

[0018] The cell analysis device of this embodiment includes a microscope observation section 1, a control and processing section 2, and an input section 3 and a display section 4 which are user interfaces. In this embodiment, the microscopic observation unit 1 is an in-line holographic microscope, and is equipped with a light source unit 10 including a laser diode or the like, and an image sensor 11, and a culture plate 12 containing cells 13 is placed between the light source unit 10 and the image sensor 11.

[0019] The control / processing unit 2 controls the operation of the microscopic observation unit 1 and processes the data acquired by the microscopic observation unit 1, and includes the following functional blocks: an imaging control unit 20, a hologram data storage unit 21, a phase information calculation unit 22, an image reconstruction unit 23, a reconstructed image data storage unit 24, a differential image creation unit 25, a gradient score calculation unit 26, an orientation score calculation unit 27, and a cell function evaluation information output unit 28.

[0020] Usually, the entity of the control / processing unit 2 is a personal computer or a high-performance workstation in which a specific software (computer program) is installed, or a computer system including a high-performance computer connected to such a computer via a communication line. That is, the functions of each block included in the control / processing unit 2 can be realized by processing using various data stored in a computer or computer system, which is implemented by executing software installed in a single computer including a CPU, RAM, external storage device such as HDD or SDD, or a computer system including multiple computers.

[0021] The cell analysis device of this embodiment can observe and analyze various cells, but as an example, the observation subject is assumed to be a mesenchymal stem cell, which is one type of pluripotent stem cell. However, the evaluation of cell function as described below is not limited to mesenchymal stem cells, but can also be applied to other stem cells that exhibit the same or similar behavior as mesenchymal stem cells.

[0022] [Photographing cells and acquiring IHM phase images] Fig. 2 is a flow chart showing the procedure and processing flow in the evaluation of cell functions using the cell analysis device of this embodiment, and Fig. 3 is a conceptual diagram for explaining the imaging operation and image reconstruction processing.

[0023] FIG. 3(a) is a schematic top view of a culture plate 12 used in the cell analysis device of this embodiment. Six wells 12a, which are circular when viewed from above, are formed in the culture plate 12, and cells are cultured in each of the wells 12a. Here, the entire culture plate 12, that is, the entire rectangular area including the six wells 12a, is the observation target area. The microscope observation unit 1 includes four pairs of light source units 10 and image sensors 11. As shown in FIG. 3(a), each pair of light source units 10 and image sensors 11 is responsible for collecting hologram data of four quarter-divided areas 81 obtained by dividing the entire culture plate 12 into four equal parts. In other words, the four pairs of light source units 10 and image sensors 11 are responsible for collecting hologram data over the entire culture plate 12.

[0024] 3(b) and (c), the range that one set of light source units 10 and image sensors 11 can capture at one time is a range equivalent to one imaging unit 83 obtained by dividing a substantially square range 82 including one well 12a in a four-divided range 81 into 10 equal parts in the X-axis direction and 12 equal parts in the Y-axis direction. One four-divided range 81 includes 15×12=180 imaging units 83. The four light source units 10 and the four image sensors 11 are each disposed near the four vertices of a rectangle that is the same size as the four-divided range 81 within the XY plane including the light source units 10 and image sensors 11, and holograms are simultaneously obtained for four different imaging units 83 on the culture plate 12.

[0025] When collecting hologram data, the operator first sets the culture plate 12 in which the cells 13 to be observed are cultured at a predetermined position in the microscope observation unit 1, inputs information such as an identification number for identifying the culture plate 12 and the measurement date and time from the input unit 3, and then issues an instruction to perform measurement. Upon receiving this instruction, the imaging control unit 20 controls each part of the microscope observation unit 1 to perform imaging (step S1).

[0026] That is, one light source unit 10 irradiates a coherent light having a spread of a small angle (for example, about 10°) onto a predetermined area (one imaging unit 83) of the culture plate 12. The coherent light (object light 15) that has passed through the cells 13 on the culture plate 12 reaches the image sensor 11 while interfering with light (reference light 14) that has passed through an area (usually a culture medium) around the cells 13 on the culture plate 12. The object light 15 is light whose phase has changed when passing through the cells 13, while the reference light 14 is light that does not pass through the cells 13 and is not subject to a phase change caused by the cells 13. Therefore, on the detection plane (image plane) of the image sensor 11, an interference image between the object light 15 whose phase has been changed by the cells 13 and the reference light 14 whose phase has not been changed, that is, a hologram, is formed, and two-dimensional light intensity distribution data (hologram data) corresponding to this hologram is output from the image sensor 11.

[0027] As described above, the four light source units 10 emit coherent light toward the culture plate 12 at approximately the same time, and the four image sensors 11 acquire hologram data of areas corresponding to different imaging units 83 on the culture plate 12. Every time a measurement at one measurement position is completed, the light source units 10 and the image sensors 11 are moved stepwise in the X-axis direction and the Y-axis direction by a distance corresponding to one imaging unit 83 in the XY plane by a moving unit (not shown). As a result, measurements are performed at 180 imaging units 83 included in the four-division range 81, and the entire culture plate 12 is measured by the four sets of light source units 10 and image sensors 11. The hologram data obtained by the four image sensors 11 of the microscopic observation unit 1 in this manner is stored in the hologram data storage unit 21 together with attribute information such as the measurement date and time.

[0028] When the series of measurements (photographing) as described above is completed, the phase information calculation unit 22 sequentially reads out the hologram data from the hologram data storage unit 21, and performs a light wave propagation calculation process (phase recovery process) to restore the phase information and amplitude information at each two-dimensional position. The spatial distribution of this information is obtained for each imaging unit 83. When the phase information and amplitude information for all imaging units 83 have been obtained, the image reconstruction unit 23 forms a phase image of the entire observation region, that is, an IHM phase image, based on the phase information and amplitude information (step S2).

[0029] That is, the image reconstruction unit 23 reconstructs an IHM phase image of each imaging unit 83 based on the spatial distribution of phase information calculated for each imaging unit 83. Then, a tiling process (see FIG. 3(d)) is performed to join together the IHM phase images of the narrow range to form an IHM phase image of the observation target region, that is, the entire culture plate 12. The data constituting the IHM phase image is stored in the reconstructed image data storage unit 24. Note that, during the tiling process, it is advisable to perform an appropriate correction process so that the IHM phase images at the boundaries of the imaging units 83 are smoothly joined.

[0030] When calculating the phase information and reconstructing the image as described above, it is possible to use well-known algorithms disclosed in, for example, Patent Documents 3 and 4. In general, IHM phase images make it easier to see the outlines (boundaries) of cells and their internal patterns, which are transparent and difficult to see with an optical microscope.

