Cell image analysis device and cell image analysis method

Abstract

To provide a cell image analysis technique that can evaluate cell images for image feature quantities with high accuracy after setting standards.SOLUTION: Provided is a cell image analysis device comprising: an image acquisition unit that acquires time-series images that are acquired at two or more different time points in a time series of cell culture, each of which includes an identical cell colony; a region extraction unit that extracts a region of the identical cell colony included in each of the time-series images; a feature quantity extraction unit that extracts image feature quantities from the region extracted by the region extraction unit; a feature quantity estimation unit that estimates image feature quantities for the identical cell colony at a predetermined time point when the time-series images are not acquired, using the image feature quantities extracted by the feature quantity extraction unit; and an evaluation unit that evaluates at least one selected from the group consisting of the identical cell colony and conditions of the cell culture, using the image feature quantities estimated by the feature quantity estimation unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The disclosure of this specification relates to a cell-image analyzing device, a cell-image analyzing method, a cell-image analyzing program, and a recording medium. [Background technology]

[0002] In recent years, in the field of cell culture, cell image analysis techniques that can perform non-invasive total inspection, in-process monitoring of cell culture conditions, etc. by capturing images of cells to be cultured and analyzing the images have been attracting attention. As a specific cell image analysis method, a method has been proposed in which image features such as the size, shape, and texture of a cell colony are extracted from an image of the cell colony, and the culture of the cell colony is evaluated based on these image features.

[0003] The image feature amount is not only a feature to be evaluated for the cell colony, but also a value that changes simply as the cell colony grows. Therefore, in order to appropriately perform evaluation based on the image feature amount, it is preferable to use the image feature amount for evaluation after taking into consideration the change in the image feature amount associated with the growth of the cell colony.

[0004] Patent Document 1 discloses a technique for extracting features related to the maturity of cells, such as area, from images of the cells and combining them with other image features to evaluate the differentiation state of the cells. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2015-61516 A Summary of the Invention [Problem to be solved by the invention]

[0006] The maturity of colonies generated during cell culture may differ from one another even in the same culture vessel. Therefore, the method disclosed in Patent Document 1 has a problem in that when evaluating colonies by comparing feature amounts between different colonies, it is not possible to compare the maturity of colonies derived from the colony area and the like by aligning them to the same value.

[0007] Another possible method is to acquire time-series images at high frequency and select and compare images of different colonies with the same degree of maturity as possible. However, in order to acquire time-series images at high frequency, it is necessary to increase the data acquisition speed, and in order to increase the data acquisition speed, it is necessary to reduce the imaging area or the resolution. As a result, there are problems in that the number of colonies to be evaluated is insufficient or the quality of the images used for analysis is insufficient.

[0008] The present invention has been made in view of the above problems, and aims to provide a cell image analysis technique that can evaluate cell images with high accuracy by standardizing image feature amounts that change during cell culture. [Means for solving the problem]

[0009] A cell image analysis device according to one embodiment of the present invention is characterized in that it includes an image acquisition unit that acquires time-series images, the images being acquired at two or more different time points in a time series of cell culture, each of which includes an identical cell colony; an area extraction unit that extracts an area of ​​the identical cell colony included in each of the time-series images; a feature extraction unit that extracts image features from the area extracted by the area extraction unit; a feature estimation unit that uses the image features extracted by the feature extraction unit to estimate image features for the identical cell colony at a predetermined time point when the time-series images were not acquired; and an evaluation unit that uses the image features estimated by the feature estimation unit to evaluate at least one selected from the group consisting of the identical cell colony and the cell culture conditions.

[0010] In addition, a cell image analysis device according to another embodiment of the present invention is characterized in that it comprises an image acquisition unit that acquires time-series images, the images being acquired at two or more different time points in a time series of cell culture, each of which includes an identical cell colony; an area extraction unit that extracts an area of ​​the identical cell colony included in each of the time-series images; a feature extraction unit that extracts, from the area extracted by the area extraction unit, a first image feature corresponding to a feature that changes depending on the culture time of the cell culture and a second image feature corresponding to a feature used for evaluation; a feature estimation unit that estimates the second image feature corresponding to a predetermined value for the first image feature; and an evaluation unit that uses the second image feature estimated by the feature estimation unit to evaluate at least one selected from the group consisting of the identical cell colony and the conditions of the cell culture.

[0011] Furthermore, a cell image analysis method according to another aspect of the present invention includes an image acquisition step of acquiring time-series images, the images being acquired at two or more different time points in a time series of cell culture, each of which includes the same cell colony; an area extraction step of extracting an area of ​​the same cell colony included in each of the time-series images; a feature extraction step of extracting image features from the area of ​​the same cell colony in a plurality of images included in the time-series images; a feature estimation step of estimating image features for the same cell colony at a predetermined time point when the time-series images were not acquired using the image features extracted in the feature extraction step; and an evaluation step of evaluating the same cell colony or the conditions of the cell culture using the image features estimated in the feature estimation step.

[0012] In addition, a cell image analysis method according to another aspect of the present invention includes an image acquisition step of acquiring time-series images, the images being acquired at two or more different time points in a time series of cell culture, each including the same cell colony; a region extraction step of extracting an area of ​​the same cell colony included in each of the time-series images; a feature extraction step of extracting, from the area extracted in the region extraction step, a first image feature corresponding to a feature that changes depending on the culture time of the cell culture and a second image feature corresponding to a feature used for evaluation; a feature estimation step of estimating the second image feature corresponding to a predetermined value for the first image feature; and an evaluation step of evaluating at least one selected from the group consisting of the same cell colony and the conditions of the cell culture, using the second image feature estimated by the feature estimation step. Effect of the Invention

[0013] According to the disclosure of this specification, a cell image analysis technique is provided that is capable of evaluating cell images with high accuracy by establishing standards for image feature amounts that change during the cell culture process. [Brief description of the drawings]

[0014] [Figure 1] 1 is a diagram illustrating an example of the configuration of a cell-image analyzing apparatus according to a first embodiment. [Diagram 2] 1 is a block diagram showing a hardware configuration of a cell-image analyzing apparatus according to a first embodiment. [Diagram 3] 1 is a flowchart outlining a cell image analyzing method according to a first embodiment. [Figure 4] 11A to 11C are diagrams for explaining an example of specific processing when estimation is performed by interpolation processing in the cell image analyzing method according to the first embodiment. [Diagram 5] 11A to 11C are diagrams for explaining an example of a specific process when estimation is performed by extrapolation processing in the cell image analyzing method according to the first embodiment. [Figure 6]FIG. 11 is a diagram illustrating an example of the configuration of a cell-image analyzing apparatus according to an example of a second embodiment. [Figure 7] 13 is a flowchart outlining an outline of a cell image analyzing method according to an example of a second embodiment. [Figure 8] 13A to 13C are diagrams for explaining an example of a specific process when estimation is performed by an interpolation process in a cell image analyzing method according to an example of the second embodiment. [Figure 9] FIG. 13 is a diagram illustrating an example of the configuration of a cell-image analyzing apparatus according to a modified example of the second embodiment. [Figure 10] 13 is a flowchart outlining an outline of a cell image analyzing method according to a modified example of the second embodiment. [Figure 11] FIG. 1 shows examples of the results of evaluating the culture state of colonies of iPS cell lines. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] Hereinafter, the embodiments of the present invention will be described in detail with reference to the accompanying drawings. Note that the embodiments exemplified below do not limit the present invention according to the claims, and all of the combinations of features described in the embodiments are not necessarily essential to the solution of the present invention.

