Apparatus, method, and program

Abstract

To provide an apparatus, method and program for estimating a proportion of detection targets in a sample containing cells or microorganisms as image-capturing targets.SOLUTION: An apparatus is provided comprising: an acquisition unit that acquires an image of a sample; an estimation unit that estimates a proportion of detection targets to image-capturing targets in a sample shown in an image acquired by the acquisition unit, based on a group of an image of a reference sample containing image-capturing targets and a proportion of cells or microorganisms as the detection targets to the image-capturing targets in the reference sample; an output unit that outputs the estimated proportion; and a determination unit that determines an extraction condition for extracting image-capturing targets in an image of the group according to the proportion of the group, among extraction conditions of image-capturing targets, which are settable by a parameter indicating a shape or texture of an image-capturing target in the image, based on the group of an image of the reference sample and the proportion of the detection targets to the image-capturing targets in the reference sample; where the estimation unit estimates the proportion regarding a sample shown by an input image, by using the extraction condition determined by the determination unit.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an apparatus, a method, and a program. [Background technology]

[0002] Patent Document 1 and other documents state that "Whether a cell is a stem cell or not is determined based on the results of comparing the optical path length inside and outside the cell nucleus region" (Claim 1 of Patent Document 1). [Prior art document] [Patent documents] [Patent Document 1] JP 2015-097498 A [Patent Document 2] JP 2012-231709 A Summary of the Invention

[0003] In a first aspect of the present invention, there is provided an apparatus comprising: an acquisition unit that acquires an image of a sample containing cells or microorganisms as imaging targets; an estimation unit that estimates the proportion of the detection target relative to the imaging target in the sample shown in the image acquired by the acquisition unit based on a set of an image of a reference sample containing the imaging target and the proportion of the cells or microorganisms as the detection target relative to the imaging target in the reference sample; and an output unit that outputs the proportion estimated by the estimation unit.

[0004] The above-mentioned device may further include a determination unit that determines, based on the set of an image of the reference sample and the proportion for the reference sample, extraction conditions for the imaged object that can be set by parameters indicating the shape or texture of the imaged object in the image, such that the imaged object corresponding to the proportion of the set is extracted in the image of the set, and the estimation unit may use the extraction conditions determined by the determination unit to estimate the proportion for the sample shown in the input image.

[0005] In the above-mentioned device, the determination unit may determine extraction conditions based on multiple sets of sets having different ratios of the detection target to the imaging target, such that the imaging target is extracted in the image according to the ratio for each of the multiple sets.

[0006] In any of the above devices having a determination unit, the parameters may include at least one of the area, diameter, circumference, circularity, compactness, ratio of major axis to minor axis, skeletal length, number of branches, or number of branches of the imaging object.

[0007] In the device of the first aspect described above, the estimation unit may perform estimation using a learning model generated by a learning process using learning data including a pair of an image of the reference sample and the proportion for the reference sample, and which outputs the proportion for the sample shown in the input image.

[0008] The device of the first aspect described above may further include a learning processing unit that generates a learning model that outputs the proportion for the sample shown in the input image using learning data that includes the pair of the image of the reference sample and the proportion for the reference sample, and the estimation unit may perform estimation using the learning model.

[0009] In any of the above devices, the acquisition section may acquire the image in which the detection target is imaged in a non-destructive and non-stained state.

[0010] Any of the above devices may further include an identification unit that identifies the ratio in the reference sample by at least one of destroying and staining cells or microorganisms contained in the reference sample.

[0011] Any of the above devices may further comprise a pre-processing unit that performs pre-processing on a sample using a cell sorter to generate the reference sample.

[0012] Any of the above devices may further include an imaging unit that captures an image of the sample.

[0013] In any of the above devices, the detection target may be a predetermined type of cell or microorganism.

[0014] In any of the above devices, the detection target may be a cell or a microorganism in a predetermined state.

[0015] In a second aspect of the present invention, there is provided a method comprising: acquiring an image of a sample containing a cell or a microorganism as an imaging target; estimating the ratio of the detection target to the imaging target in the sample shown in the acquired image based on a set of an image of a reference sample containing the imaging target and the ratio of the cells or microorganisms as the detection target to the imaging target in the reference sample; and outputting the estimated ratio.

[0016] In a third aspect of the present invention, there is provided a program that, when executed by a computer, causes the computer to function as an acquisition unit that acquires an image of a sample containing cells or microorganisms as the imaging target, an estimation unit that estimates the ratio of the detection target to the imaging target in the sample shown in the image acquired by the acquisition unit based on a pair of an image of a reference sample containing the imaging target and the ratio of cells or microorganisms as the detection target to the imaging target in the reference sample, and an output unit that outputs the ratio estimated by the estimation unit.

[0017] The above summary of the invention does not list all of the necessary features of the present invention, and subcombinations of these features may also constitute inventions. [Brief explanation of the drawings]

[0018] [Figure 1] 1 shows a device 1 according to a first embodiment. [Figure 2] The operation of the device 1 is shown. [Figure 3] The tip shape of the imaging target is shown. [Figure 4] The hole shape of the imaging target is shown. [Figure 5]The watershed shape of the imaged area is shown. [Figure 6] The valley shape of the imaging target is shown. [Figure 7] The edge shape of the object being imaged is shown. [Figure 8] 1 shows the saddle shape of the imaging object. [Figure 9] 1 shows a device 1A according to a second embodiment. [Figure 10] The operation of the device 1A is shown. [Figure 11] 12 illustrates an example computer 1200 in which aspects of the present invention may be embodied, in whole or in part. DETAILED DESCRIPTION OF THE INVENTION

[0019] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention according to the claims. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0020] (First embodiment) ((Device 1)) 1 shows an apparatus 1 according to this embodiment. As will be described in detail later, the apparatus 1 according to this embodiment estimates the proportion of the target substance contained in a sample for evaluation based on an image of a reference sample containing cells as the target substance and the proportion of the target substance contained in the reference sample. The apparatus 1 may include a preprocessing unit 10, a holding unit 11, an identifying unit 12, an imaging unit 13, an acquiring unit 14, a determining unit 15, an estimating unit 16, and an output unit 17.

[0021] Here, the sample may be liquid, and cells to be imaged may be dispersed therein. As an example, the sample may be a cell culture medium. The cells to be imaged may be cells that can be imaged by the imaging unit 13. The cells to be imaged may be cells of human origin, or may be cells of other organisms. As an example, the cells to be imaged may be iPS cells that have been subjected to a differentiation process, and may include at least one of undifferentiated cells and differentiated cells.