[0031] In addition to the phase information, intensity information and pseudo-phase information that merges the phase information and intensity information can also be calculated based on the hologram data, and the image reconstruction unit 23 can create a reconstructed image (IHM intensity image, IHM pseudo-phase image) based on these.

[0032] Next, the image reconstruction unit 23 executes a process of removing background regions where it is clear that no cells exist from the created IHM phase image (step S3). There are various methods that can be used to remove the background, one example being texture analysis. Texture analysis can be broadly divided into structural texture analysis and statistical texture analysis, with the latter being more appropriate for extracting the features of patterns in localized areas within an image. There are various methods for statistical texture analysis, such as the density histogram method that uses a density histogram, the density level difference method that uses difference statistics, the spatial density level dependent method that uses a co-occurrence matrix, and the method that uses a density co-occurrence matrix, and an appropriate method can be selected. The pixel values ​​of the parts that have been removed as background regions can be set to a fixed value such as 0.

[0033] Specifically, the entire IHM phase image is divided into multiple small regions, and a predetermined texture analysis is performed for each small region to calculate the texture feature. Since there is a clear difference in the texture feature between a case where the small region is only a background region and a case where the small region includes at least a part of a cell region, it is sufficient to find and remove the small region that corresponds to the background region from the texture feature. Of course, such background removal processing is not essential and may be omitted.

[0034] Furthermore, the image reconstruction unit 23 performs a process of removing noise components from the IHM phase image after the background removal process using a Gaussian filter or the like (step S4). This noise removal process can also be omitted, similar to the above-mentioned background removal process. The above processing results in an IHM phase image with noise and background removed, meaning that the target cells can be clearly observed.

[0035] [Calculation of gradient score of cell contour based on IHM phase image] Then, based on the created IHM phase image, a gradient score is calculated that indexes the degree of gradient in the thickness direction at the cell contour. That is, the differential image creating unit 25 applies a differential filter to the signal value (pixel value) of each pixel of the IHM phase image after removing noise and the like as described above, and calculates a differential value for each pixel. Then, a differential image consisting of the differential values ​​of all the pixels is created (step S5). As the differential filter, for example, a Laplacian filter that is often used for edge detection in an image can be used.

[0036] FIG. 6 shows an example of a typical 3×3 pixel Laplacian filter using the four neighboring pixels on the top, bottom, left, and right. Also, a Sobel filter, which combines a vertical filter and a horizontal filter as shown in FIG. 7(a) and (b), may be used as a differential filter. Once the differential value is obtained for each pixel, it is used to create a differential image corresponding to the IHM phase image. The Sobel filter is also used when calculating the cell orientation score, which will be described later.

[0037] Figure 8 shows an example of an IHM phase image of human umbilical cord derived-mesenchymal stem cells and a differential image obtained from it. The signal value of each pixel in the IHM phase image indicates the phase delay of the light that passed through the cell, which reflects the optical thickness of the cell. Therefore, the steeper the gradient in the thickness direction of the contour of each cell, the larger the differential value of the pixel corresponding to that contour. In the differential image of Figure 8(b), the display brightness range has been appropriately adjusted, which makes the contour of each cell more clearly depicted compared to the interior side.

[0038] Based on the differential values ​​of all pixels constituting the differential image, the gradient score calculation unit 26 creates a differential value histogram in which the horizontal axis is the differential value and the vertical axis is the occurrence number (number of pixels) (step S6). In order to easily visually grasp the state of the change in the occurrence number relative to the increase in the differential value, it is preferable to make the horizontal axis of the histogram a linear axis and the vertical axis a logarithmic axis.

[0039] FIG. 9 shows an example of a differential histogram. A and B in the figure show two samples with different culture conditions (specifically, the presence or absence of an activator that activates cells). As shown in FIG. 9, the differential histogram shows an asymmetric peak with a steep slope on the low differential value side (left side in the figure) and a gentle slope on the high differential value side (right side in the figure). The pixels included in the range on the low differential value side including the peak top of this peak are considered to be pixels that exist mainly in background regions that are not cells or in parts of the cells that are relatively flat in thickness. On the other hand, the pixels included in the range with a gentle slope on the high differential value side are considered to be pixels that exist mainly on the outline of the cells. Therefore, the degree of inclination of the slope on the high differential value side in the differential histogram reflects the degree of the gradient in the thickness direction at the outline of the cells.

[0040] That is, among many cells appearing in the original IHM phase image, the greater the proportion of cells with a steep gradient in the thickness direction at the cell contour, the greater the proportion of pixels with relatively high differential values. The greater the proportion of pixels with relatively high differential values, the gentler the slope on the right side of the peak in the differential value histogram. In FIG. 9, the peak of sample 1 has a gentler slope than the peak of sample 2, and the greater the proportion of cells with a steep gradient in the thickness direction at the cell contour. Therefore, in the cell analysis device of this embodiment, an index value that allows the degree of inclination of the slope in different differential value histograms to be compared with each other is calculated as a gradient score.

[0041] Specifically, the gradient score calculation unit 26 first performs a smoothing process on the differential value histogram by taking a moving average in the direction of change of the differential value in order to reduce variations and errors. In the high differential value range where the difference in the gradient in the thickness direction at the cell outline appears prominently in the differential value histogram, the decreasing slope can be approximated by an exponential function. As described above, when the vertical axis is a logarithmic axis, the above slope becomes almost linear, so that the exponential function can be approximated by linear approximation on the graph. Therefore, here, two arbitrary differential values ​​are selected within a predetermined high differential value range, and the slope between the two differential values ​​is calculated as y=a·e -bx (where a and b are arbitrary constants) is used to approximate the slope. In other words, the constants a and b that minimize the approximation error for the slope should be searched for. The constant b of the exponent part found in this way reflects the gradient of the slope, which reflects the tendency of the gradient in the thickness direction at the contour of each cell, so this constant b of the exponent part should be used as the gradient score. In this case, the gradient score is not affected by the density of the cells, making it convenient for comparison between different samples.

[0042] [Another example of how gradient scores are calculated] The method for calculating the gradient score from the differential value histogram is not limited to the above method, and other methods such as those described below may also be used.