[0016] First Embodiment The cell image analyzing device according to the present embodiment is a device that extracts image features from time-series images of a cell colony, and estimates image features of the same cell colony at a predetermined time point when no time-series images have been acquired, using the image features thus obtained. Hereinafter, a cell colony may simply be referred to as a colony.

[0017] FIG. 1 shows an example of the configuration of an apparatus for carrying out a cell image analyzing method according to this embodiment. As shown in FIG. 1, the cell image analysis device 10 according to the present embodiment includes an image acquisition unit 110, a region extraction unit 120, a feature extraction unit 130, a feature estimation unit 140, an evaluation unit 150, an instruction acquisition unit 160, and a storage unit 170. The cell image analysis device 10 is externally connected to a data server 30 and an output device 40. The imaging device 20 is, for example, a phase-contrast microscope having an imaging function and a cell incubation function, and has a function of generating time-series images of a cell colony by imaging, for example, every 24 hours while culturing the cells. The output device 40 is a device that outputs information from the cell image analysis device 10 to the outside, and is typically a user interface for presenting information to a user. Examples of the output device 40 include a display and a speaker.

[0018] The data server 30 holds time-series images of cell colonies captured by the imaging device 20. The data server 30 is, for example, a recording medium such as a hard disk drive (HDD) or a solid state drive (SDD) connected to the imaging device 20 and the cell-image analyzing device 10 via a network. The data server 30 has a function of saving each of the cell colony images captured by the imaging device 20. The configuration of the cell-image analyzing device 10 according to this embodiment is not limited to the above configuration, and for example, the cell-image analyzing device 10 may include some or all of the imaging device 20, the data server 30, and the output device 40 therein.

[0019] 2 is a block diagram showing the hardware configuration of the cell-image analyzing device 10 according to this embodiment. The cell-image analyzing device 10 has a CPU 201, a RAM 202, a ROM 203, a HDD (Hard Disk Drive) 204, a communication I / F 205, and an input device 206. These components are connected to each other via a bus or the like.

[0020] The CPU 201 is a processor that reads out to the RAM 202 programs stored in the ROM 203 and the HDD 204, executes them, and performs calculation processing and controls each part of the cell-image analyzing device 10. The processing performed by the CPU 201 may include extraction of a cell colony region, estimation of image feature amounts, evaluation of the cell colony or culture conditions, and the like.

[0021] The RAM 202 is a volatile storage medium and functions as a work memory when the CPU 201 executes a program. The ROM 203 is a non-volatile storage medium and stores firmware and the like necessary for the operation of the cell-image analyzing device 10. The HDD 204 is a non-volatile storage medium and stores programs used in processes such as extraction of cell colony regions and estimation of image feature amounts in this embodiment, as well as image data and the like.

[0022] The communication I / F 205 is a communication device based on standards such as Wi-Fi (registered trademark), Ethernet (registered trademark), Bluetooth (registered trademark), etc. The communication I / F 205 is used for communication with the imaging device 20, other computers, etc.

[0023] The input device 206 is a device for inputting information to the cell-image analyzer 10, and is typically a user interface for a user to operate the cell-image analyzer 10. Examples of the input device 206 include a keyboard, a button, a mouse, a touch panel, and the like.

[0024] The hardware configuration of the cell image analyzer 10 described above is an example, and can be changed as appropriate. For example, examples of processors that can be mounted on the cell image analyzer 10 include GPU, ASIC, FPGA, etc., in addition to the above-mentioned CPU 201. A plurality of these processors may be provided, and the plurality of processors may perform processing in a distributed manner. The function of storing information such as image data in the HDD 204 may be provided in the data server 30, not in the cell image analyzer 10. The HDD 204 may be a storage medium such as an optical disk, a magneto-optical disk, or an SSD (Solid State Drive).

[0025] An example of a cell-image analyzing method using the cell-image analyzing apparatus 10 according to this embodiment will be described below with reference to a flow chart of FIG.

[0026] First, in step S300, which is an image acquisition step, the image acquisition unit 110 receives a processing request from a user and acquires time-series images of cell colonies captured by the imaging device 20 and stored in the data server 30. Here, the time-series images of cell colonies are images captured at two or more different time points in the time series of cell culture, each of which includes the same cell colony. Examples of the time-series images of cell colonies include time-series images captured at predetermined intervals by whole-well imaging, which captures the entire well from the time the cells are seeded to the time the cells are subcultured. The image acquisition unit 110 is connected to the data server 30 via a network or the like, and can acquire multiple time-series images of cell colonies stored in the data server 30.

[0027] When acquiring each time-series image, the image acquiring unit 110 also acquires a time stamp associated with each image.

[0028] The cell image analysis device 10 has an instruction acquisition unit 160 that can accept instruction information input by a user, and the image acquisition unit 110 may acquire time-series images from the data server 30 based on the instruction information acquired by the instruction acquisition unit 160.

[0029] In addition, the time series images of the cell colony may be stored in advance in the memory unit 170 of the cell image analysis device 10, and the image acquisition unit 110 may acquire the time series images of the cell colony that have been stored in advance from the memory unit 170.

[0030] Next, in step S310, which is a region extraction step, the region extraction unit 120 extracts the region of the same cell colony included in each of the time-series images of the cell colony acquired by the image acquisition unit 110 in step S300 across each image frame. Thereafter, the same colony is identified across each image frame for each extracted colony region by a tracking process or the like. At this time, the colony from which the region is extracted is not limited to only one type of colony, and regions can be extracted from multiple types of colonies as necessary. The colony to be specifically extracted may be selected based on instruction information input by the user and acquired by the instruction acquisition unit 160, or, for example, a process of automatically extracting all colonies contained in the well may be performed. Also, a process of automatically extracting only colonies that satisfy a predetermined condition, rather than all colonies contained in the well, may be performed. At this time, extraction is performed so that the same cell colony across each image frame can be distinguished from different cell colonies by assigning a colony ID to each colony, or the like.

[0031] Here, the same cell colony across each image frame refers to a colony formed by the same cells dividing and growing during the culture process after the cells are seeded for culture. For example, if five cells are seeded at the time of seeding and each cell divides to form a cell colony, five independent cell colonies are formed, and each colony is distinguished as a different cell colony across each image frame.

[0032] As a specific example of the region extraction process, it can be realized by region division process for each image frame included in the time series images of the cell colony. In the region division process, the tendency of the pixel values ​​of the cell colony to be extracted and the pixel values ​​of the background are different, and the colony region and background region are divided by forming a region by integrating adjacent pixels having similar pixel values ​​for each pixel.

[0033] As a specific example of the tracking process, for each colony area of ​​each image frame included in the cell colony time series images, it is possible to determine whether it is the same colony as the colony in the previous frame based on the overlap amount with each colony area in the previous frame. The calculation of the overlap amount will be described below when the cell colony area area included in the current image frame is S(A1) and the cell colony area area included in the previous image frame is S(A2). The overlap amount of the cell colony in this case is calculated by the following formula.

number

[0034] The colony area with the largest overlap area calculated by the above formula is determined to be the same colony area. The above same colony discrimination method is only an example, and other means, such as a method for obtaining optical flow such as the Lucas-Kanade method, may be used to discriminate the same colony.

[0035] The area information of each colony area extracted in each image frame may be stored in the storage unit 170 in the form of, for example, cropping the target colony area from the cell colony time-series images acquired by the image acquisition unit 110. Also, it may be stored in the storage unit 170 as an image mask expressing the contour shape of the target colony area.

