Method and device for estimating therapeutic effect of stem cell transplantation

Abstract

This method for estimating the therapeutic effect of stem cell transplantation comprises: a step (S11) for acquiring a plurality of cell images in which a stem cell group in culture is imaged at a plurality of predetermined time points within a cell culturing period for stem cell transplantation for regenerative medicine purposes; a step (S12) for identifying stem cell regions respectively in the plurality of cell images; a step (S13) for acquiring, on the basis of the identified stem cell regions, a plurality of types of index values associated with the stem cell group with respect to each of the plurality of cell images; and a step (S14) for estimating, before the transplantation of the stem cell group obtained by the cell culturing, the therapeutic effect of the transplantation of the stem cell group on the basis of the acquired plurality of types of index values. The plurality of types of index values acquired in step (S13) include a plurality of types of index values associated with the morphologies of the stem cells.

Description

Method and apparatus for estimating the effect of stem cell transplantation

[0001] The present invention relates to a technique for estimating the effect of stem cell transplantation therapy.

[0002] In regenerative medicine, stem cell therapy is performed in which a group of stem cells cultured and proliferated in vitro from stem cells established from tissues extracted from the body are transplanted back into the body in order to improve or repair damaged tissues and organs of patients. Currently, various techniques for manufacturing and processing stem cells used in such regenerative medicine have been developed (see Patent Document 1 below). In addition, as in Patent Document 2 below, a method for evaluating the quality of stem cells such as differentiation ability has also been proposed.

[0003] Japanese Patent Application Laid-Open No. 2017-221186, Japanese Patent Application Laid-Open No. 2018-117582

[0004] One aspect of the present invention is a method for estimating the effect of stem cell transplantation therapy. The estimation method according to this aspect includes an image acquisition step of acquiring a plurality of cell images in which a group of stem cells in culture are imaged at a plurality of predetermined times during the period of cell culture for stem cell transplantation in regenerative medicine, a region identification step of identifying each individual stem cell region in each of the plurality of cell images, an index acquisition step of acquiring a plurality of types of index values regarding the group of stem cells for each of the plurality of cell images based on the identified stem cell regions, and an estimation step of estimating the treatment effect of transplantation of the group of stem cells obtained by the cell culture before transplantation based on the plurality of types of index values acquired, and the plurality of types of index values acquired in the index acquisition step include a plurality of types of index values regarding the morphology of stem cells.

[0005] Another aspect of the present invention is a stem cell transplantation treatment effect estimation device including at least one or more processors and one or more memories, and capable of executing the above-described method for estimating the effect of stem cell transplantation treatment when a computer program stored in the memory is executed by the processor. As another aspect, it may be a computer program that causes one or more information processing devices including at least one or more processors and one or more memories to execute the above-described method for estimating the effect of stem cell transplantation treatment, or it may be a storage medium storing such a computer program.

[0006] Figure 1 is a flowchart of the method for estimating the efficacy of stem cell transplantation therapy according to this embodiment (this estimation method). Figure 2 is a diagram showing an example of a cell image and an example of a stem cell region identified within the cell image. Figure 3 is a conceptual diagram showing an example of the hardware configuration of an information processing device (stem cell transplantation therapy efficacy estimation device) capable of executing the method for estimating the efficacy of stem cell transplantation therapy according to this embodiment. Figure 4 shows a heatmap graph of index values ​​obtained from cell images in the first half of culture. Figure 5 shows a heatmap graph of index values ​​obtained from cell images in the second half of culture. Figure 6 shows a heatmap graph of index values ​​obtained from cell images in the first half of culture. Figure 7 shows a heatmap graph of index values ​​obtained from cell images in the second half of culture. Figure 8 is a graph showing the results of principal component analysis of 20 index values ​​obtained from cell images in the second half of culture. Figure 9 is a graph showing the results of principal component analysis of 12 index values ​​obtained from cell images in the second half of culture. Detailed description of the invention