[0043] 10 to 12 are diagrams illustrating other examples of the method of calculating the gradient score. In Fig. 10, the half-width of the peak in the differential histogram is used as the gradient score. The half-width here means the width of the differential value at 1 / 2 the number of occurrences of the peak tops of the peaks. For example, in Fig. 10, the half-width Wa is obtained for the peak of sample 1, and the half-width Wb is obtained for the peak of sample 2. However, as described above, since only the slope on the high differential value side is meaningful here, the difference between the differential value at the peak top and the differential value corresponding to the point where the horizontal line indicating (the number of occurrences of peak tops) x (1 / 2) intersects with the slope on the high differential value side may be used instead of the half-width.

[0044] In Fig. 11, the gradient score is the rate of decrease in the number of occurrences on the decreasing slope of the peak on the differential value histogram. Specifically, in a predetermined high differential value range, any k number of differential values ​​(where k is an integer of 2 or more) are selected, and the ratio of the number of occurrences corresponding to the k number of differential values ​​is calculated as the gradient score. Generally, k can be 2 as shown in Fig. 11, and for two differential values ​​D1 and D2, the gradient score is calculated as the number of occurrences ratio Xa [%] for the peak of sample 1 and the number of occurrences ratio Xb [%] for the peak of sample 2.

[0045] In FIG. 12, the difference in differential values ​​corresponding to a certain difference in the number of occurrences is taken as the gradient score. Specifically, in the range of the number of occurrences corresponding to the slope of the high differential value range, any L number of occurrences (where L is an integer of 2 or more) is selected, and the difference in differential values ​​(the width of the differential value) corresponding to the L number of occurrences is calculated as the gradient score. Generally, as shown in FIG. 12, L may be 2, and for two numbers of occurrences P1 and P2, a differential value difference La is calculated for the peak of sample 1, and a differential value difference Lb is calculated for the peak of sample 2. Note that since the absolute value of the number of occurrences depends on the cell density, in order to eliminate the influence of differences in cell density, for example, a process of normalizing the number of occurrences may be performed.

[0046] Furthermore, according to the inventor's experimental study, it has been found that in the high differential value range in which the difference in the gradient in the thickness direction at the cell contour is conspicuous in the differential value histogram, the number of occurrences is proportional to the number of cells or the confluency (the ratio of the area of ​​the cells to the entire image area) in the particularly high differential value range. Therefore, when the number of occurrences in such a high differential value range is normalized by the number of cells or the confluency, the gradient score becomes independent of them.

[0047] Therefore, the confluency or cell count in the cell image (IHM phase image) to be observed is obtained by a measurement method different from, that is, independent of, the measurement method by this device. For example, cell counting using a hemocytometer can be used as the measurement method. Then, the number of occurrences at any differential value in the high differential value range in the differential value histogram is obtained, and this number of occurrences is divided by the cell number or confluency obtained as described above to calculate the normalized number of occurrences as the gradient score.

[0048] [Calculation of cell orientation score based on IHM phase images] Next, an orientation score is calculated by indexing the degree of alignment of the orientations of a large number of cells observed in the IHM phase image. Figures 3 and 4 are flowcharts showing the procedure and processing for calculating the orientation score from the IHM phase image, and Figure 13 is an explanatory diagram of the method for calculating the orientation score.

[0049] The orientation score calculation unit 27 reads the IHM phase image that has been subjected to noise removal and the like in step S4 (step S21), and divides this image into a number of rectangular small regions as shown in Fig. 13(a) (step S22). In Fig. 13(a), the number of divisions is 5 x 7 = 35, but this is just an example, and the size and number of the small regions may be set appropriately depending on the size of the cells in the phase image, the cell density, and the like.

[0050] Next, the orientation score calculation unit 27 calculates the strength and direction of the orientation of the cells included in each small region obtained by division (step S23). The process for one small region will be specifically described with reference to FIG.

[0051] The orientation score calculation unit 27 first reads the small region image (step S31) and performs noise removal processing (step S32). This may be the same processing as step S4, and can be omitted. After that, a contour detection processing is performed using a differential filter or the like on the signal value (pixel value) of each pixel. Here, the vertical contour detection using the vertical Sobel filter shown in FIG. 7(a) and the horizontal contour detection using the horizontal Sobel filter shown in FIG. 7(b) are performed for each pixel (steps S33, S34), and the strength and direction of the change in luminance in that pixel are calculated (steps S35, S36).

[0052] Specifically, when the signal value of a pixel, org(x, y), after processing by a vertical Sobel filter is denoted as sobelH(x, y) and the signal value after processing by a horizontal Sobel filter is denoted as sobelV(x, y), the strength of the luminance change Strength(x, y) can be calculated using the following equation (1), and the direction of the luminance change Angle(x, y) can be calculated using the following equation (2). Strength(x, y)=√(sobelH(x, y) 2 +sobelV(x, y) 2 ) …(1) Angle(x, y)=tan -1 (sobelH(x, y) / sobelV(x, y)) …(2)

[0053] After the processes of steps S33 to S36 are performed for all pixels included in the small region, the distribution of the angle of the brightness change is calculated in order to convert the information on the direction of the brightness change into information on the direction of the cell (step S37). To do this, first, the continuous numerical value Angle(x, y) obtained by the formula (2) is converted to a numerical value at a fixed angle interval, for example, every 1 degree, by rounding or the like. Then, the distribution of the converted angles, that is, a histogram, is obtained. In this case, the number of pixels whose brightness change strength Strength(x, y) is equal to or greater than a predetermined threshold value is counted for each angle, and the counted value is used as the frequency of the angle distribution. In addition, the sum of the brightness change strength Strength(x, y) of all pixels whose brightness change direction is at that angle may be calculated for each angle, and the value of the sum may be used as the frequency in the angle distribution.

[0054] Next, the angle of the angle distribution obtained in step S37 is rotated by 90° (step S38). This is because the direction of the change in brightness and the orientation of the cells are perpendicular to each other. Note that the shape of the mesenchymal stem cells assumed here is an elongated ellipse or needle, and the longitudinal direction is the orientation of the cells. Of course, the order of steps S37 and S38 can be reversed. After that, the frequency of the angle distribution is normalized to enable numerical evaluation (step S39), and the accuracy of the distribution is improved by calculating the moving average of the angle distribution (step S40).

[0055] FIG. 13(d) shows an example of angle distribution when the orientation of cells is uniform (see 13(b)) and when it is not uniform (see 13(c)). As shown in this figure, when the orientation of cells is uniform overall, the frequency at a particular angle is high, so there is a difference in frequency between when the orientation of cells is uniform and when it is not uniform. Therefore, the orientation score calculation unit 27 determines the angle with the highest frequency in the angle distribution after performing the moving average, that is, the most frequent angle, and determines this as the orientation direction of cells in that small region (step S41). In addition, the frequency corresponding to the most frequent angle is determined as the strength of the orientation of cells in that small region (step S42).