[0036] Next, in step S320, which is a feature extraction step, the feature extraction unit 130 extracts image features from the cell colony region extracted by the region extraction unit 120. Here, the image feature is a numerical representation of information on each feature obtained from the cell colony image. The type of feature extracted by the feature extraction unit 130 is not limited to one type, and two or more types of image features corresponding to two or more types of features may be extracted. The image feature extracted by the feature extraction unit 130 may be an image feature corresponding to a morphological feature such as the area, length, perimeter, or circularity of the target colony, or a texture feature of the target colony may be extracted as an image feature.

[0037] Examples of texture features include the average value, variance, contrast, skewness, kurtosis, energy, entropy, etc. of pixels derived from the pixel histogram or density co-occurrence matrix of the target colony. In addition, a power spectrum obtained by subjecting the target colony image to a discrete Fourier transform may be extracted as an image feature. The above image features are merely examples, and the image features to be extracted are not limited to these. In addition, new image features may be synthesized from these image features by principal component analysis or the like.

[0038] The feature extraction unit 130 extracts image features for each of the cell colony regions extracted from each image frame by the region extraction unit 120. At this time, the feature extraction unit 130 may extract image features from an image in which only the colony region from which the image features are to be extracted is masked. Alternatively, the feature extraction unit 130 may extract image features by combining the original image acquired by the image acquisition unit 110 and information on the cell colony region extracted by the region extraction unit 120. When extracting image features in the feature extraction unit 130, an appropriate extraction method can be selected depending on the type of image feature to be extracted.

[0039] Next, in step S330 which is a feature estimation step, the feature estimation unit 140 estimates image features of a cell colony at a predetermined time point when no time-series images have been acquired, using the image features extracted by the feature extraction unit 130. The image features used by the feature estimation unit 140 during estimation (image features extracted by the feature extraction unit 130) are image features extracted of the same cell colony as the cell colony whose image features are to be estimated.

[0040] In step S330, the predetermined time point at which the time-series images related to the estimation by the feature estimation unit 140 have not been acquired may be a time point determined in advance. Furthermore, the instruction information acquired by the instruction acquisition unit 160 may include information specifying the predetermined time point at which the time-series images have not been acquired, and the feature estimation unit 140 may estimate the image feature amount for the same cell colony at the time point specified by the instruction information. At this time, the instruction information acquired by the instruction acquisition unit 160 may be stored in the storage unit 170, or may be input by a user via an arbitrary input interface.

[0041] When the above-mentioned specified time point at which no time-series images have been acquired is within a time range that includes the time point at which the time-series images were acquired in the cell culture time series, the feature estimation unit 140 performs estimation by interpolation processing.

[0042] A specific example of processing when the feature estimation unit 140 performs estimation by interpolation processing will be described with reference to FIG. 4. FIG. 4 shows an example in which the image feature extracted by the feature extraction unit 130 in step S320 and estimated by the feature estimation unit 140 in step S330 is the area value of the same cell colony. In the example shown in FIG. 4, the area value of the same cell colony as the image feature is extracted from time-series images acquired every 24 hours from the seeding of cells. More specifically, the area value of the same cell colony as the image feature is extracted based on time-series images consisting of four image frames captured at 24 hours, 48 ​​hours, 72 hours, and 96 hours from the seeding. In the example shown in FIG. 4, the area of ​​the cell colony at a time 60 hours after the seeding, which is defined as the above-mentioned predetermined time when the time-series images are not acquired, is estimated. Note that the example shown in FIG. 4 is merely an example for explanation, and the image feature to be estimated is not limited to the area value of the cell colony, and may be other morphological features, texture features, or a combination of a plurality of these features.

[0043] In the estimation process, the feature estimation unit 140 selects a plurality of values ​​to be used for estimation from the image feature extracted by the feature extraction unit 130. The values ​​selected at this time are preferably image feature values ​​possessed by time-series images acquired at two or more consecutive time points including time points immediately before and after the above-mentioned predetermined time point at which no time-series images have been acquired. Specifically, in the example shown in FIG. 4, in order to estimate the image feature value at 60 hours after sowing, it is preferable to use image feature values ​​extracted from image frames acquired at two or more consecutive time points including image frames acquired 48 hours and 72 hours after sowing.

[0044] Here, we explain the estimation process using image feature amounts extracted from two image frames acquired 48 hours and 72 hours after sowing.

[0045] In the example shown in Figure 4, the area value of the cell colony extracted from the image frame 48 hours after seeding was 30 μm 2and the area value of the cell colony extracted from the image frame 72 hours after seeding was 70 μm 2 In the example shown in Fig. 4, the feature estimation unit 140 assumes that there is a linear relationship between the area value of the cell colony and the culture time from 48 hours after seeding to 72 hours after seeding, and that the area value of the cell colony 60 hours after seeding is 50 µm 2 It is estimated that.

[0046] In the above, an example was described in which it was assumed that the area value of the cell colony and the culture time have a linear relationship when the feature estimation unit 140 performs the estimation, but how the image feature value is assumed to change according to the culture time when performing the estimation is not limited to the above example. The assumption about the mode of change of the image feature value according to the culture time can be appropriately changed according to the feature corresponding to the image feature value to be estimated and the specific value of the image feature value used for the estimation. For example, if it is already known or can be expected that the image feature value to be estimated follows a specific function with respect to the culture time, the specific function can be used for the estimation. Also, for example, when image feature values ​​extracted from image frames acquired at three or more consecutive time points are used for the estimation, a polynomial approximation may be performed on the relationship between the elapsed time after seeding and the image feature value, and the obtained polynomial may be used for the estimation. In the example shown in FIG. 4, specifically, for example, the area value at 24 hours, 48 ​​hours, 72 hours, and 96 hours after seeding may be used to perform spline interpolation to estimate the area value at 60 hours after seeding. Compared to linear interpolation, spline interpolation can obtain smoother interpolation results because each data point is connected by a smooth curve whose curvature changes continuously. In addition, it is also possible to perform interpolation by expressing the relationship between the image capture time and the area value as a polynomial using Lagrange interpolation, etc.

[0047] In the above example, the image feature amounts extracted from the image frames acquired at two points in time immediately before and after the predetermined point in time at which no time-series images are acquired are used for estimation, but the combination of image frames used for estimation is not limited to this. For example, the feature amount estimation unit 140 may use all image feature amounts extracted from all image frames included in the time-series images for estimation.

[0048] In the example shown in Fig. 4, in estimating image feature amounts at timings included in a time range in which time-series images are captured, an interpolation process is performed using image feature amounts extracted from image frames immediately before and after the timing. However, in the present invention, the timings for which image feature amounts are estimated are not limited to within the time range in which time-series images are captured, and image feature amounts at timings included outside the time range in which time-series images are captured may be estimated. In other words, when the above-mentioned predetermined time point at which no time-series images are acquired is outside the time range including the time points at which time-series images are acquired in the time series in cell culture, the feature amount estimation unit 140 performs estimation by extrapolation process.

[0049] A specific example of processing when the feature estimation unit 140 performs estimation by extrapolation processing will be described with reference to FIG. 5. In the example shown in FIG. 5, the processing will be described when the image feature extracted by the feature extraction unit 130 in step S320 and estimated by the feature estimation unit 140 in step S330 is the area value of the same cell colony, as in the example shown in FIG. 4 above. In the example shown in FIG. 5, the area value of the same cell colony as the image feature is extracted based on time-series images consisting of four image frames captured at 24 hours, 48 ​​hours, 72 hours, and 96 hours after seeding, as in the example shown in FIG. 4 above. In the example shown in FIG. 5, the area of ​​the cell colony at 120 hours after seeding, which is defined as the above-mentioned predetermined time point at which time-series images are not acquired, is estimated.