[0007] As described in the background technology section, although methods for evaluating stem cells have been proposed, existing evaluation methods only evaluate the stem cells themselves, or the number of stem cells or their proliferation rate. They are merely aimed at improving the manufacturing quality (culture quality) and manufacturing efficiency of stem cells, and do not measure the effectiveness of transplantation therapy in considering the recovery state of the patient's condition after stem cell transplantation. Currently, it is impossible to predict before stem cell transplantation whether the patient's condition or disability will improve when manufactured (cultured and proliferated) stem cells are transplanted into the patient, that is, whether the manufactured stem cells will have a medical effect when transplanted into the patient.

[0008] In this context, the inventors have discovered that by utilizing stem cell culture data, biochemical test data, physical test data, and evaluation data of daily living activities (ADL, FIM, etc.) from patients after stem cell transplantation, it is possible to predict whether a group of stem cells will be effective in treating a patient's condition based on images of the stem cell group before transplantation. This invention was made in view of these circumstances and provides a technology that enables the estimation of the therapeutic effect of stem cell transplantation before transplantation.

[0009] Preferred embodiments of the present invention (hereinafter sometimes referred to as "this embodiment") will be described below with reference to the drawings. The embodiments listed below are illustrative, and the present invention is not limited to the configuration of the embodiments below.

[0010] [Method for Estimating the Therapeutic Efficacy of Stem Cell Transplantation (This Estimation Method)] One aspect of this embodiment is a method for estimating the therapeutic effect of stem cell transplantation. The "stem cell transplantation" for which the therapeutic effect is estimated in this embodiment is a treatment method in which stem cells established from tissue extracted from the human body are cultured and proliferated outside the body and then transplanted back into the body. This embodiment can estimate the therapeutic effect of transplantation targeting various stem cells such as mesenchymal stem cells, iPS cells (induced pluripotent stem cells), and ES cells (embryonic stem cells), but is particularly suitable for estimating the therapeutic effect of transplantation targeting mesenchymal stem cells. Mesenchymal stem cell transplantation is mainly aimed at the regeneration of bone, cartilage, muscle, and nerve, and is carried out by methods such as administering cultured mesenchymal stem cell groups by intravenous infusion or directly transplanting them into the defective area (injured area). Furthermore, this embodiment can estimate the therapeutic effect of both so-called autologous transplantation and allogeneic transplantation.

[0011] Figure 1 is a flowchart of the method for estimating the efficacy of stem cell transplantation therapy according to this embodiment (hereinafter sometimes abbreviated as "this estimation method"). As shown in Figure 1, this estimation method includes at least an image acquisition step (S11), a region identification step (S12), an index acquisition step (S13), and an estimation step (S14).

[0012] In the image acquisition step (S11), multiple cell images are acquired. The multiple cell images acquired are images of the cultured stem cell population at multiple predetermined points in time within the period of cell culture for stem cell transplantation in regenerative medicine. That is, each cell image acquired shows the cultured stem cell population at multiple different timings within the culture period.

[0013] The timing of imaging each cell image can be at any of several predetermined times within the culture period. If the culture period is approximately one month, the imaging timing may be at multiple days in the first half of the culture period (for example, any multiple days from 10 to 15 days after the start of culture), or at multiple days in the second half (for example, any multiple days from 16 to 30 days after the start of culture), or at multiple days in both the first and second halves. However, it is preferable that the multiple cell images obtained include multiple cell images taken at multiple predetermined times in the second half of the cell culture period, and it is more preferable that they include multiple cell images taken at multiple predetermined times in the first half and multiple predetermined times in the second half of the cell culture period, respectively. The inventors have found that the therapeutic effect of transplantation can be appropriately estimated by using cell images taken at multiple predetermined times in the second half of the culture period, and that the therapeutic effect of transplantation can be estimated more reliably by further using cell images taken in the first half of the culture period.