[0056] By carrying out the above process for each small region, a numerical value indicating the direction and strength of the orientation of the cells can be obtained for each small region. The orientation score calculation unit 27 calculates the average value of the orientation strength in all small regions as an orientation score indicating the tendency of the orientation of the cells in the entire phase image, and obtains the maximum value of the orientation strength as the orientation score. In addition, an evaluation image is created for the user to intuitively evaluate the orientation (step S24). Specifically, the orientation score is visualized for each small region by a two-dimensional vector in which the orientation direction is the vector direction and the orientation strength is the length of the vector (scalar quantity). Then, an image in which the two-dimensional vector is superimposed on the IHM phase image is created as an evaluation image.

[0057] FIG. 14 shows an example of an evaluation image of the orientation of two types of samples, one in which the cells are aligned and the other in which they are not aligned. Small regions are shown on the IHM phase image, and lines indicating the strength and direction of the orientation are superimposed on each small region. When viewed locally, the cells of sample 2 are more aligned than those of sample 1. The difference in the characteristics of the images is reflected in the difference in the length of the lines drawn in each small region. Therefore, by displaying this evaluation image on the display unit 4 and presenting it to the user, the user can intuitively grasp the degree of alignment of the cells.

[0058] Figure 15 shows a comparative example of the orientation scores for the two types of samples shown in Figure 14. The difference in the image features described above is clearly reflected in both the average and maximum orientation scores. In other words, the orientation score makes it possible to numerically compare and evaluate the degree of cell alignment.

[0059] [Evaluation of cell function] The cell function evaluation information output unit 28 displays the gradient score and orientation score calculated as described above, as well as the evaluation image related to the orientation, in a predetermined format on the display unit 4. The user evaluates the cell function based on these displays. The cell functions that can be evaluated here can include five types, namely, maintenance of undifferentiated state, migration ability, tissue repair ability, angiogenesis ability, and immune control ability. In addition, the state of the cytoskeleton (actin filament) related to cell hypertrophy and flattening can also be evaluated. The fact that the above scores can be used to evaluate the functions and states of these cells will be explained based on the results of experiments conducted by the inventors.

[0060] The subject of evaluation in this experimental example is human umbilical cord-derived mesenchymal stem cells (hereinafter sometimes abbreviated as UC-MSC). The method of collecting the subject cells and the method of identifying the cell functions are as follows.

[0061] Figure 16 shows a light microscopy image of the tissue from the cell harvest site. This is a hematoxylin and eosin stained cross section of the umbilical cord. In the following explanation, cells collected from a thin slice of the amniotic membrane side of the umbilical cord are referred to as Am-MSCs, and cells collected from the interstitial tissue (Wharton's jelly gelatin) near the umbilical artery and vein are referred to as Wj-MSCs. The tissues collected from each of the above-mentioned parts of the umbilical cord were cut into pieces of about 3 to 5 mm square and placed in a petri dish for cell culture. The petri dish was left to stand in an incubator at 37°C and 5% CO2 for 30 minutes, and when the tissue pieces attached to the bottom of the culture dish, a basal medium containing fetal bovine serum (FBS) was added and cultured. Cells that had migrated out of the tissue pieces and attached to the culture dish were collected, and the cells were then passaged and cultured.

[0062] In order to produce clear differences in cell functions, the cultured cells were divided into those with and without the addition of a cell activator (Wharton's jelly-derived extract, hereafter sometimes referred to as WJ) after five repeated passages. Cells in which WJ was added to Am-MSC are referred to as Am-MSC+WJ, and cells in which WJ was added to Wj-MSC are referred to as Wj-MSC+WJ. Am-MSC and Wj-MSC are the reference cells, and Am-MSC+WJ and Wj-MSC+WJ are the test cells.

[0063] (1) Expression of cell surface antigens First, we used existing methods to confirm the cell functions of samples containing the four types of cells, Am-MSC, Wj-MSC, Am-MSC+WJ, and Wj-MSC+WJ, which were derived from four different specimens (A, B, C, and D).

[0064] The expression of surface antigens of UC-MSCs after 48 hours of culture with or without the addition of WJ was analyzed by flow cytometry. Figure 17 shows the analysis results of the expression of cell surface antigens for specimen A. Figures 21, 25, and 29 show the analysis results of the expression of cell surface antigens for specimens B, C, and D, respectively. In each of these figures, the absence of WJ is indicated as WJ-, and the presence of WJ is indicated as WJ+. Furthermore, CD (Cluster of Differentiation) 90, CD73, CD105, CD34, and CD45 are types of antigens. As is well known, the left peak is an isotope control and the right peak is the antigen of interest; if the antigen is not expressed, the two peaks will overlap.

[0065] In these results, both Am-MSC and Wj-MSC, which were not added with activator, expressed three types of antigens, CD90, CD73, and CD105, but did not express CD34 or CD45. Similarly, both Am-MSC+WJ and Wj-MSC+WJ, which were added with activator, expressed three types of antigens, CD90, CD73, and CD105, but did not express CD34 or CD45. These results were common to samples A, B, C, and D. In general, it has been reported that CD90, CD73, and CD105 antigens are positive, and CD34 and CD45 antigens are negative in human mesenchymal stem cells. The above analysis results were consistent with this report, and it was confirmed that the UC-MSC used in this experiment had the characteristics of cell surface antigens of human mesenchymal stem cells, and that the characteristics did not change even when WJ was added.

[0066] (2) Expression of nucleic acids related to various cell functions After 72 hours of culture with or without activators, the expression of nucleic acids (genes) related to various functions of UC-MSCs was analyzed using real-time PCR (polymerase chain reaction). In this study, we analyzed the following factors related to UC-MSC undifferentiation (factors indicating the maintenance of cell undifferentiation): OCT4 (Octamer-binding transcription factor 4) and NANOG; SDF-1 (Stromal derived factor-1) and CXCR4 (CXC chemokine receptor type 4); TGFβ (Transforming growth factor-β) as a tissue repair-related factor indicating high tissue repair ability; αSMA (α-smooth muscle actin) as a cytoskeleton-related factor; VEGF (Vascular endothelial growth factor) as an angiogenesis-related factor; and PD-L1 (Programmed cell death - Ligand 1) as an immune regulation-related factor. Among these, VEGF is a vascular endothelial cell growth factor that induces the division, migration, and differentiation of endothelial cells and promotes angiogenesis. PD-L1 is an immune checkpoint protein that cooperates with the PD-1 receptor expressed in immune cells to suppress the activation of immune cells (e.g., T cell response) and is considered to be one of the main factors that contribute to the immune control ability of mesenchymal stem cells. αSMA is a factor related to cell shape, and it is known that excessive accumulation within a cell causes the cell body to expand and enlarge, resulting in a decrease in the migration ability of the cell, or inhibiting the distribution of the cell to target organs in the body (see, for example, Non-Patent Document 1).