[0022] The reference sample may be a sample for determining conditions (also referred to as extraction conditions) used to estimate the proportion of the target substance contained in the evaluation sample. The target substance may be a cell whose content is to be detected among the cells to be imaged. The target substance may be a predetermined type of cell (e.g., hematopoietic stem cells (HPCs) differentiated from iPS cells) or a cell in a predetermined state (e.g., cells in a normal or abnormal state, cells in a specific cell cycle, cells in a state where a specific gene is expressed or not expressed, iPS cells in a state where a specific cell is differentiated or not differentiated). The proportion of the target substance contained in the sample may be the proportion of the target substance to be imaged in the sample.

[0023] The evaluation sample may be a sample in which the proportion of the target cell is estimated. The evaluation sample may be a sample different from the reference sample. The target cells contained in the evaluation sample may be the same type of cells as the target cells contained in the reference sample, so long as they are the same type of cells as the target cells contained in the reference sample, the source of acquisition, culture conditions, and cell state may be the same or different. Additionally or alternatively, the target cells contained in the evaluation sample may be the same type of cells as the target cells contained in the reference sample, so long as they are ... the source of acquisition, culture method, and cell type may be the same or different.

[0024] (((Pre-processing unit 10))) The pre-treatment section 10 performs pre-treatment on the sample. The pre-treatment section 10 may acquire the sample from an external source.

[0025] The pretreatment unit 10 may perform pretreatment on the acquired sample to generate a reference sample before determining the extraction conditions. The pretreatment unit 10 may perform processing using a cell sorter. The cell sorter may separate target cells from the sample by utilizing differences in cell morphology, surface antigens, fluorescent protein amounts, etc.

[0026] After determining the extraction conditions, the pretreatment unit 10 may supply the acquired sample to the holding unit 11 as a sample for evaluation without pretreatment. The pretreatment unit 10 may supply the sample to the holding unit 11 while maintaining it in a closed environment, for example, by supplying the sample to the holding unit 11 while it is contained in a container such as a culture bag. The sample maintained in a closed environment may be a sample in the middle of culturing. However, the pretreatment unit 10 may also generate a sample for evaluation by pretreatment and supply it to the holding unit 11.

[0027] (((holding part 11))) The holder 11 holds the sample (i.e., the reference sample or the sample for evaluation) supplied from the pretreatment unit 10 within an imaging area for the imaging unit 13. The holder 11 may be formed in the shape of a container and contain the sample therein, or may be formed in a cylindrical shape and allow the sample to flow inside. In these cases, the holder 11 may have a light-transmitting window member facing the imaging unit 13 so that the sample inside can be imaged by the imaging unit 13. When a sample contained in a container and maintained in a closed environment is supplied to the holder 11, the holder 11 may be formed in the shape of a platform and support the sample together with the container from below.

[0028] Before the extraction conditions are determined, the holding unit 11 may supply the sample (i.e., the reference sample) imaged by the imaging unit 13 to the identification unit 12. After the extraction conditions are determined, the holding unit 11 may discharge the imaged sample (i.e., the sample for evaluation) to the outside. When a sample that is contained in a container and maintained in a closed environment is supplied, the holding unit 11 may discharge the sample to the outside of the device 1 while it is maintained in the closed environment (i.e., while it is contained in the container).

[0029] (((Specific part 12))) The identifying unit 12 identifies the proportion of the target substance contained in the reference sample (also referred to as the "actual proportion" to distinguish it from the estimated proportion). The identifying unit 12 may identify the proportion of the target substance by at least one of destroying and staining cells contained in the reference sample. As an example, the identifying unit 12 may identify the proportion of the target substance using at least one of flow cytometry (FCM), quantitative PCR (qPCR), RNA sequencing, and immunostaining. The identifying unit 12 may identify the proportion of the target substance contained in the reference sample in the portion imaged by the imaging unit 13, or may identify the proportion of the target substance contained in the reference sample in the portion not imaged. The identifying unit 12 may supply the identified proportion of the target substance (i.e., the actual proportion) to the determining unit 15.

[0030] (((imaging unit 13))) The imaging unit 13 images the sample. The imaging unit 13 may image each of the reference sample and the evaluation sample. The imaging unit 13 may image the sample held in the holding unit 11. The imaging unit 13 may image at least the evaluation sample while the detection target is in a non-destructive and non-stained state. In the present embodiment, as an example, the imaging unit 13 may image each of the reference sample and the evaluation sample while the detection target is in a non-destructive and non-stained state. The detection target being non-destructive may mean that no destructive treatment has been performed on each part of the detection target. The detection target being unstained may mean that no staining treatment has been performed on each part of the detection target.

[0031] The imaging unit 13 may include a microscope for magnifying and capturing an image of an object. The microscope may be any type of microscope, such as a bright-field microscope. The imaging unit 13 may provide the captured image of the sample to the acquisition unit 14.

[0032] (((Acquisition part 14))) The acquisition unit 14 acquires an image of a sample containing cells or microorganisms as an imaging target. The acquisition unit 14 may acquire an image of a reference sample before determining the extraction conditions. The acquisition unit 14 may acquire an image of the reference sample in which the detection target is captured in a non-destructive and non-stained state as the image of the reference sample. The acquisition unit 14 may acquire the image of the reference sample from the imaging unit 13 and supply it to the determination unit 15.

[0033] After determining the extraction conditions, the acquisition unit 14 may acquire an image of the sample for evaluation. The acquisition unit 14 may acquire, as the image of the sample for evaluation, an image of the detection target captured in a non-destructive and non-stained state. The acquisition unit 14 may acquire the image of the sample for evaluation from the imaging unit 13 and supply it to the estimation unit 16.

[0034] (((Determining unit 15))) The determination unit 15 determines extraction conditions used to estimate the proportion of the detection target. The determination unit 15 may determine the extraction conditions based on a pair (also referred to as a correspondence set) of an image of a reference sample and a proportion of cells as the detection target relative to the imaged target in the reference sample (also referred to as a proportion of the detection target contained in the reference sample). The determination unit 15 may associate the image of the reference sample supplied from the acquisition unit 14 with the measured proportion supplied from the identification unit 12 to form a correspondence set.

[0035] The extraction conditions may be conditions for extracting at least a portion of the imaging target from the image, i.e., extraction conditions for imaging target cells, but do not necessarily need to be extraction conditions for detection target cells. The extraction conditions may be conditions that can be set using parameters (also referred to as condition parameters) that indicate the shape or texture of the imaging target in the image, and among such conditions, may be conditions that extract imaging targets corresponding to the measured proportion of the correspondence set in the image. In this way, when imaging targets that meet the extraction conditions are detected in the image, the proportion of the detected imaging targets may be close to the measured proportion of the correspondence set.