[0050] In the estimation process, the feature estimation unit 140 selects a plurality of values ​​to be used for estimation from the image feature extracted by the feature extraction unit 130. The values ​​selected at this time are preferably image feature values ​​possessed by time-series images acquired at two or more consecutive time points immediately before the above-mentioned predetermined time point at which no time-series images have been acquired. Specifically, in the example shown in FIG. 5, it is preferable to use image feature values ​​extracted from image frames acquired at two or more consecutive time points including image frames acquired 72 hours and 96 hours after sowing.

[0051] Here, we explain the process of extrapolating the colony area 120 hours after seeding using image features extracted from two image frames acquired 72 hours and 96 hours after seeding.

[0052] In the example shown in Figure 5, the area value of the cell colony extracted from the image frame 72 hours after seeding was 70 μm 2 and the area value of the cell colony extracted from the image frame 96 hours after seeding was 150 μm 2 5, the feature amount estimation unit 140 assumes that there is a linear relationship between the area value of the cell colony and the culture time from 72 hours after seeding to 120 hours after seeding, and the area value of the cell colony 120 hours after seeding is 230 μm 2 It is estimated that.

[0053] When the feature estimation unit 140 performs estimation by extrapolation, how the image feature changes according to the culture time is assumed in the estimation is not limited to the above example, as in the case of the example of estimation by interpolation shown in Fig. 4. In other words, the assumption about the mode of change of the image feature according to the culture time can be changed as appropriate according to the feature corresponding to the image feature to be estimated and the specific value of the image feature used for the estimation.

[0054] In addition, even when estimating by extrapolation, the combination of image frames used for estimation is not limited to the above, as in the case of estimating by interpolation, and for example, all image frames included in the time-series images may be used.

[0055] In the example shown in FIG. 5, when performing estimation by extrapolation, the case where the culture time and the image feature amount are assumed to have a linear relationship has been described. However, in the case of estimation by extrapolation, similar to the case of estimation by interpolation, a relationship other than a linear relationship can be used for the relationship between the image feature amount and the culture time. For example, spline interpolation can be performed using the area values ​​at 24 hours, 48 ​​hours, 72 hours, and 96 hours after seeding, and the area value at 120 hours after seeding can be estimated using the obtained spline. In addition, extrapolation processing can be performed using a polynomial that expresses the relationship between the imaging time and the area value by Lagrange interpolation or the like.

[0056] When the feature estimation unit 140 performs estimation by extrapolation, if the above-mentioned predetermined time point at which the time-series image is not acquired is far from the time range including the time point at which the time-series image is acquired, it may be difficult to make an appropriate estimation. Therefore, the feature estimation unit 140 may have a function of generating a result indicating that estimation is impossible when the above-mentioned predetermined time point at which the time-series image is not acquired is far from the time range including the time point at which the time-series image is acquired by a predetermined time or more. Here, the extent to which the above-mentioned predetermined time point at which the time-series image is not acquired is far from the time range including the time point at which the time-series image is acquired to generate a result indicating that estimation is impossible can be appropriately set according to the specific situation.

[0057] By performing estimation by extrapolation in the feature estimation unit 140, it is possible to estimate the area value of a colony at an image capture timing outside the time range including the time point at which the time series images were acquired. This makes it possible, for example, to predict the future area value of a colony at an early stage of culture, or to verify the original growth level of a colony whose culture has failed midway due to an external factor.

[0058] When the feature amount extraction section 130 extracts two or more types of image feature amounts, the feature amount estimation section 140 can also estimate two or more types of image feature amounts.

[0059] Next, in step S340, which is an evaluation step, the evaluation unit 150 performs evaluation of at least one selected from the group consisting of cell colonies and cell culture conditions, using the image features estimated by the feature estimation unit 140. The evaluation unit 150 can evaluate the image features to be evaluated based on an evaluation criterion, and may determine whether the image features of the cell colony estimated by the feature estimation unit 140 exceed a predetermined reference value included in the evaluation criterion. Examples of the contents of the evaluation include determining whether the colony is currently differentiated or non-differentiated, and evaluating the colony properties such as the differentiation ability and growth rate of the target colony.

[0060] Hereinafter, an example will be described in which the evaluation unit 150 analyzes the growth rate of a colony based on the area value estimated by the feature amount estimation unit 140. In the example described here, the evaluation unit 150 performs an analysis of the growth rate of a colony by determining whether the area value of the colony is 60 μm 40 hours after the seeded colony starts to grow. 2 The criterion used in the evaluation in step S340 is not limited to the area value, and may be other morphological features, texture features, or a combination of a plurality of these features.

[0061] In general, cells do not necessarily start growing immediately after seeding, and the timing at which they start growing differs for each single cell that becomes the source of each colony. Therefore, in the case of an imaging setting in which imaging is performed discontinuously, for example, every 24 hours, as in the example described above using Figures 4 and 5, time-series images captured exactly 40 hours after the colony started to grow are not necessarily acquired.

[0062] Therefore, for example, the user inputs instruction information specifying that the feature estimation unit 140 should estimate the image feature amount 40 hours after the seeded colony starts to grow, via an arbitrary input interface. The cell-image analysis device 10 acquires the instruction information by the instruction acquisition unit 160, and the feature estimation unit 140 estimates the area of ​​the cell colony based on the instruction information. If the target colony starts to grow 20 hours after seeding, then in the example shown in FIG. 4, 40 hours after the seeded colony starts to grow corresponds to 60 hours after seeding. Therefore, the feature estimation unit 140 estimates the area of ​​the cell colony 60 hours after seeding, and the area value is 50 μm 2 The evaluation unit 150 estimates that the area value of the seeded colony is 60 μm 40 hours after the seeded colony starts to grow. 2 Since the evaluation is based on the criterion that if the growth rate is below 100%, it is judged to be a growth-deficient colony, the colony to be evaluated can be evaluated as a growth-deficient colony. In the above example, the information that "the target colony started to grow 20 hours after seeding" may be included in the instruction information, or the cell-image analyzer 10 may have a function of automatically predicting the growth rate based on the image feature extracted by the feature extractor 130.

[0063] Although the above describes an example in which evaluation is performed using one type of estimated image feature, it is also possible to estimate two or more types of image feature in feature estimation unit 140 and perform evaluation using the two or more types of image feature thus estimated.

[0064] The evaluation unit 150 may evaluate the degree of correlation between the image feature amount estimated by the feature amount estimation unit 140 and at least one selected from the group consisting of predetermined features of the cell colony and cell culture conditions. Examples of the predetermined features of the cell colony include the type of cell line, production or non-production of a specific substance, and the presence or absence of cancer. Examples of the cell culture conditions include the presence or absence or concentration of a specific medium component, pH, temperature, humidity, the presence or absence or concentration of a specific atmospheric composition component such as carbon dioxide gas, and illuminance.

[0065] When the evaluation unit 150 evaluates the degree of correlation, for example, two or more cell cultures are performed in which the predetermined characteristics of the cell colonies or the cell culture conditions, which are the targets of the evaluation of the degree of correlation, are different from each other. Then, time-series images are captured in each cell culture, and the cell-image analysis method according to this embodiment is carried out using the cell-image analysis device 10 according to this embodiment.

[0066] A specific example will be described below in which the image feature amount used in evaluating the degree of correlation in the evaluation unit 150 is the area of ​​a cell colony, and the cell culture condition to be evaluated is pH.