[0067] 18, 22, 26 and 30 are graphs showing the results of analysis of nucleic acid expression of cell function-associated factors for samples A, B, C and D, respectively. Although the amount of increase in the expression of SDF-1 and TGFβ varies depending on the sample, an increase in expression can be confirmed in all of samples A to D by the addition of an activator. Furthermore, the expression of αSMA is suppressed by the addition of an activator in all of samples A to D. Furthermore, the expression of VEGF and PD-L1 is increased by the addition of an activator in all of samples A to D, except for Wj-MSC in sample C. In Wj-MSC in sample C, the expression is slightly decreased by the addition of an activator, but the decrease is only mild.

[0068] The above results show that cells to which activators have been added have improved undifferentiation, cell migration ability, tissue repair ability, angiogenesis ability, and immune control ability, while actin components involved in cell migration, orientation, cell enlargement, and flattening are reduced. In addition, the results of morphological observation of these cells show that cells to which activators have been added have changed to an elongated spindle shape, supporting the idea that the addition of activators improves the migration ability of cells.

[0069] (3) Measurement of cell migration ability The migration ability of cells was evaluated using a common scratch assay. Specifically, each UC-MSC was cultured in a 35 mm diameter culture dish, and then a linear open wound was formed perpendicular to the cultured cells. The area of ​​the nine intersections of each cell was measured for the area where no cells were present (residual open area) at 0 hours, immediately after the open wound was formed, and after 9 hours of culture after the open wound was formed. The ratio of the remaining open area at 9 hours to the open area at 0 hours was then calculated, and the degree of remnant wound area was used to evaluate the migration ability of UC-MSC cells. The higher the migration ability of the cells, the smaller the area of ​​the open area.

[0070] Figures 19, 23, 27, and 31 are box plots showing the quantitative values ​​of nine regions in each cell for specimens A, B, C, and D, respectively. These figures show that in all specimens A to D, except for Am-MSC in specimen B, the area of ​​the remaining open wound was smaller in the activator-added group than in the non-added group. In addition, in Am-MSC in specimen B, the area of ​​the remaining open wound was almost the same in the activator-free group and the activator-added group. These results confirmed that the migration ability of cells was generally improved by the addition of an activator.

[0071] (4) Cell gradient score and orientation score In the IHM phase images obtained under each condition 48 hours and 72 hours after the addition of the activator to the cells, 15 fields including the center (within a radius of about 20 mm including the center of the culture dish) were extracted, and the gradient score and orientation score were calculated by the above-mentioned processing. Figures 20, 24, 28, and 32 are box plots showing the gradient score and orientation score of four types of MSCs for specimens A, B, C, and D, respectively. Since the gradient score indicates the inclination of the slope of the peak in the above-mentioned differential value histogram, a decrease in the gradient score indicates an increase in cells with a steeper contour.

[0072] From these figures, it can be seen that in all of specimens A to D, the gradient score in the non-activator-added group increases over time, in other words, the optical thickness of the cells decreases. On the other hand, it can be seen that the increase in gradient score over time is suppressed in the activator-added group. Furthermore, it can be seen that the gradient score in the activator-added group is already lower at 48 hours, in other words, the optical thickness of the cells has increased, and this tendency is even more pronounced at 72 hours. In other words, the addition of an activator clearly reduces the gradient score over time.

[0073] Furthermore, in all of the specimens A to D, except for the Am-MSC in specimen C, the group with the activator added showed a clear tendency for the orientation score to increase after either 48 hours or 72 hours, or both, compared to the group without the activator added. Only for the Am-MSC in specimen C, the change in the orientation score due to the addition of the activator was small. The increase in the orientation score indicates that the cells are changing to be aligned in their orientation.

[0074] As described above, it was confirmed that the addition of the activator reduced the gradient score of the cells and increased the optical thickness. The gradient score of the cells was clearly reduced in all of the specimens A to D, and this reduction in gradient score was linked to the changes in undifferentiation-related factors, cell migration-related factors, cytoskeleton-related factors, and tissue repair-related factors evaluated in nucleic acid expression, and the cell migration ability confirmed by the scratch assay. In addition, some, but not all, of the angiogenesis-related and immune control-related factors were clearly linked to the reduction in gradient score.

[0075] On the other hand, the cell orientation score of sample C hardly changed even when the activator was added, but it was confirmed that the orientation was increased in the activator-added group in the other samples A, B (after 48 hours), and D. In sample C (especially Wj-MSC), not only the orientation but also the expression of undifferentiation-related factors, migration-related factors, tissue repair-related factors, angiogenesis-related factors, and immune control-related factors showed little change or different expression trends compared to other samples. From this, it is speculated that the cell orientation is highly likely to affect the expression of undifferentiation-related factors, migration-related factors, tissue repair-related factors, angiogenesis-related factors, and immune control-related factors in cells.

[0076] Therefore, even if the function or state of a cell is unknown, by comparing the gradient scores and orientation scores of a reference cell and a test cell, it is possible to evaluate whether the cell functions of the test cell, such as cellular undifferentiation, migration ability, tissue repair ability, angiogenesis ability, and immune control ability, are in a high state compared to the reference cell.

[0077] In the cell analysis device of this embodiment, the cell function evaluation information output unit 28 displays, for example, the gradient score and orientation score obtained for the test cell, and the evaluation image related to orientation, and the gradient score and orientation score obtained for the reference cell, and the evaluation image related to orientation, on the display unit 4 in a predetermined format that is easy to compare (step S9). At this time, in addition to the IHM phase image, the differential image created in step S5 and the differential value histogram created in step S6 may also be displayed. The user relatively evaluates the cell functions of the test cell, such as cell undifferentiation, migration ability, tissue repair ability, angiogenesis ability, and immune control ability, of the test cell relative to the reference cell based on the displayed gradient score, orientation score, or evaluation image (step S10).