[0036] Here, the imaging target corresponding to the actual measurement ratio of the correspondence set may be an imaging target with a ratio within an error range centered on the actual measurement ratio. As an example, if the actual measurement ratio of the correspondence set is 50% and the width of the error range is 2Δ% (Δ is any natural number), the imaging target corresponding to the actual measurement ratio may be an imaging target with a ratio of (50-Δ)% to (50+Δ)%. However, the imaging target corresponding to the actual measurement ratio of the correspondence set may also be an imaging target with the actual measurement ratio itself.

[0037] The parameters for the conditions indicating the shape of the object in the image are the area of ​​the object (μm 2 ), diameter (μm), perimeter (μm), circularity, compactness, ratio of major axis to minor axis, skeletal length, number of branches, or number of branches. Compactness may be an index showing how packed the internal structure of a cell (such as the cytoplasm or nucleus) is, and is also referred to as Compactness. The ratio of major axis to minor axis may be the ratio of the length of the cell in the major axis direction to the length of the cell in the minor axis direction, and is also referred to as Anisometry. The number of branches may be the number of branch points, and the number of branches may be the number of branches at the branched ends. The determination unit 15 may detect cells in the image by image recognition and calculate the condition parameters from the detected cells.

[0038] The texture of the object in the image may be a marking or pattern of the object, or may be caused by the shape of the object in the imaging direction. The parameter for the condition indicating the texture of the object in the image may include the number of at least one preset texture. The number of textures may be the number of areas indicating the corresponding texture. Examples of preset textures include a texture indicating a peak shape (Peak), a texture indicating a hole shape (Hole), a texture indicating a watershed shape (Ridge), a texture indicating a valley shape (Valley), a texture indicating an edge shape (Edge), and a texture indicating a saddle shape (Saddle). Each texture will be described in detail below.

[0039] The condition that can be set by the parameter for the condition may be a numerical range of at least one, preferably two or more, parameter for the condition. For example, the condition that can be set by the parameter for the condition is a circularity of more than 0.4 and an area of ​​100 μm 2 It may be greater than. The determination unit 15 may supply the determined extraction conditions to the estimation unit 16.

[0040] (((estimation part 16))) The estimation unit 16 estimates the proportion of the detection target relative to the imaged target in the sample shown in the image acquired by the acquisition unit 14 (also referred to as the proportion of the detection target contained in the sample). In response to input of an image of an evaluation sample different from the reference sample, the estimation unit 16 may estimate the proportion of the detection target contained in the evaluation sample. The estimation unit 16 may perform the estimation based on a correspondence set between the image of the reference sample and the proportion of the detection target contained in the reference sample. In the present embodiment, as an example, the estimation unit 16 may estimate the proportion of the detection target for the sample shown in the input image using extraction conditions determined by the determination unit 15 based on the correspondence set.

[0041] The estimation unit 16 may estimate the proportion of imaging targets that meet the extraction conditions among imaging targets in the input image as the proportion of detection targets. The estimation unit 16 may calculate parameters for conditions from cells recognized in the image of the evaluation sample, in the same manner as the determination unit 15, and calculate the proportion of cells that meet the extraction conditions. However, the estimation unit 16 does not necessarily have to calculate parameters for conditions that are not included in the extraction conditions. The estimation unit 16 may supply the estimated ratio to the output unit 17.

[0042] (((output unit 17))) The output unit 17 outputs the ratio estimated by the estimation unit 16. The output unit 17 may output the ratio estimated by the estimation unit 16 in response to supplying the image of the sample for evaluation to the estimation unit 16.

[0043] According to the above-described device 1, the proportion of the target cell in the sample for evaluation shown in the acquired image is estimated and output based on a correspondence set between the image of the reference sample and the proportion of the target cell in the reference sample. Therefore, by capturing an image of the sample for evaluation, the proportion of the target cell in the sample can be obtained, and the proportion of the target cell can be obtained while maintaining the cells in the sample.

[0044] Furthermore, based on a correspondence set between an image of a reference sample and the proportion of the target substance contained in the reference sample, extraction conditions are determined from among extraction conditions for the target substance that can be set using parameters for the conditions of the target substance in the image, under which the target substance corresponding to the proportion of the corresponding set is extracted in the image of the corresponding set, and the proportion of the target substance for the sample shown in the input image is estimated using the determined extraction conditions. Therefore, by imaging the sample, the proportion of the target substance in the sample can be estimated with high accuracy.

[0045] Furthermore, the parameters for the conditions indicating the shape of the object in the image include at least one of the area, diameter, circumference, circularity, compactness, ratio of major axis to minor axis, skeleton length, number of branches, or number of branches of the object, and the parameters for the conditions indicating the texture of the object in the image include the number of at least one preset texture, so that the object that meets the extraction conditions can be reliably identified in the input image, and the proportion of the object to be detected can be reliably estimated.

[0046] Furthermore, since the image of the sample for evaluation is captured while the target is in a non-destructive and non-stained state, the proportion of cells in the target can be obtained without damaging the state of the target by destruction or staining.

[0047] Furthermore, the proportion of the target substance contained in the reference sample is determined by at least one of destroying and staining the cells contained in the reference sample, thereby enabling accurate determination of the proportion of the target substance contained in the reference sample.

[0048] Furthermore, because a reference sample is generated by pre-processing the sample using a cell sorter, the proportion of the target substance can be determined with a low content of imaging targets other than the target substance in the reference sample, preventing imaging targets other than the target substance from being mistakenly identified as the target substance, and enabling accurate determination of the proportion of the target substance contained in the reference sample.

[0049] Furthermore, since the sample is imaged within the device 1, the process from imaging the sample for evaluation to estimating the proportion of the target to be detected can be carried out inline.

[0050] Furthermore, since the detection target is a predetermined type of cell, it is possible to estimate the proportion of the cell type. Furthermore, since the detection target is a cell in a predetermined state, it is possible to estimate the proportion of the cells in that state.

[0051] ((action)) 2 shows the operation of the device 1. The device 1 performs the processes of steps S11 to S41 to estimate the proportion of the target substance contained in the sample for evaluation.

[0052] In step S11, the pre-processing unit 10 acquires a reference sample. The pre-processing unit 10 may generate the reference sample by pre-processing a sample acquired from outside. As an example, the pre-processing unit 10 may generate the reference sample by sorting the cells to be detected using a cell sorter.