[0067] First, for a certain cell line a, cell culture A is performed at a pH of 5.0, and cell culture B is performed at a pH of 7.0, and time-series images of each culture are captured by the imaging device 20. Next, the image analysis described above is performed on the time-series images captured in each cell culture. At this time, the area of ​​the cell colony 40 hours after the seeded colony starts to grow is estimated by the feature estimation unit 140.

[0068] Next, the evaluation unit 150 compares the area A of the cell colony estimated for cell culture A with the area B of the cell colony estimated for cell culture B, and can evaluate that the greater the difference between area A and area B, the greater the correlation between the cell colony area and pH. Here, the degree of correlation can be evaluated more accurately by using the average value of areas estimated from a plurality of mutually different cell colonies for area A and area B. By performing such evaluation, it becomes possible to evaluate the effect of pH during culture on the growth rate of cell line a based on the evaluation results obtained by the evaluation unit 150.

[0069] In the above example, the evaluation unit 150 uses only one type of image feature, the colony area, when evaluating the degree of correlation. However, a value obtained by combining two or more types of image features using principal component analysis or the like may be used for the evaluation.

[0070] The state evaluation unit 150 can output the evaluation result to the output device 40 via a network or the like, and present the evaluation result to the user.

[0071] Second Embodiment In the first embodiment, the feature estimation unit 140 estimates image feature amounts using image feature amounts extracted from images captured at different times among time-series images acquired by the image acquisition unit 110. However, in the method according to the first embodiment, when it is desired to compare image feature amounts of different colonies with the same degree of growth of the colonies, the user needs to specify the optimal imaging timing for comparison of the respective colonies, which requires a lot of effort.

[0072] In the method according to the second embodiment described below, when an image feature reflecting the degree of colony growth, such as a colony area value, takes a predetermined value, other image feature values ​​are compared, which makes it possible to easily compare the colonies with their respective growth degrees brought closer together.

[0073] 6 is a diagram showing an example of the configuration of an apparatus for carrying out a cell-image analyzing method according to an example of the second embodiment. A cell-image analyzing apparatus 60 according to an example of this embodiment has the same configuration as the cell-image analyzing apparatus 10 according to the first embodiment, except that it has a time point estimation unit 600. The hardware configuration of the cell-image analyzing apparatus 60 can be the same as that of the cell-image analyzing apparatus 10 according to the first embodiment.

[0074] An example of a cell image analysis method according to an example of this embodiment will be described below with reference to the flowchart of Fig. 7. In Fig. 7, steps in which the same processing as in the cell image analysis method according to the first embodiment is performed are given the same step numbers as in Fig. 3. There is substantially no difference between the processing in steps S300, S310, and S340 in the cell image analysis method according to an example of this embodiment and the cell image analysis method according to the first embodiment, and therefore the description thereof will be omitted.

[0075] In one example of this embodiment, in step S700, the feature extraction unit 130 extracts a first image feature and a second image feature from the cell colony region extracted by the region extraction unit 120. Here, the first image feature is an image feature corresponding to a feature that changes depending on the culture time of the cell culture. Also, the second image feature is an image feature corresponding to a feature used for evaluation by the user based on the result obtained by the evaluation unit 150 in the subsequent processing.

[0076] As the first image feature, an image feature related to the size of a colony can be preferably used. Specifically, the first image feature is preferably at least one selected from the group consisting of a colony area, a colony volume, a colony contour length, and an average cell area.

[0077] The first image feature and the second image feature extracted by the feature extraction unit 130 are not limited to one type each, and the feature extraction unit 130 may extract two or more types of image feature as the first image feature or the second image feature.

[0078] Next, in step S710, which is a time point estimation step, the time point estimation unit 600 estimates a target time point, which is a time point at which a first image feature amount possessed by a cell colony becomes a predetermined value. Specifically, the time point estimation unit 600 estimates the target time point from the first image feature amount extracted by the feature amount extraction unit 130 and a time stamp associated with a time series image including a region of a cell colony from which each of the first image feature amounts was extracted. Note that the time point estimation unit 600 does not necessarily need to estimate the target time point up to a specific time point, and may only estimate, for example, a relative positional relationship in the time series between the target time point and the time point at which the time series image is acquired.

[0079] The predetermined value for the first image feature may be a value that has been determined in advance. Alternatively, the instruction information acquired by the instruction acquisition unit 160 may include information that specifies the predetermined value for the first image feature, and the feature estimation unit 140 may estimate the image feature based on the instruction information. In this case, the instruction information acquired by the instruction acquisition unit 160 may be stored in the storage unit 170, or may be input by the user via an arbitrary input interface.

[0080] Next, in step S720, which is a feature estimation step, the feature estimation unit 140 determines an image frame to be used to estimate a second image feature from the information about the target time point estimated by the time point estimation unit 600 in step S710, and performs the estimation.

[0081] When the target time point is within a time range including the time point at which the time-series images were acquired on the time series in the cell culture, the feature estimation unit 140 estimates a second image feature amount corresponding to the predetermined value for the first image feature amount by an interpolation process. At this time, it is preferable that the feature estimation unit 140 performs the interpolation process using second image feature amounts possessed by time-series images acquired at two or more consecutive time points including the time points immediately before and after the target time point.

[0082] When the target time point is outside the time range including the time point at which the time-series images were acquired on the time series in the cell culture, the feature estimation unit 140 estimates the second image feature amount corresponding to the predetermined value for the first image feature amount by extrapolation. At this time, it is preferable that the feature estimation unit 140 performs the extrapolation process using the second image feature amount possessed by the time-series images acquired at two or more consecutive time points on the time series immediately before the target time point.

[0083] A specific example of processing when the feature estimation unit 140 performs estimation by interpolation processing will be described with reference to FIG. 8. FIG. 8 shows an example in which the first image feature extracted by the feature extraction unit 130 in step S320 is the area value of the cell colony, and the second image feature is a contrast value calculated from a histogram of the image texture. In the example shown in FIG. 8, the first image feature and the second image feature are extracted from time-series images acquired every 24 hours from cell seeding. More specifically, each image feature is extracted based on time-series images consisting of four image frames captured 24 hours, 48 ​​hours, 72 hours, and 96 hours after seeding. In the example shown in FIG. 8, the above-mentioned predetermined value for the colony area as the first image feature is 50 μm 2 The contrast value when is estimated. Note that the first image feature value serving as a criterion for the degree of growth is not limited to the area value, and may be other morphological features or texture features. Also, the second image feature value to be estimated is not limited to the contrast value, and may be other morphological features or texture features, or a combination of a plurality of these features.

[0084] At this time, the time point estimation unit 600 determines that the colony area is 50 μm2 or less based on the area values ​​extracted by the feature extraction unit 130 and the elapsed time since seeding corresponding to each area value. 2 The time point at which the time elapsed since sowing occurs is estimated to be between 48 hours and 72 hours. The method of estimation by the time point estimation unit 600 is not particularly limited, and can be appropriately determined according to the features corresponding to the image features to be estimated and the specific values ​​of the image features used for the estimation.

[0085] The feature estimation unit 140 selects a plurality of image frames necessary for estimating the second image feature, based on the information of the target time point estimated by the time point estimation unit 600. Since the target time point estimated by the time point estimation unit 600 is between 48 hours and 72 hours after sowing, the image frames captured 48 hours and 72 hours after sowing, which are the time points before and after the target time point, are used here.