[0078] After the steps of evaluating cell function as described above have been completed, the following steps, for example, can be carried out. That is, cells are sorted (collected) based on the evaluation results of cell function. For example, a fluorescence activated cell sorter (FACS) can be used to separate the cells. Then, cells with different evaluation results of cell function are cultured again in different culture modes. By subculturing this culture, it is confirmed that cells that are unsuitable for transplantation, for example, cells with low migration ability or low tissue repair function, do not grow (do not adhere or proliferate) in the next generation. In this way, the evaluation of cell functions as described above is useful for confirming and verifying proliferation ability and the like due to differences in cell functions.

[0079] In addition, instead of a user, i.e., a human being, evaluating the cell functions based on the displayed image or score, for example, by comparing the gradient scores and orientation scores of a reference cell and a test cell and judging the magnitude relationship between them, a process of relatively evaluating the cell functions of the test cell, such as the cell undifferentiation, migration ability, tissue repair ability, angiogenesis ability, and immune control ability, relative to the reference cell, may be performed in the control / processing unit 2, i.e., on a computer including a CPU or the like. In this case, only the final evaluation result may be output to the display unit 4, or the evaluation result may be output together with the score or image. In addition, various machine learning methods such as deep learning may be used for such judgment and evaluation.

[0080] The test cell to be evaluated and the reference cell may be different types of cells, but in practical applications, they are often the same type of cells cultured under different conditions. Of course, the test cell and the reference cell do not need to be one each, and for example, three or more test cells may be compared with each other to perform a relative evaluation of cell function. Data such as the gradient score and orientation score for the reference cell, and evaluation images related to orientation may be stored in advance in the control and processing unit 2 as evaluation criteria, and the stored evaluation criteria and the gradient score and orientation score newly calculated based on the hologram data obtained by the microscopic observation unit 1 for the test cell may be displayed in a format that allows comparison.

[0081] 1, all processing is performed in the control and processing unit 2, but in general, calculation of phase information based on hologram data and image reconstruction processing require a huge amount of calculation. Therefore, a computer system may be used in which a personal computer connected to the microscope observation unit 1 is used as a terminal device and this terminal device and a server, which is a high-performance computer, are connected via a communication network such as the Internet or an intranet, and the above-mentioned complicated calculations and processing are performed by the high-performance computer, while the control of the microscope observation unit 1 and display processing using the processed data are performed by a relatively low-performance personal computer.

[0082] In addition, in the cell analysis device of the above embodiment, an in-line holographic microscope is used as the microscopic observation unit 1, but it is natural that this can be replaced with other types of holographic microscopes, such as off-axis type and phase shift type, as long as the microscope can obtain holograms.

[0083] In addition, in the cell analysis device of the above embodiment, both a score of the thickness gradient at the cell contour and a score indicating the cell orientation are calculated and presented to the user, but it is also possible to calculate at least one of them and present it to the user, who can then evaluate the cell function based on that. To calculate the gradient score in the thickness direction at the contour of the cell, the observed image of the cell needs to include information in the thickness direction (i.e., three-dimensional shape information), such as an IHM phase image. On the other hand, to calculate a score indicating the orientation of the cell, information in the thickness direction is not necessary, and it is sufficient to obtain an image in which the cell can be clearly observed. Therefore, in that case, the microscopic observation unit 1 does not need to be a holographic microscope, and a general phase-contrast microscope or the like may be used.

[0084] Furthermore, the above-mentioned method for evaluating cell function is also applicable to other types of stem cells in addition to mesenchymal stem cells for the following reasons. That is, according to the study by the present inventors, the number of actin fibers in the cell is reduced, and the flat and spread shape is changed to a thin spindle shape, and the localization of intracellular organelles and proteins is also changed, and the thickness of the cell increases (increased gradient), and as a result, it has become clear that the function of sensing shear stress present on the side near the apex of the nucleus works more sensitively, the orientation is improved, and the migration ability and tissue repair ability of the cell are enhanced. In other words, it is predicted that the function of sensing shear stress of the cell is related to the migration ability and cell repair ability of the cell, and the present invention has made it clear that evaluating the orientation is important in evaluating the function of sensing shear stress. From these facts, it is clear that the present invention can be applied to cells having the function of sensing shear stress, not limited to mesenchymal stem cells, to evaluate cell functions such as migration ability and tissue repair ability of the cell.

[0085] Furthermore, the above embodiment and various modified examples are merely examples of the present invention, and it is clear that further appropriate modifications, amendments, and additions made within the spirit of the present invention will also be encompassed within the scope of the claims of this application.

[0086] [Various aspects] It will be appreciated by those skilled in the art that the exemplary embodiments described above are examples of the following aspects.

[0087] (Item 1) A method for evaluating a cell function according to one aspect of the present invention is a method for evaluating a cell function of a test cell which is a stem cell, comprising the steps of: The test cells were photographed , including information reflecting the thickness of the test cell. An image acquisition step of acquiring a cell image; From the cell image, a contour of the test cell is In the thickness direction Information about the gradient News a feature extraction step of extracting a feature of the test cell; an evaluation step of evaluating a cellular function of the test cell based on the feature amount of the test cell; It has the following. (Item 2) In the method according to item 1, the feature extraction step can further include extracting information about the orientation of the test cell from the cell image as the feature of the test cell.

[0088] (First 1 A cell analysis device according to one aspect of the present invention is a cell analysis device for evaluating a cell function of a test cell that is a stem cell, comprising: The test cells were photographed , including information reflecting the thickness of the test cell. an image acquisition unit for acquiring a cell image; From the cell image, a contour of the test cell is In the thickness direction Information about the gradient News a feature extraction unit that extracts a feature of the test cell; an evaluation processing unit that relatively evaluates a cellular function of the test cell based on the feature amount of the test cell; It is equipped with the following. (Item 12) In the device described in Item 11, the feature extraction section can further extract information regarding the orientation of the test cell from the cell image as a feature of the test cell.

[0089] (Part 3 Section) Section 1 or paragraph 2In the method for evaluating a cell function described in the above, the cell function may include at least one of the following: maintenance of undifferentiated state, migration ability, tissue repair ability, angiogenesis ability, and immune control ability.

[0090] (First 3 Section) No. 1 Paragraph 1 or 12 In the cell analysis device described in the above, the cell functions may include at least one of the following: maintenance of undifferentiated state, migration ability, tissue repair ability, angiogenesis ability, and immune control ability.