[0053] In step S13, the imaging unit 13 images the reference sample. The imaging unit 13 may image the reference sample that has been preprocessed and held in the holding unit 11. The captured image may be supplied to the determining unit 15 via the acquiring unit 14.

[0054] In step S15, the determination unit 15 detects, from the image, parameters for the conditions of the cells to be imaged. The parameters for the conditions may be average values ​​of multiple cells to be imaged. The determination unit 15 may detect multiple types of parameters for the conditions (for example, area, diameter, perimeter, circularity, compactness, ratio of major axis to minor axis, skeletal length, number of branches, and number of branches).

[0055] In step S17, the determination unit 15 determines whether the number of data items for the condition parameters has reached the required number of data items. The number of data items for the condition parameters may be the number of condition parameters detected in step S15 for the same reference sample. For example, if two condition parameters (for example, area and diameter) are detected in step S15, the number of data items for the condition parameters may be two. The required number of data items may be the number of types of condition parameters (for example, area, diameter, circumference, circularity, compactness, ratio of major axis to minor axis, skeletal length, number of branches, and number of branches).

[0056] If it is determined that the number of data items for the condition parameters has not reached the required number of data items (step S17; No), the process may proceed to step S13. As a result, an image of the reference sample may be captured again to detect the condition parameters. If it is determined that the number of data items for the condition parameters has reached the required number of data items (step S17; Yes), the process may proceed to step S19.

[0057] In step S19, the identification unit 12 identifies the proportion of the target substance contained in the reference sample acquired in step S11 (i.e., the measured proportion). The identification unit 12 may identify the proportion of the target substance by at least one of destroying and staining cells contained in the reference sample. The identified proportion may be supplied to the determination unit 15. In this embodiment, as an example, the image supplied to the determination unit 15 in step S13 and the proportion supplied to the determination unit 15 in step S19 may be assigned identification information for each reference sample, and the determination unit 15 may associate the image and the measured proportion for the same reference sample based on the identification information to form a correspondence set.

[0058] In step S21, the determining unit 15 determines whether the number of data items of the actual proportions identified in step S19 has reached the required number of data items. The number of data items of proportions may be the number of actual proportions identified in step S19 for each reference sample acquired in step S11. The required number of data items may be any number set in advance. The determining unit 15 may determine whether the number of data items of proportions falling within each predetermined numerical range (for example, 80 to 90%, 90 to 100%, etc.) has reached the required number of data items.

[0059] If it is determined that the number of proportion data has reached the required number of data (step S21; Yes), the process may proceed to step S23. If it is determined that the number of proportion data has not reached the required number of data (step S21; No), the process may proceed to step S11. As a result, another reference sample is acquired, and the processes of steps S13 to S19 are performed. As long as the cells to be detected contained in the reference sample in each step S11 are the same type of cells, the source of acquisition, culture conditions, cell state, and liquid composition of the cells to be imaged may be the same or different. Additionally or alternatively, as long as the cells to be detected contained in the reference sample are the same state of cells, the source of acquisition, culture conditions, cell type, and liquid composition of the cells to be imaged may be the same or different.

[0060] In step S23, the determination unit 15 provisionally determines extraction conditions. The determination unit 15 may provisionally determine, as extraction conditions, conditions that can be set using parameters for the conditions of the imaging target in the image.

[0061] In step S25, the determination unit 15 determines whether the estimation accuracy based on the provisionally determined extraction conditions falls within an allowable range. The determination unit 15 may calculate, for at least one correspondence set, the proportion of imaging targets extracted from the images of the correspondence set based on the provisionally determined extraction conditions, i.e., the estimated proportion of detection targets, and calculate the estimation accuracy relative to the actual proportion of the correspondence set (i.e., the proportion of actual detection targets).

[0062] The estimation accuracy may be the rate of agreement between the estimated proportion and the actual proportion of the detection target. In this embodiment, as an example, the estimation accuracy is a value calculated by (actual proportion / estimated proportion)×100, and the smaller the error, the closer it is to 100%. The allowable range is a range set according to the error range of the estimated proportion, and when the width of the error range is 2Δ%, it may be {actual proportion / (estimated proportion-Δ)}×100 to {actual proportion / (estimated proportion+Δ)}×100.

[0063] The determining unit 15 may calculate the estimation accuracy for each of a plurality of correspondence sets having different actual measurement proportions, and determine whether all of the calculated estimation accuracies fall within an allowable range. For example, the determining unit 15 may determine whether the estimation accuracy calculated for a first correspondence set including an image of a reference sample having an actual measurement proportion of 50% falls within the range {50 / (estimated proportion-Δ)}×100 to {50 / (estimated proportion+Δ)}×100, and whether the estimation accuracy calculated for a second correspondence set including an image of a reference sample having an actual measurement proportion of 60% falls within the range {60 / (estimated proportion-Δ)}×100 to {60 / (estimated proportion+Δ)}×100.

[0064] If it is determined that the estimation accuracy is not within the allowable range (step S25; No), the process may proceed to step S23. As a result, other extraction conditions are tentatively determined. If it is determined that the estimation accuracy is within the allowable range (step S25; Yes), the process may proceed to step S27.

[0065] In step S27, the determination unit 15 determines the extraction conditions provisionally determined in the most recent step S25 as the extraction conditions for the target to be used. As a result, among the conditions that can be set by the parameters for the conditions of the target to be captured in the image, the conditions under which the target to be captured according to the actual measurement proportion of the correspondence set is extracted in the image of the correspondence set are determined as the extraction conditions. In addition, by determining extraction conditions for which the estimation accuracy for multiple correspondence sets is within an allowable range, extraction conditions are determined under which the target to be captured according to the actual measurement proportion (i.e., the proportion of the actual target to be detected) in each of the multiple correspondence sets is extracted in the image of the correspondence set, based on multiple correspondence sets having mutually different actual measurement proportions of the target to be detected.

[0066] The determined extraction conditions may be supplied to the estimation unit 16. Through the above steps S11 to S27, the estimation unit 16 can estimate the proportion of the detection target contained in a sample when an image of the sample for evaluation is input.

[0067] In step S31, the pre-processing unit 10 acquires a sample for evaluation. The pre-processing unit 10 may acquire the sample for evaluation itself from an external source.

[0068] In step S33, the imaging unit 13 images the sample for evaluation. The imaging unit 13 may image the sample for evaluation held in the holding unit 11. The captured image may be supplied to the acquisition unit 14. As a result, the image of the sample for evaluation is acquired by the acquisition unit 14 and supplied from the acquisition unit 14 to the estimation unit 16.

[0069] In step S35, the estimation unit 16 detects, from the image, parameters for conditions included in the extraction conditions, among parameters for conditions of the cell as the imaging target.