[0086] In one example of this embodiment, the interpolation process performed by the feature estimation unit 140 is similar to the interpolation process performed by the feature estimation unit 140 in the first embodiment, and only the criterion used for estimation is different. That is, in the estimation process in the first embodiment, estimation is performed based on information about a time point linked to a time-series image as a criterion, whereas in this embodiment, estimation is performed based on a first image feature.

[0087] In the above example, the first image feature (area) 48 hours after seeding is 30 μm 2 The second image feature amount (contrast) at this time is 1300. The first image feature amount (area) 72 hours after seeding is 70 μm 2 and the second image feature (contrast) at this time is 1400. Therefore, as shown in FIG. 8, if it is assumed that there is a linear relationship between the area of ​​the cell colony and the contrast between 48 hours and 72 hours after seeding, the above-mentioned predetermined value for the first image feature is 50 μm 2 When , the contrast is estimated to be 1350. Note that the combination of image frames used for estimation is not limited to the above example. For example, all image frames included in the time-series images may be used.

[0088] The method of estimation by interpolation processing has been described in detail above, but estimation by extrapolation processing can also be realized in the same manner as shown in the first embodiment, except that only the criteria used for estimation are different.

[0089] In addition, although the above description has been given as an example in which the cell image analysis device 60 is equipped with a time point estimation unit 600 that estimates the target time point, the feature estimation unit 140 may have the functions of the time point estimation unit 600 described above.

[0090] (Modification of the second embodiment) In the second embodiment, instead of using the function of the time point estimation unit 600 described in the above example, a regression model obtained from the first image feature amount and the second image feature amount extracted by the feature amount extraction unit 130 may be used. In this case, the feature amount estimation unit 140 can estimate the second image feature amount corresponding to a predetermined value for the first image feature amount based on the regression model created by the regression model creation unit.

[0091] 9 is a diagram showing an example of the configuration of an apparatus for carrying out a cell-image analyzing method according to a modified example of the second embodiment. A cell-image analyzing apparatus 90 according to this modified example has the same configuration as the cell-image analyzing apparatus 60, except that it has a regression model creating unit 900 instead of the time point estimating unit 600. The hardware configuration of the cell-image analyzing apparatus 90 can be the same as that of the cell-image analyzing apparatus 10 according to the first embodiment.

[0092] Fig. 10 is a flow chart showing an outline of an example of the cell image analysis method according to this modification. In Fig. 10, the same step numbers as in Fig. 7 are given to steps that perform the same processes as in the cell image analysis method described for the second embodiment with reference to Fig. 7. Since there is no substantial difference in the processes in steps S300, S310, S340, and 700 between the cell image analysis method according to this modification and the cell image analysis method described for the second embodiment with reference to Fig. 7, the description thereof will be omitted.

[0093] In this modification, in step S1000, which is a regression model creation step, regression model creation unit 900 creates a regression model using the first image feature and the second image feature extracted by feature extraction unit 130. Then, in step S1010, which is a feature estimation step, feature estimation unit 140 estimates a second feature corresponding to the above-mentioned predetermined value for the first image feature, using the regression model created in step S1000.

[0094] For example, in the case of the first image feature (area) and the second image feature (contrast) extracted by the feature extraction unit 130 shown in Fig. 8, the regression model creation unit 900 creates a regression model for the relationship between area and contrast. For example, when a regression model of "(contrast) = 181 x Ln (area) + 620" is created, the feature estimation unit 140 uses the regression model to estimate the area of ​​50 μm 2 When , the contrast can be estimated to be 1328. Here, the regression model creation unit 900 does not necessarily need to use all of the first image feature amount and the second image feature amount to create the regression model, and may perform processing such as removing values ​​that are thought to include clearly large errors.

[0095] The regression model creation unit 900 may have a function of evaluating the validity or accuracy of the created regression model. In this case, only when the created regression model has a predetermined level of accuracy or higher, it may be used in the estimation process in the feature estimation unit 140. Furthermore, when there are a plurality of first image feature amounts and / or a plurality of second image feature amounts, the regression model creation unit 900 may create regression models for a plurality of combinations. Furthermore, in this case, the validity of the created plurality of regression models may be evaluated, and only the regression models having a predetermined level of accuracy or higher may be used in the estimation process in the feature estimation unit 140.

[0096] Although the above description has been given as an example in which the cell image analysis device 90 further includes a regression model creation unit 900 that creates a regression model, the feature estimation unit 140 may have the functions of the regression model creation unit 900 described above.

[0097] It should be noted that the above-mentioned embodiments are merely examples of the embodiment of the present invention, and the technical scope of the present invention should not be interpreted as being limited by these. In other words, the present invention can be implemented in various forms without departing from its technical idea or main features. For example, it should be understood that an embodiment in which a part of the configuration of any of the embodiments is added to another embodiment, or an embodiment in which a part of the configuration of any of the embodiments is replaced with a part of the configuration of another embodiment, is also an embodiment to which the present invention can be applied.

[0098] The present invention may be realized by executing the following process. That is, software (programs) for causing a computer (or a CPU, MPU, etc.) to execute one or more functions of the above-described embodiments is supplied to a system or device via a network or various storage media. The computer (or a CPU, MPU, etc.) of the system or device then reads and executes the program. The present invention may also be realized, for example, by a circuit (e.g., ASIC) that executes one or more functions.

[0099] A cell image analysis program according to one aspect of the present invention is a program for causing a computer to execute the cell image analysis method according to the present invention.

[0100] A recording medium according to another aspect of the present invention stores the above-mentioned cell image analysis program in a computer-readable format. EXAMPLES

[0101] An example will be described below showing that the use of the cell image analysis method according to the present invention enables accurate evaluation based on image feature amounts extracted from time-series images.

[0102] Three types of iPS cell lines, A, B, and C, were used for the evaluation. These three types of iPS cell lines were generated from different cell types, and are known to have differences in their properties, such as differentiation potential.

[0103] The iPS cell lines were cultured with some modifications of the protocol published on the website of the Center for iPS Cell Research and Application, Kyoto University (https: / / www.cira.kyoto-u.ac.jp / j / research / img / protocol / hipsprotocolFf_140311.pdf). Specifically, a 6-well cell culture plate (Corning, Tissue culture treated) and StemFit AK03N (Ajinomoto) were used, and the amount of medium during culture was 4.5 mL / well. Time-series images were taken using a BZ-X810 (Keyence) with a 10x objective lens every 24 hours from the 5th to 9th day after seeding. In order to remove floating dead cells, the plate was washed once with 1.5 mL / well of medium before image acquisition, and then 4.5 mL / well of medium was added before taking the time-series images.

[0104] Next, image features were extracted from the obtained time-series images, and the culture state of the colonies was evaluated. The evaluation was carried out based on the area value, which is the image feature value that is the standard for the growth degree of each colony, of 300,000 μm2 or more. 2 The purpose of this study was to compare the culture conditions between iPS cell lines, using the time point at which the cells reach maturity as a reference point.

[0105] As image features for the culture state, multiple image features were used, including morphological features such as colony circumference and circularity, as well as colony texture features, and the first principal component PC1 and second principal component PC2 were synthesized by principal component analysis and used for evaluation.

[0106] When the cell image analysis method according to the present invention is not used for the evaluation, the above-mentioned reference area value is exactly 300,000 μm 2There are no time-series images taken at the time when the area value is 300,000 μm, and it is not possible to obtain image features for the culture state at that time. 2 The image feature values ​​obtained from the time-series images captured at the time closest to the time when the saturation occurred were used for the evaluation.