[0091] Section 1 or paragraph 2 The method for evaluating a cell function according to the first aspect of the present invention Paragraph 1 or 12 According to the cell analysis device described in the above, the functions of stem cells, such as maintenance of undifferentiated state, migration ability, tissue repair ability, angiogenesis ability, and immune control ability, can be quantitatively evaluated based on image information obtained non-invasively and non-destructively from the stem cells. Since such evaluation of cell functions can be performed using numerical values, even a person who has little experience or knowledge in cell analysis or evaluation can perform an accurate and reliable evaluation. In addition, it is possible to avoid the variation in results depending on the person, which is common in evaluations that rely on the sense of the person in charge. In addition, it is also easy to verify the reasons for the evaluation. Furthermore, it is also possible to improve the efficiency of the evaluation of cell functions.

[0092] (Part 4 Section) Section 1 Any one of the following items In the method for evaluating a cellular function described above, the cellular image may be a phase image.

[0093] (First 4 Section) No. 1 Any one of clauses 1 to 13 In the cell analysis apparatus described in the above, the cell image may be a phase image.

[0094] In phase imaging, cells that are generally transparent and difficult to see with an optical microscope can be captured relatively clearly. 4The method for evaluating a cell function according to the first aspect of the present invention 4 According to the cell analysis device described in the paragraph, it is possible to extract features of a test cell with high accuracy and to appropriately evaluate the function of the cell.

[0095] (Part 5 Section) No. 4 In the method for evaluating a cellular function according to the above item, the image acquisition step comprises: a measuring step of irradiating a test cell with laser light and detecting interference light between the light that has passed through the test cell and the light that has passed around the test cell to obtain hologram data; a phase image creating step of performing an image reconstruction process based on the hologram data to create a phase image of the test cell; It may include:

[0096] (First 5 Section) No. 1 4 In the cell analysis device according to the above item, the image acquisition unit is a measurement execution unit that irradiates a test cell with laser light and detects interference light between the light that has passed through the test cell and the light that has passed around the test cell, thereby acquiring hologram data; a phase image creating unit that performs image reconstruction processing based on the hologram data and creates a phase image of the test cell; It may include:

[0097] The phase image created based on the hologram data in the phase image creating step includes not only two-dimensional information about the test cell but also information about the thickness direction of the cell. 5 The method for evaluating a cell function according to the first aspect of the present invention 5 According to the cell analysis device described in the item 1, the outline of the test cell is detected from the phase image. In the thickness direction It is possible to extract information regarding the gradient of the above-mentioned activity with high accuracy, and based on this information, it is possible to accurately evaluate the functions of cells, such as the maintenance of undifferentiated state, migration ability, tissue repair ability, angiogenesis ability, and immune control ability.

[0098] (Part 6 Section) Sections 1 to 1 5 In the method for evaluating a cell function according to any one of the items, A reference image acquisition step of acquiring a reference image which is a cell image of a reference cell whose cell function is known; a reference feature extraction step of extracting, from the reference image, a feature corresponding to the feature extracted in the feature extraction step; The method may further include: in the evaluation step, evaluating a cellular function of the test cell by comparing a feature amount for the test cell with a feature amount for the reference cell.

[0099] (First 6 Section) No. 1 1 Term~1st 5 In the cell analysis device described in any one of the items, the evaluation processing unit can be configured to relatively evaluate the cell function of the test cell by comparing features for the test cell with features extracted from a previously acquired cell image of a reference cell whose cell function is known.

[0100] The reference cell does not necessarily have to be a cell different from the test cell. That is, for example, the cell image of the reference cell may be a cell image previously obtained for the test cell. 6 The method for evaluating a cell function according to the first aspect of the present invention 6 According to the cell analysis device described in the paragraph, it is possible to accurately evaluate whether a test cell has superior or inferior cell functions, such as maintenance of undifferentiated state, migration ability, tissue repair ability, angiogenesis ability, and immune control ability, in comparison with a reference cell.

[0101] (Part 7 Section) No. 4 Section or Article 5 In the method for evaluating a cellular function according to the above item, the feature extraction step comprises: a differential image generating step of generating a differential image by calculating a spatial differential value for each pixel from the phase image; a gradient information extraction step of extracting information related to a gradient in a thickness direction at a contour portion of a test cell from a cellular region including the test cell in the differential image; It may include:

[0102] (First 7 Section) No. 1 4 Item or 1 5 In the cell analysis apparatus according to the above item, the feature extraction unit obtains a spatial differential value for each pixel from the phase image to generate a differential image, and extracts a contour portion of the test cell from a cellular region including the test cell in the differential image. In the thickness direction The method may extract information related to the gradient of

[0103] No. 7 The method for evaluating a cell function according to the first aspect of the present invention 7 According to the cell analysis device described in the item 1, In the thickness direction It is possible to extract information regarding the gradient of the above-mentioned activity with high accuracy, and based on this information, it is possible to accurately evaluate the functions of cells, such as the maintenance of undifferentiated state, migration ability, tissue repair ability, angiogenesis ability, and immune control ability.

[0104] (Part 8 Section) No. 7 In the method for evaluating a cell function according to the above item, The feature extraction step includes: a pixel number acquiring step of calculating the number of pixels corresponding to the cellular region in the differential image; a cell number / confluency acquisition step of acquiring the number or confluency of cells from the phase image; The number of pixels is divided by the number or confluency of the cells to obtain a contour of the test cell. In the thickness direction The method may seek information related to the gradient of

[0105] (First 8 Section) No. 1 7In the cell analysis device according to the above item, the feature extraction unit calculates the number of pixels corresponding to the cell region in the differential image, obtains the number of cells or the confluency from the phase image, and divides the number of pixels by the number of cells or the confluency to obtain a contour of the test cell. In the thickness direction The method may seek information related to the gradient of

[0106] No. 8 The method for evaluating a cell function according to the first aspect of the present invention 8 According to the cell analysis device described in the paragraph, even if the densities of the test cells to be evaluated and the reference cells serving as the standard for comparison differ, it is possible to accurately perform comparative evaluation of cell functions.

[0107] (Part 9 Section) No. In item 2 In the method for evaluating a cellular function described above, the feature extraction step comprises: an image division step of dividing the cell image into a plurality of small regions; an orientation information calculation step of calculating, for each of the small regions, a direction in which brightness changes and a magnitude of the change by applying a differential filter to each pixel contained in the small region, and obtaining information regarding the orientation of the test cell from the direction and magnitude of the brightness change for each of the small regions; It may include:

[0108] (First 9 Section) No. 1 In item 2 In the described cell analysis device, the feature extraction unit divides the cell image into a number of small regions, and for each small region, applies a differential filter to each pixel contained in that small region to calculate the direction in which brightness changes and the magnitude of that change, and obtains information regarding the orientation of the test cell from the direction and magnitude of the brightness change for each small region.