[0070] In step S37, the estimation unit 16 determines whether the number of data items for the condition parameters has reached the required number of data items. The number of data items for the condition parameters may be the number of condition parameters detected in step S35 for the same evaluation sample. The required number of data items may be the number of condition parameters included in the extraction conditions.

[0071] If it is determined that the number of data items for the condition parameters has not reached the required number of data items (step S37; No), the process may proceed to step S33. As a result, an image of the sample for evaluation may be captured again to detect the condition parameters. If it is determined that the number of data items for the condition parameters has reached the required number of data items (step S37; Yes), the process may proceed to step S39.

[0072] In step S39, the estimation unit 16 estimates the proportion of the detection target contained in the evaluation sample. As a result, the proportion of the detection target contained in the evaluation sample shown in the acquired image is estimated by the estimation unit 16. The estimation unit 16 may estimate, as the proportion of the detection target, the proportion of the imaged targets that meet the extraction conditions among the imaged targets in the image of the evaluation sample.

[0073] Then, in step S41, the output unit 17 outputs the estimated ratio of the detection target. The output unit 17 may display the estimated ratio or may transmit it to an external device.

[0074] According to the above operation, based on a plurality of correspondence sets having mutually different actual measurement ratios (i.e., the ratio of the actual detection target), extraction conditions are determined for extracting the imaging target in the image according to the actual measurement ratio in each of the plurality of correspondence sets. Therefore, regardless of the ratio of the detection target contained in an image of a sample input, the ratio of the detection target in the sample can be estimated with high accuracy.

[0075] ((Example)) (((Preparation of standard sample))) Adherent culture of iPS cells (strain 201B7) was performed for 7 days using a culture microplate for adherent cells (manufactured by Iwaki Corporation). The cultured iPS cells were detached from the microplate using the enzyme "TryPLE" (registered trademark) (manufactured by Thermo Fisher Scientific), suspended in StemFit medium (manufactured by Ajinomoto Co., Inc.), and subcultured to prepare single iPS cells. Y-27632 (manufactured by Fujifilm Wako Pure Chemical Industries, Ltd.) was added to the medium to a final concentration of 10 μM.

[0076] The prepared single cells were seeded into a well-equipped culture bag (Toyo Seikan Group Holdings Co., Ltd.) that had been treated for low cell adhesion and cultured for one day to aggregate the cells and produce spheres. The iPS cell spheres were differentiated into hematopoietic stem cells (HPCs) using the embryoid body (EB) method, and the resulting cells were harvested. The harvested cells were seeded into a 24-well plate (Iwaki Co., Ltd.) at 4,000 cells per well to serve as a reference sample.

[0077] (((Detecting parameters for conditions))) Cell images of each well were taken using an "Image Cytometer CQ-1" (Yokogawa Electric Corporation). The obtained images were analyzed using the image analysis software "CellPathfinder" (Yokogawa Electric Corporation), which eliminated non-cellular impurities and allowed cells to be image-recognized. Shape analysis was performed on the recognized cells, and the area (μm 2 ), diameter (μm), circumference (μm), circularity, compactness, long axis / short axis ratio, skeleton length, number of branches, and number of branches were detected as condition parameters.

[0078] (((Specifying the percentage))) Flow cytometry (FCM) analysis was performed on a portion of the cells from the reference sample to determine the percentage of hematopoietic stem cells (HPCs) marked by the CD34 marker (i.e., the actual percentage of detected cells), which was 66.3% of all cells.

[0079] (((Determining extraction conditions))) Among the conditions that can be set by the condition parameters, the circularity and area (μm 2 ) was used to determine the condition of circularity > 0.465 and area (μm 2 )>100.

[0080] (((Estimated proportion of samples for evaluation))) iPS cells were differentiated into hematopoietic stem cells (HPCs) in the same manner as the reference sample, and single cells were collected. Cell images of the collected cells in each well were taken using an "Image Cytometer CQ-1" (Yokogawa Electric Corporation), and the image analysis software "CellPathfinder" (Yokogawa Electric Corporation) was used to eliminate non-cellular impurities and recognize the cells in the image. Shape analysis was performed on the recognized cells to detect the circularity and area of ​​each cell. Cells in the image with a circularity of >0.465 and an area (μm 2 The percentage of cells that met the extraction condition of >100, i.e., the estimated percentage of hematopoietic stem cells, was 73.66%.

[0081] (((Verification of estimation accuracy))) Flow cytometry (FCM) analysis of a portion of the evaluation sample identified the percentage of hematopoietic stem cells (HPCs) marked by the CD34 marker, which was 65.1% of the total cells, resulting in a concordance rate of 88.4% (=65.1 / 73.66 × 100) between the estimated and actual percentages.

[0082] As in the above examples, the extraction conditions were set to compactness > 1.08 and area (μm 2 ) > 100, the agreement rate between the estimated proportion of the evaluation sample and the measured proportion was 94.8%. 2 ) > 100, the agreement rate between the estimated proportion in the evaluation sample and the actual measured proportion was 92.8%.

[0083] ((Texture of the object being imaged)) Fig. 3 shows the shape of the tip of the imaging target. In Fig. 3 to Fig. 8, the outer shape of the imaging target is shown as a mesh pattern, the imaging direction is the up-down direction, and the plane perpendicular to the imaging direction is shown as a white rectangle.

[0084] The tip shape may be a cone-like shape protruding in the imaging direction. The texture of the tip shape may be generated by the tip shape, and the brightness may change with distance from the center point.

[0085] 4 shows a hole shape of an imaging target. The hole shape may be a cone-shaped depression in the imaging direction. The texture of the hole shape may be generated by the hole shape, and the brightness may change with distance from the center point.

[0086] 5 shows the watershed shape of the imaged object. The watershed shape may be a ridge-like shape that protrudes in the imaging direction. The texture of the watershed shape may be generated by the watershed shape, and the brightness may change with distance from the center line (i.e., ridge line, crest line).

[0087] 6 shows a valley shape of an imaged object. The valley shape may be a shape that is recessed in the imaging direction. The texture of the valley shape may be generated by the valley shape, and the brightness may change with distance from the center line (i.e., the valley line).

[0088] 7 shows the edge shape of an imaged object. The edge shape may be the shape of the edge of the imaged object. The texture of the edge shape may be generated by the edge shape and may be a texture whose brightness changes toward the edge portion of the imaged object.