[0107] As an example, an evaluation was performed by applying the cell image analysis method according to the second embodiment described above. At this time, the area value as the image feature quantity serving as a reference for the degree of colony growth was set as the first image feature quantity, and the first principal component PC1 and the second principal component PC2 obtained from the image feature quantity regarding the culture state were set as the second image feature quantity.

[0108] Fig. 11 shows examples of the results of evaluating the culture state of colonies of three iPS cell lines, where Fig. 11(a) shows the results obtained in a comparative example, and Fig. 11(b) shows the results obtained in an embodiment. Fig. 11 shows a scatter plot in which the first principal component PC1 and the second principal component PC2 obtained for the three iPS cell lines are plotted on the horizontal and vertical axes, respectively.

[0109] In FIG. 11(a) showing the results obtained in the comparative example, the area of ​​the colony from which the image feature used for the evaluation was extracted was not standardized to a standard value, so the deviation from the standard area value significantly affected the image feature for evaluating the culture state, and the difference between the strains was small. Therefore, it can be seen that it is difficult to classify each strain, such as by clustering, in the evaluation in the comparative example. On the other hand, in FIG. 11(b) showing the results obtained by applying the cell image analysis method according to the present invention, the variance of each strain is small, and the difference between the strains of the three types of iPS cells is large compared to the results obtained in the comparative example. In other words, the comparison between the evaluation results obtained in the example and the results obtained in the comparative example shows that the application of the cell image analysis method according to the present invention allowed for a highly accurate evaluation of the differences in the properties of the three types of iPS cells.

[0110] From the above results, it can be seen that by applying the cell image analysis method of the present invention to evaluate the state of colonies, it is possible to reduce the influence of area differences between each colony on other image features, making it possible to compare colonies with high accuracy.

[0111] Disclosure according to an embodiment of the present invention includes the following configurations and methods. (Configuration 1) an image acquisition unit that acquires time-series images of the same cell colony, the images being acquired at two or more different time points in a time series of cell culture; an area extraction unit that extracts an area of ​​the same cell colony included in each of the time-series images; a feature extraction unit that extracts image features from the region extracted by the region extraction unit; a feature amount estimation unit that estimates an image feature amount for the same cell colony at a predetermined time point when the time-series images are not acquired, using the image feature amount extracted by the feature amount extraction unit; an evaluation unit that evaluates at least one selected from the group consisting of the same cell colony and the cell culture conditions, using the image feature amount estimated by the feature amount estimation unit; A cell image analysis device comprising: (Configuration 2) An instruction acquisition unit for acquiring instruction information is further provided, the instruction information includes information specifying a predetermined time point at which the time-series images are not acquired, 2. The cell-image analyzing apparatus according to configuration 1, wherein the feature amount estimating unit estimates an image feature amount for the same cell colony at a time point specified by the instruction information. (Configuration 3) 3. The cell-image analyzing device according to configuration 1 or 2, wherein the feature estimation unit estimates image feature amounts for the same cell colony at a predetermined time point when the time-series images are not acquired by an interpolation process. (Configuration 4) The cell image analysis device according to configuration 3, wherein the feature estimation unit performs the interpolation process using the image feature amounts possessed by the time-series images acquired at two or more consecutive time points including time points immediately before and after a predetermined time point at which the time-series images have not been acquired. (Configuration 5) 3. The cell-image analyzing device according to configuration 1 or 2, wherein the feature estimation unit estimates image feature amounts for the same cell colony at a predetermined time point when the time-series images are not acquired by extrapolation processing. (Configuration 6) The cell image analysis device according to configuration 5, wherein the feature estimation unit performs the extrapolation process using the time-series images acquired at two or more consecutive time points on the time series immediately preceding a predetermined time point at which the time-series images are not acquired. (Configuration 7) an image acquisition unit that acquires time-series images of the same cell colony, the images being acquired at two or more different time points in a time series of cell culture; an area extraction unit that extracts an area of ​​the same cell colony included in each of the time-series images; a feature extraction unit that extracts, from the region extracted by the region extraction unit, a first image feature corresponding to a feature that changes depending on a culture time of the cell culture and a second image feature corresponding to a feature used for evaluation; a feature estimation unit that estimates the second image feature corresponding to a predetermined value for the first image feature; an evaluation unit that evaluates at least one selected from the group consisting of the same cell colony and the cell culture conditions, using the second image feature estimated by the feature estimation unit; A cell image analysis device comprising: (Configuration 8) 8. The cell-image analyzing apparatus according to claim 7, wherein the first image feature is an image feature related to a size of a colony. (Configuration 9) 9. The cell-image analyzing apparatus according to configuration 8, wherein the first image feature amount is at least one selected from the group consisting of a colony area, a colony volume, a colony contour length, and an average cell area. (Configuration 10) An instruction acquisition unit for acquiring instruction information is further provided, the instruction information includes information defining the predetermined value for the first image feature amount, The cell image analysis device according to any one of configurations 7 to 9, wherein the feature estimation unit estimates the second image feature based on the predetermined value for the first image feature defined by the instruction information. (Configuration 11) a time estimation unit that estimates a target time point that is a time point at which the first image feature amount of the same cell colony becomes the predetermined value, 11. The cell image analyzing device of any one of configurations 7 to 10, wherein when the target time point is within a time range including a time point on the time series at which the time-series images are acquired, the feature estimation unit estimates the second image feature amount corresponding to a predetermined value for the first image feature amount by an interpolation process. (Configuration 12) The cell image analysis device of configuration 11, wherein the feature estimation unit performs the interpolation process using the second image feature possessed by the time-series images acquired at two or more consecutive time points including the time points immediately before and after the target time point. (Configuration 13) a time estimation unit that estimates a target time point that is a time point at which the first image feature amount of the same cell colony becomes the predetermined value, The cell image analyzing device according to any one of configurations 7 to 10, wherein when the target time point is outside a time range including a time point on the time series that includes the time point at which the time-series images were acquired, the feature estimation unit estimates the second image feature amount corresponding to a predetermined value for the first image feature amount by extrapolation processing. (Configuration 14) The cell image analysis device according to configuration 13, wherein the feature estimation unit performs the extrapolation process using the second image feature possessed by the time-series images acquired at two or more consecutive time points on the time series immediately preceding the target time point. (Configuration 15) a regression model creation unit that creates a regression model using the first image feature amount and the second image feature amount; The cell image analysis device according to any one of configurations 7 to 10, wherein the feature estimation unit estimates the second image feature corresponding to a predetermined value for the first image feature based on the regression model created by the regression model creation unit. (Configuration 16) The cell image analysis device according to any one of configurations 1 to 15, wherein the evaluation unit evaluates at least one selected from the group consisting of the state of the identical cell colony and the cell culture conditions based on a predetermined evaluation criterion. (Configuration 17) 16. A cell image analysis device according to any one of configurations 1 to 15, wherein the evaluation unit evaluates the degree of correlation between the image feature estimated by the feature estimation unit and at least one selected from the group consisting of a predetermined feature possessed by the same cell colony and the cell culture conditions. (Method 1) an image acquiring step of acquiring time-series images acquired at two or more different time points in a time series of cell culture, each of which includes the same cell colony; a region extraction step of extracting a region of the same cell colony included in each of the time-series images; a feature extraction step of extracting image features from the region of the same cell colony in a plurality of images included in the time-series images; a feature amount estimation step of estimating an image feature amount for the same cell colony at a predetermined time point when the time-series images are not acquired, using the image feature amount extracted in the feature amount extraction step; an evaluation step of evaluating the same cell colony or the conditions of the cell culture using the image feature amount estimated in the feature amount estimation step; A cell image analysis method comprising: (Method 2) an image acquiring step of acquiring time-series images acquired at two or more different time points in a time series of cell culture, each of which includes the same cell colony; a region extraction step of extracting a region of the same cell colony included in each of the time-series images; a feature extraction step of extracting, from the region extracted in the region extraction step, a first image feature corresponding to a feature that changes depending on a culture time of the cell culture and a second image feature corresponding to a feature used for evaluation; a feature estimation step of estimating the second image feature corresponding to a predetermined value for the first image feature; an evaluation step of evaluating at least one selected from the group consisting of the same cell colony and the cell culture conditions, using the second image feature amount estimated in the feature amount estimation step; A cell image analysis method comprising: (Configuration 20) A cell image analysis program for causing a computer to execute the cell image analysis method according to Method 1 or 2. (Configuration 21) A recording medium storing the cell image analysis program according to aspect 20 in a computer-readable format. [Explanation of symbols]