[0109] No. 9 The method for evaluating a cell function according to the first aspect of the present invention 9According to the cell analysis device described in the paragraph, it is possible to obtain reliable quantitative information regarding cell orientation, i.e., the degree to which cells are aligned, thereby enabling accurate evaluation of cell function.

[0110] (Part 10 Section) Sections 1 to 1 9 In the method for evaluating a cell function according to any one of the above items, the stem cells may be mesenchymal stem cells.

[0111] (Part 20 Section) No. 1 1 Term~1st 9 In the cell analysis device according to any one of the above items, the stem cells may be mesenchymal stem cells.

[0112] No. 10 The method for evaluating cell function according to the present invention 20 According to the cell analysis device described in the paragraph, in particular, the function of cells can be accurately evaluated. [Explanation of symbols]

[0113] 1. Microscopic observation section 10...Light source section 11...Image sensor 12…Culture plate 12a…Well 13...Cell 14...Reference light 15…Object light 2. Control and processing section 20...Shooting control unit 21...Hologram data storage unit 22...Phase information calculation unit 23...Image reconstruction unit 24...Reconstructed image data storage unit 25…Differential image creation section 26...Gradient score calculation section 27...Orientation score calculation unit 28…Cell function evaluation information output section 3. Input section 4…Display part

Claims

1. A method for evaluating a cellular function of a test cell which is a stem cell, comprising the steps of: an image acquiring step of acquiring a cell image including information reflecting the thickness of the test cell by photographing the test cell; a feature extraction step of extracting information regarding a gradient in a thickness direction at a contour portion of a test cell from the cell image as a feature of the test cell; an evaluation step of evaluating a cellular function of the test cell based on the feature amount of the test cell; A method for evaluating the function of a cell having

2. A method for evaluating cell function as described in claim 1, wherein the feature extraction step further comprises extracting information regarding the orientation of the test cell from the cell image as a feature of the test cell.

3. The method for evaluating a cell function according to claim 1 or 2, wherein the cell function includes at least one of the following: maintenance of undifferentiated state, migration ability, tissue repair ability, angiogenesis ability, and immune control ability.

4. The method for evaluating a cell function according to any one of claims 1 to 3, wherein the cell image is a phase image.

5. The image acquisition step includes: a measuring step of irradiating a test cell with laser light and detecting interference light between the light that has passed through the test cell and the light that has passed around the test cell to obtain hologram data; a phase image creating step of performing an image reconstruction process based on the hologram data to create a phase image of the test cell; The method for evaluating a cell function according to claim 4 .

6. A reference image acquisition step of acquiring a reference image which is a cell image of a reference cell whose cell function is known; a reference feature extraction step of extracting, from the reference image, a feature corresponding to the feature extracted in the feature extraction step; The method for evaluating cellular function according to any one of claims 1 to 5, further comprising: in the evaluation step, evaluating the cellular function of the test cell by comparing a characteristic amount for the test cell with a characteristic amount for the reference cell.

7. The feature extraction step includes: a differential image generating step of generating a differential image by calculating a spatial differential value for each pixel from the phase image; a gradient information extraction step of extracting information about a gradient in a thickness direction at a contour portion of a test cell from a cellular region including the test cell in the differential image; The method for evaluating a cell function according to claim 4 or 5, comprising:

8. The feature extraction step includes: a pixel number acquiring step of calculating the number of pixels corresponding to the cellular region in the differential image; a cell number / confluency acquisition step of acquiring the number or confluency of cells from the phase image; The method for evaluating a cell function according to claim 7 , further comprising: determining information regarding a gradient in the thickness direction at a contour portion of the test cell by dividing the number of pixels by the number or confluency of the cells.

9. The feature extraction step includes: an image division step of dividing the cell image into a plurality of small regions; an orientation information calculation step of calculating, for each of the small regions, a direction in which brightness changes and a magnitude of the change by applying a differential filter to each pixel contained in the small region, and obtaining information regarding the orientation of the test cell from the direction and magnitude of the brightness change for each of the small regions; The method for evaluating a cell function according to claim 2 .

10. The method for evaluating a cell function according to any one of claims 1 to 9, wherein the stem cells are mesenchymal stem cells.

11. A cell analysis device for evaluating a cell function of a test cell which is a stem cell, comprising: an image acquiring unit that acquires a cell image obtained by photographing a test cell and that includes information reflecting the thickness of the test cell; a feature extraction unit that extracts information regarding a gradient in a thickness direction at a contour portion of a test cell from the cell image as a feature of the test cell; an evaluation processing unit that relatively evaluates a cellular function of the test cell based on the feature amount of the test cell; A cell analysis device comprising:

12. The cell analysis device of claim 11, wherein the feature extraction unit further extracts information regarding the orientation of the test cell from the cell image as a feature of the test cell.

13. The cell analysis device according to claim 11 or 12, wherein the cell functions include at least one of the following: maintenance of undifferentiated state, migration ability, tissue repair ability, angiogenesis ability, and immune control ability.

14. The cell analysis device according to any one of claims 11 to 13, wherein the cell image is a phase image.

15. The image acquisition unit includes: a measurement execution unit that irradiates a test cell with laser light and detects interference light between the light that has passed through the test cell and the light that has passed around the test cell, thereby acquiring hologram data; a phase image creating unit that performs image reconstruction processing based on the hologram data and creates a phase image of the test cell; The cell analysis device according to claim 14 .

16. The cell analysis device according to any one of claims 11 to 15, wherein the evaluation processing unit relatively evaluates the cell function of the test cell by comparing features for the test cell with features extracted from a previously acquired cell image of a reference cell whose cell function is known.

17. The cell analysis device according to claim 14 or 15, wherein the feature extraction unit generates a differential image by calculating a spatial differential value for each pixel from the phase image, and extracts information regarding a gradient in the thickness direction at a contour portion of the test cell from a cellular region containing the test cell in the differential image.

18. 18. The cell analysis device according to claim 17, wherein the feature extraction unit calculates a number of pixels corresponding to the cellular region in the differential image, obtains a number of cells or a confluency from the phase image, and obtains information regarding a gradient in a thickness direction at a contour portion of the test cell by dividing the number of pixels by the number of cells or the confluency.

19. The cell analysis device of claim 12, wherein the feature extraction unit divides the cell image into a plurality of small regions, and for each small region, calculates the direction of brightness change and the magnitude of that change by applying a differential filter to each pixel contained in that small region, and obtains information regarding the orientation of the test cell from the direction and magnitude of the brightness change for each small region.

20. The cell analysis device according to any one of claims 11 to 19, wherein the stem cells are mesenchymal stem cells.

Images