[0089] FIG. 8 shows a saddle-shaped object. A saddle-shaped object may be a shape that is concave in the imaging direction and has curvature in two directions, like a horse's saddle. In the figure, the curvature in the depth direction is different from the curvature in the left-right direction. The saddle-shaped texture may be generated by the saddle shape, and may be a texture in which the brightness changes with distance from the center of the object, and the brightness change is anisotropic.

[0090] (Second embodiment) ((Device 1A)) 9 shows a device 1A according to a second embodiment. The device 1A includes a learning processing unit 18A and an estimation unit 16A. In the device 1A according to this embodiment, components that are substantially the same as those in the device 1 shown in FIG. 1 are designated by the same reference numerals, and descriptions thereof will be omitted.

[0091] The learning processing unit 18A generates the learning model 160A through a learning process. The learning model 160A uses learning data including a correspondence set between an image of a reference sample and the proportion of the target substance contained in the reference sample, and outputs the proportion of the target substance contained in the sample shown in the input image. The learning processing unit 18A may generate the learning model 160A through machine learning such as deep learning, for example. In the machine learning, learning data including images of various cells or microorganisms and information indicating the type of the captured cells or microorganisms may be used. The learning processing unit 18A may supply the generated learning model 160A to the estimation unit 16.

[0092] The estimation unit 16A estimates the proportion of the target substance contained in the sample shown in the image acquired by the acquisition unit 14. In response to input of an image of an evaluation sample different from the reference sample, the estimation unit 16A may estimate the proportion of the target substance contained in the evaluation sample.

[0093] The estimation unit 16A according to the present embodiment may perform estimation using the learning model 160A. Note that, as an example, the present embodiment will be described assuming that the estimation unit 16A has the learning model 160A built therein, but the learning model 160A may be provided outside the estimation unit 16A. The estimation unit 16A may use the proportion of images output from the learning model 160A in response to supplying the images supplied from the acquisition unit 14 to the learning model 160A as the proportion of the detection target.

[0094] According to the above-described device 1A, a learning model 160A is generated by a learning process using learning data including a correspondence set between an image of a reference sample and the proportion of the target substance contained in the reference sample, and the learning model 160A outputs the proportion of the target substance contained in the sample in response to an input image of the sample. Therefore, by capturing an image of the sample, the proportion of the target substance in the sample can be estimated with high accuracy.

[0095] ((action)) 10 shows the operation of the device 1 A. The device 1 A estimates the proportion of the target substance contained in the sample for evaluation by performing the processes of steps S51 to S81.

[0096] In step S51, the pre-processing unit 10 acquires a reference sample. The pre-processing unit 10 may generate the reference sample in the same manner as in step S11 described above.

[0097] In step S53, the imaging unit 13 images the reference sample. The imaging unit 13 may image the reference sample in the same manner as in step S13 described above. The captured image may be supplied to the learning processing unit 18A via the acquisition unit 14.

[0098] In step S59, the identification unit 12 identifies the proportion of the detection target contained in the reference sample acquired in step S51 (i.e., the actual measured proportion). The identification unit 12 may identify the proportion of the detection target in the same manner as in step S19 described above. The identified proportion may be supplied to the learning processing unit 18A. Note that in this embodiment, as an example, the image supplied to the learning processing unit 18A in step S53 and the proportion supplied to the learning processing unit 18A in step S59 may be assigned identification information for each reference sample, and the learning processing unit 18A may associate the image and the actual measured proportion for the same reference sample based on the identification information to form a correspondence set.

[0099] In step S61, the learning processing unit 18A determines whether the number of data items with the actual measurement ratios identified in step S59 has reached the required number of data items. The learning processing unit 18A may make this determination in the same manner as the determination unit 15 in step S21 described above.

[0100] If it is determined that the number of proportion data has reached the required number of data (step S61; Yes), the process may proceed to step S63. If it is determined that the number of proportion data has not reached the required number of data (step S61; No), the process may proceed to step S51. As a result, another reference sample is acquired, and the processes of steps S53 to S59 are performed. As long as the cells to be detected contained in the reference sample in each step S51 are the same type of cells, the source of acquisition, culture conditions, cell state, and liquid composition of the cells to be imaged may be the same or different. Additionally or alternatively, as long as the cells to be detected contained in the reference sample are the same state of cells, the source of acquisition, culture conditions, cell type, and liquid composition of the cells to be imaged may be the same or different.

[0101] In step S63, the learning processing unit 18A performs a learning process on the learning model 160A. The learning processing unit 18A may perform the learning process on the learning model 160A using learning data including a correspondence set between an image of a reference sample and the proportion of the target substance contained in the reference sample, so as to output the proportion of the target substance contained in the sample shown in the input image. The learning processing unit 18A may perform the learning process using all of the correspondence sets generated in steps S51 to S59, or may perform the learning process using some of the correspondence sets.

[0102] In step S65, the learning processing unit 18A determines whether the estimation accuracy by the learning model 160A falls within an acceptable range. The learning processing unit 18A may supply images of at least one correspondence set (for example, a correspondence set not used in the learning process) to the learning model 160A to output an estimated proportion of the detection target, and calculate the estimation accuracy for the actually measured proportion of the correspondence set (i.e., the proportion of the actual detection target). The learning processing unit 18A may calculate the estimation accuracy and determine whether the estimation accuracy falls within an acceptable range, similar to the determination unit 15 in step S25 described above.

[0103] If it is determined that the estimation accuracy is not within the allowable range (step S65; No), the process may proceed to step S51. As a result, a reference sample is acquired again and the learning process is performed. If it is determined that the estimation accuracy is within the allowable range (step S65; Yes), the learning model 160A may be supplied to the estimation unit 16, and the process may proceed to step S71.

[0104] In step S71, the pre-processing unit 10 acquires a sample for evaluation. The pre-processing unit 10 may acquire the sample for evaluation itself in the same manner as in step S31 described above.

[0105] In step S73, the imaging unit 13 images the sample for evaluation. The imaging unit 13 may image the sample for evaluation in the same manner as in step S33 described above. The captured image may be supplied to the acquisition unit 14. As a result, the image of the sample for evaluation is acquired by the acquisition unit 14 and supplied from the acquisition unit 14 to the estimation unit 16A.

[0106] In step S79, the estimation unit 16A estimates the proportion of the detection target contained in the evaluation sample. As a result, the estimation unit 16A estimates the proportion of the detection target contained in the evaluation sample shown in the acquired image. The estimation unit 16A may estimate, as the proportion of the detection target, the proportion output from the learning model 160A in response to supplying the image of the evaluation sample to the learning model 160A.

[0107] Then, in step S81, the output unit 17 outputs the estimated ratio of the detection target. The output unit 17 may output the ratio in the same manner as in step S41 described above.