[0112] 10, 60, 90: Cell image analysis device 20: Imaging device 30: Data Server 40: Output device 110: Image acquisition unit 120: Area extraction part 130: Feature extraction unit 140: Feature estimation unit 150: Evaluation section 600: Time Estimation Department 900: Regression model creation section

Claims

1. an image acquisition unit that acquires time-series images of the same cell colony, the images being acquired at two or more different time points in a time series of cell culture; an area extraction unit that extracts an area of ​​the same cell colony included in each of the time-series images; a feature extraction unit that extracts image features from the region extracted by the region extraction unit; a feature amount estimation unit that estimates an image feature amount for the same cell colony at a predetermined time point when the time-series images are not acquired, using the image feature amount extracted by the feature amount extraction unit; an evaluation unit that evaluates at least one selected from the group consisting of the same cell colony and the cell culture conditions, using the image feature amount estimated by the feature amount estimation unit; A cell image analysis device comprising:

2. An instruction acquisition unit for acquiring instruction information is further provided, the instruction information includes information specifying a predetermined time point at which the time-series images are not acquired, The cell-image analyzing device according to claim 1 , wherein the feature amount estimating section estimates an image feature amount for the same cell colony at a time point specified by the instruction information.

3. The cell-image analyzing device according to claim 1 , wherein the feature amount estimating section estimates the image feature amount for the same cell colony at a predetermined time point when the time-series images are not acquired by an interpolation process.

4. 4. The cell image analyzing device according to claim 3, wherein the feature estimation unit performs the interpolation process using the image feature amounts possessed by the time-series images acquired at two or more consecutive time points including time points immediately before and after a predetermined time point at which the time-series images have not been acquired.

5. The cell-image analyzing device according to claim 1 , wherein the feature amount estimating section estimates the image feature amount for the same cell colony at a predetermined time point when the time-series images are not acquired by extrapolation processing.

6. The cell image analysis device according to claim 5 , wherein the feature estimation unit performs the extrapolation process using the time-series images acquired at two or more consecutive time points on the time series immediately preceding a predetermined time point at which the time-series images have not been acquired.

7. an image acquisition unit that acquires time-series images of the same cell colony, the images being acquired at two or more different time points in a time series of cell culture; an area extraction unit that extracts an area of ​​the same cell colony included in each of the time-series images; a feature extraction unit that extracts, from the region extracted by the region extraction unit, a first image feature corresponding to a feature that changes depending on a culture time of the cell culture and a second image feature corresponding to a feature used for evaluation; a feature estimation unit that estimates the second image feature corresponding to a predetermined value for the first image feature; an evaluation unit that evaluates at least one selected from the group consisting of the same cell colony and the cell culture conditions, using the second image feature estimated by the feature estimation unit; A cell image analysis device comprising:

8. The cell-image analyzing apparatus according to claim 7 , wherein the first image feature amount is an image feature amount relating to a size of a colony.

9. 9. The cell-image analyzing apparatus according to claim 8, wherein the first image feature amount is at least one selected from the group consisting of a colony area, a colony volume, a colony contour length, and an average cell area.

10. An instruction acquisition unit for acquiring instruction information is further provided, the instruction information includes information defining the predetermined value for the first image feature amount, The cell-image analyzing apparatus according to claim 7 , wherein the feature amount estimating section estimates the second image feature amount based on the predetermined value for the first image feature amount defined by the instruction information.

11. a time estimation unit that estimates a target time point that is a time point at which the first image feature amount of the same cell colony becomes the predetermined value, 8. The cell image analyzing device according to claim 7, wherein when the target time point is within a time range including a time point on the time series at which the time-series images are acquired, the feature estimation unit estimates the second image feature amount corresponding to a predetermined value for the first image feature amount by an interpolation process.

12. The cell image analysis device according to claim 11, wherein the feature estimation unit performs the interpolation process using the second image feature possessed by the time-series images acquired at two or more consecutive time points including time points immediately before and after the target time point.

13. a time estimation unit that estimates a target time point that is a time point at which the first image feature amount of the same cell colony becomes the predetermined value, 8. The cell image analyzing device according to claim 7, wherein when the target time point is outside a time range on the time series that includes a time point at which the time-series images were acquired, the feature estimation unit estimates the second image feature amount corresponding to a predetermined value for the first image feature amount by extrapolation processing.

14. The cell image analysis device according to claim 13, wherein the feature estimation unit performs the extrapolation process using the second image feature of the time-series images acquired at two or more consecutive time points on the time series immediately before the target time point.

15. a regression model creation unit that creates a regression model using the first image feature amount and the second image feature amount; The cell image analyzing device according to claim 7 , wherein the feature estimation unit estimates the second image feature corresponding to a predetermined value for the first image feature based on the regression model created by the regression model creation unit.

16. The cell image analysis device according to any one of claims 1 to 15, wherein the evaluation unit evaluates at least one selected from the group consisting of a state of the identical cell colony and a condition of the cell culture based on a predetermined evaluation criterion.

17. The cell image analysis device according to any one of claims 1 to 15, wherein the evaluation unit evaluates a degree of correlation between the image feature estimated by the feature estimation unit and at least one selected from the group consisting of a predetermined feature possessed by the same cell colony and the cell culture conditions.

18. an image acquiring step of acquiring time-series images acquired at two or more different time points in a time series of cell culture, each of which includes the same cell colony; a region extraction step of extracting a region of the same cell colony included in each of the time-series images; a feature extraction step of extracting image features from the region of the same cell colony in a plurality of images included in the time-series images; a feature amount estimation step of estimating an image feature amount for the same cell colony at a predetermined time point when the time-series images are not acquired, using the image feature amount extracted in the feature amount extraction step; an evaluation step of evaluating the same cell colony or the conditions of the cell culture using the image feature amount estimated in the feature amount estimation step; A cell image analysis method comprising:

19. an image acquiring step of acquiring time-series images acquired at two or more different time points in a time series of cell culture, each of which includes the same cell colony; a region extraction step of extracting a region of the same cell colony included in each of the time-series images; a feature extraction step of extracting, from the region extracted in the region extraction step, a first image feature corresponding to a feature that changes depending on a culture time of the cell culture and a second image feature corresponding to a feature used for evaluation; a feature estimation step of estimating the second image feature corresponding to a predetermined value for the first image feature; an evaluation step of evaluating at least one selected from the group consisting of the same cell colony and the cell culture conditions, using the second image feature amount estimated in the feature amount estimation step; A cell image analysis method comprising:

20. A cell image analyzing program for causing a computer to execute the cell image analyzing method according to claim 18 or 19.

21. A recording medium storing the cell image analysis program according to claim 20 in a computer-readable format.

Images