[0108] (Variation) In the above embodiment, the device 1 has been described as including the pre-processing unit 10, the holding unit 11, the imaging unit 13, the identification unit 12, and the determination unit 15, but any of these may be omitted. If the device 1 does not include the pre-processing unit 10, the imaging unit 13 may image a reference sample that has not been processed by a cell sorter. If the device 1 does not include the holding unit 11 and the imaging unit 13, the acquisition unit 14 may acquire an image of the sample from an external imaging device. If the device 1 does not include the identification unit 12, the determination unit 15 may acquire an externally identified proportion of the detection target (for example, a proportion manually determined by a user). If the device 1 does not include the determination unit 15, the estimation unit 16 may estimate the proportion using externally determined extraction conditions (for example, extraction conditions determined by a user through trial and error).

[0109] Similarly, although the device 1A has been described as including the preprocessing unit 10, the storage unit 11, the identification unit 12, the imaging unit 13, and the learning processing unit 18A, any of these may be omitted. If the device 1A does not include the identification unit 12, the learning processing unit 18A may acquire the proportion of the detection target identified externally. If the device 1A does not include the learning processing unit 18A, the estimation unit 16 may estimate the proportion using a learning model 160A generated externally.

[0110] In addition, although the devices 1 and 1A have been described as capturing an image of the reference sample and then determining the ratio of the detection target, the image capturing and the ratio determination may be performed in reverse order, or may be performed in parallel. In these cases, an image may be acquired using a portion of the reference sample, and the ratio determination may be performed using another portion.

[0111] In addition, although the determination unit 15 of the device 1 has been described as determining the provisionally determined extraction conditions as the extraction conditions to be used if the estimation accuracy of the provisionally determined extraction conditions falls within an acceptable range, the extraction conditions may be determined by other methods. For example, after provisionally determining a reference number of extraction conditions, the determination unit 15 may determine the extraction condition with the estimation accuracy closest to 100% as the extraction condition to be used.

[0112] Furthermore, although the imaging target has been described as a cell, it may also be a microorganism. In this case, the detection target may be a microorganism whose content ratio is to be detected among the microorganisms that are the imaging target. The microorganism may be a minute organism that cannot be observed with the naked eye, such as eukaryotic algae, protozoa, fungi, prokaryotic bacteria, cyanobacteria, or viruses. The bacteria may be chlamydia, rickettsia, or actinomycetes. The detection target microorganism may be a microorganism of a predetermined type (for example, microalgae) or a microorganism in a predetermined state (for example, a microorganism in a normal or abnormal state, or a microorganism in a specific cell cycle).

[0113] Various embodiments of the present invention may be described with reference to flowcharts and block diagrams, where the blocks may represent (1) stages of a process in which operations are performed or (2) sections of an apparatus responsible for performing the operations. Particular stages and sections may be implemented by dedicated circuitry, programmable circuitry provided with computer-readable instructions stored on a computer-readable medium, and / or a processor provided with computer-readable instructions stored on a computer-readable medium. Dedicated circuitry may include digital and / or analog hardware circuitry, and may include integrated circuits (ICs) and / or discrete circuits. Programmable circuitry may include reconfigurable hardware circuitry, including logical AND, OR, XOR, NAND, NOR, and other logical operations, flip-flops, registers, memory elements such as field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), and the like.

[0114] A computer-readable medium may include any tangible device capable of storing instructions that are executed by an appropriate device, such that the computer-readable medium having instructions stored thereon comprises an article of manufacture containing instructions that can be executed to create means for performing the operations specified in the flowcharts or block diagrams. Examples of computer-readable media may include electronic, magnetic, optical, electromagnetic, and semiconductor storage media. More specific examples of computer-readable media may include floppy disks, diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital versatile disc (DVD), Blu-ray disc, memory stick, integrated circuit card, and the like.

[0115] The computer readable instructions may include either assembler instructions, Instruction Set Architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, JAVA®, C++, etc., and conventional procedural programming languages ​​such as the “C” programming language or similar programming languages.

[0116] The computer-readable instructions may be provided to a processor or programmable circuit of a programmable data processing device, such as a computer, locally or over a wide area network (WAN) such as a local area network (LAN) or the Internet, and the computer-readable instructions may be executed to create means for performing the operations specified in the flowcharts or block diagrams. Here, the computer may be a personal computer (PC), a tablet computer, a smartphone, a workstation, a server computer, a general-purpose computer, a special-purpose computer, or the like, or may be a computer system in which multiple computers are connected. Such a computer system in which multiple computers are connected is also called a distributed computing system, and is a broad definition of computer. In a distributed computing system, the multiple computers collectively execute a program by each executing a portion of the program and passing data between the computers as needed during program execution.

[0117] Examples of processors include a computer processor, a central processing unit (CPU), a processing unit, a microprocessor, a digital signal processor, a controller, a microcontroller, etc. A computer may have one or more processors. In a multiprocessor system with multiple processors, each processor executes a portion of a program and passes data between processors as needed during program execution, allowing the multiple processors to collectively execute a program. For example, in multitasking, each of the multiple processors may execute a portion of each task in small chunks by switching tasks at time slice intervals. In this case, which portion of a program each processor executes changes dynamically. Which portion of a program each of the multiple processors executes may also be statically determined by multiprocessor-aware programming.

[0118] 11 illustrates an example of a computer 1200 in which aspects of the present invention may be embodied, in whole or in part. Programs installed on the computer 1200 may cause the computer 1200 to function as or perform operations associated with an apparatus or one or more sections of the apparatus according to embodiments of the present invention, and / or to perform a process or steps of a process according to embodiments of the present invention. Such programs may be executed by the CPU 1212 to cause the computer 1200 to perform specific operations associated with some or all of the blocks of the flowcharts and block diagrams described herein.

[0119] A computer 1200 according to this embodiment includes a CPU 1212, a RAM 1214, a graphics controller 1216, and a display device 1218, which are interconnected by a host controller 1210. The computer 1200 also includes input / output units such as a communication interface 1222, a storage device 1224 such as a hard disk drive, a DVD-ROM drive 1226, and an IC card drive, which are connected to the host controller 1210 via an input / output controller 1220. The computer also includes legacy input / output units such as a ROM 1230 and a keyboard 1242, which are connected to the input / output controller 1220 via an input / output chip 1240.

[0120] The CPU 1212 operates according to programs stored in the ROM 1230 and the RAM 1214, thereby controlling each unit. The graphics controller 1216 acquires image data generated by the CPU 1212 into a frame buffer or the like provided in the RAM 1214 or into the graphics controller 1216 itself, and causes the image data to be displayed on the display device 1218.

[0121] The communication interface 1222 communicates with other electronic devices via a network. The storage device 1224 stores programs and data used by the CPU 1212 in the computer 1200. The DVD-ROM drive 1226 reads programs or data from a DVD-ROM 1227 and provides the programs or data to the storage device 1224 via the RAM 1214. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0122] The ROM 1230 stores therein a boot program or the like that is executed by the computer 1200 upon activation, and / or programs that depend on the hardware of the computer 1200. The input / output chip 1240 may also connect various input / output units to the input / output controller 1220 via a parallel port, a serial port, a keyboard port, a mouse port, etc.

[0123] The programs are provided by a computer-readable medium such as a DVD-ROM 1227 or an IC card. The programs are read from the computer-readable medium, installed in the storage device 1224, RAM 1214, or ROM 1230, which are also examples of computer-readable media, and executed by the CPU 1212. Information processing described in these programs is read by the computer 1200, and causes cooperation between the programs and the various types of hardware resources described above. An apparatus or a method may be configured by implementing information manipulation or processing in accordance with the use of the computer 1200.

[0124] For example, when communication is performed between the computer 1200 and an external device, the CPU 1212 may execute a communication program loaded into the RAM 1214 and instruct the communication interface 1222 to perform communication processing based on the processing described in the communication program. Under the control of the CPU 1212, the communication interface 1222 reads transmission data stored in a transmission buffer processing area provided in the RAM 1214, the storage device 1224, the DVD-ROM 1227, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes reception data received from the network to a reception buffer processing area or the like provided on the recording medium.

[0125] The CPU 1212 may also cause all or a necessary portion of a file or database stored in an external recording medium such as the storage device 1224, the DVD-ROM drive 1226 (DVD-ROM 1227), an IC card, etc. to be read into the RAM 1214, and perform various types of processing on the data in the RAM 1214. The CPU 1212 then writes back the processed data to the external recording medium.

[0126] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and may undergo information processing. CPU 1212 may perform various types of processing on data read from RAM 1214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search / replacement, etc., as described throughout this disclosure and specified by the instruction sequences of the programs, and write the results back to RAM 1214. CPU 1212 may also search for information in a file, database, etc. on the recording medium. For example, if multiple entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored on the recording medium, CPU 1212 may search for an entry that matches a condition specified by the attribute value of the first attribute from among the multiple entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0127] The above-described programs or software modules may be stored in a computer-readable medium on or near the computer 1200. A recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can also be used as a computer-readable medium, thereby providing the programs to the computer 1200 via the network.

[0128] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention.

[0129] It should be noted that the execution order of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before," "prior to," etc., and that the processes can be performed in any order unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using "first," "next," etc. for convenience, this does not mean that the processes must be performed in this order. [Explanation of symbols]

[0130] 1,1A device 10 Pretreatment section 11 Holding part 12 Specific section 13 Imaging unit 14 Acquisition Department 15 Decision Section 16,16A Estimation part 17 Output section 18A Learning processing unit 160A Learning Model 1200 Computer 1210 host controller 1212 CPU 1214 RAM 1216 Graphics Controller 1218 Display Devices 1220 Input / Output Controller 1222 communication interface 1224 Storage device 1226 DVD-ROM drive 1227 DVD-ROM 1230 ROM 1240 Input / Output Chip 1242 keyboard

Claims

1. an acquisition unit for acquiring an image of a sample including cells or microorganisms as an imaging target; an estimation unit that estimates the ratio of the detection target to the imaging target in the sample shown in the image acquired by the acquisition unit, based on a pair of an image of a reference sample including the imaging target and a ratio of cells or microorganisms as the detection target to the imaging target in the reference sample; and an output unit that outputs the ratio estimated by the estimation unit.

2. a determination unit that determines, based on the set of the image of the reference sample and the ratio for the reference sample, an extraction condition for extracting an imaging target in the image according to the ratio of the set from among extraction conditions for the imaging target that can be set by a parameter indicating a shape or texture of the imaging target in the image, The apparatus according to claim 1 , wherein the estimation unit estimates the proportion for the sample shown in the input image using the extraction condition determined by the determination unit.

3. The device according to claim 2 , wherein the determination unit determines extraction conditions for extracting the imaging target from the image according to the ratio of the detection target to the imaging target in each of the plurality of sets, the ratios being different from one another.

4. the parameters indicating the shape of the object in the image include at least one of the area, diameter, perimeter, circularity, compactness, ratio of major axis to minor axis, skeletal length, number of branches, or number of branches of the object; The apparatus according to claim 2 , wherein the parameter indicative of the texture of the object in the image includes a number of at least one preset texture.

5. The device according to claim 1, wherein the estimation unit performs estimation using a learning model generated by a learning process using learning data including a pair of an image of the reference sample and the proportion for the reference sample, the learning model outputting the proportion for the sample shown in an input image.

6. a learning processing unit that generates a learning model that outputs the proportion for the sample shown in the input image using learning data that includes the pair of the image of the reference sample and the proportion for the reference sample, The device according to claim 1 , wherein the estimation unit performs estimation using the learning model.

7. The apparatus according to claim 1 , wherein the acquisition unit acquires the image in a non-destructive and non-stained state of the detection target.

8. The device according to claim 1 , further comprising an identification unit that identifies the ratio for the reference sample by at least one of destroying and staining cells or microorganisms contained in the reference sample.

9. The apparatus of claim 1 , further comprising a pre-processing section that performs pre-processing of a sample with a cell sorter to generate the reference sample.

10. The apparatus of claim 1 , further comprising an imaging unit for imaging the sample.

11. The device according to claim 1 , wherein the detection target is a predetermined type of cell or microorganism.

12. The device according to claim 1 , wherein the detection target is a cell or a microorganism in a predetermined state.

13. acquiring an image of a sample including cells or microorganisms as an imaging target; estimating the ratio of the detection target to the imaging target in the sample shown in the acquired image based on a set of an image of a reference sample containing the imaging target and a ratio of cells or microorganisms as the detection target to the imaging target in the reference sample; outputting the estimated proportion; A method for providing

14. When executed by a computer, the computer an acquisition unit for acquiring an image of a sample including cells or microorganisms as an imaging target; an estimation unit that estimates the ratio of the detection target to the imaging target in the sample shown in the image acquired by the acquisition unit, based on a pair of an image of a reference sample including the imaging target and a ratio of cells or microorganisms as the detection target to the imaging target in the reference sample; an output unit that outputs the ratio estimated by the estimation unit A program that functions as a

Images