[0014] Furthermore, while it is preferable that each cell image captures as many stem cells as possible, it is not necessary for all cultured stem cells to be captured. As described later, it is preferable that the stem cell population is imaged at a magnification and density such that individual stem cells are captured at a size and density sufficient to identify the stem cell region in the region identification step (S12) and to obtain indicator values ​​regarding the morphology of stem cells. The cell imaging method is not particularly limited as long as it is a method that can image the morphology of individual stem cells without fixing and staining the cells. By imaging without fixing and staining with formalin, acetone, etc., the imaged stem cell population can be safely used for transplantation. Cell images are captured, for example, using imaging methods used in phase-contrast observation. For this purpose, a phase-contrast microscope is used. However, imaging devices other than phase-contrast microscopes, such as digital cameras that can emit transmitted light, can also be used as long as they can image the morphology of individual stem cells.

[0015] In the region identification step (S12), individual stem cell regions are identified in each of the multiple cell images acquired in step (S11). Here, "stem cell region" refers to the image region that represents an individual stem cell within the cell image.

[0016] Figure 2 shows examples of cell images and stem cell regions identified within those cell images. Figure 2(a) illustrates a single cell image acquired in step (S11), and Figure 2(b) illustrates multiple stem cell regions identified in step (S12) within the cell image in Figure 2(a). In Figure 2(b), each stem cell region is shown as an area enclosed by a dashed line, but in identifying stem cell regions in step (S12), it is not necessary to add dashed lines as shown in Figure 2(b). As illustrated in Figure 2, the acquired cell image is represented in a way that distinguishes individual stem cells from other elements (background, etc.), so by recognizing the boundary between stem cells and other elements, stem cell regions can be identified. Existing image filtering or image recognition technologies can be used for identifying stem cell regions in step (S12), and the specific identification method is not limited in any way. As illustrated in Figure 2, in step (S12), multiple stem cell regions are identified within each cell image.

[0017] In the index acquisition step (S13), based on the stem cell region identified in step (S12), multiple index values ​​relating to the stem cell population are acquired for each of the multiple cell images acquired in step (S11). That is, in step (S13), for each cell image acquired in step (S11), multiple index values ​​relating to the stem cell population at a predetermined time within the culture period shown in that cell image are acquired. Here, "multiple index values" means quantitative values ​​for multiple mutually different index types. The multiple index values ​​acquired in step (S13) include multiple index values ​​relating to the morphology of stem cells. In other words, in step (S13), only multiple index values ​​relating to the morphology of stem cells may be acquired, or in addition to multiple index values ​​relating to the morphology of stem cells, one or more index values ​​relating to stem cells other than morphology may be acquired.

[0018] In step (S13), multiple indicator values ​​related to stem cells are obtained for each stem cell region identified within each cell image. The obtained indicator values ​​for each stem cell region are then aggregated on a cell image and indicator type basis, thereby obtaining multiple indicator values ​​related to the stem cell population for each of the multiple cell images. For example, either the representative value or the scattered value of the group of indicator values ​​aggregated on a cell image and indicator type basis, or both, are calculated as multiple indicator values ​​for each cell image. Here, "representative value" refers to a value that represents the group of indicator values ​​aggregated on a cell image and indicator type basis, such as the mean, mode, median, maximum, and minimum values. "Scattered value" refers to a value that indicates the degree of dispersion of the group of indicator values ​​aggregated on a cell image and indicator type basis, such as the standard deviation and variance.

[0019] Preferably, as multiple indicator values ​​related to stem cell morphology, representative and scattered values ​​of the stem cell area and perimeter are obtained for each cell image, and further, representative and scattered values ​​are obtained for each cell image for any of the following: smoothness of the stem cell contour, chord length, length in a predetermined direction, and shortest distance from the centroid to the edge. For example, in step (S13), representative and scattered values ​​of the stem cell area, representative and scattered values ​​of the stem cell perimeter, representative and scattered values ​​of the stem cell contour smoothness, representative and scattered values ​​of the stem cell chord length, representative and scattered values ​​of the stem cell length in a predetermined direction, and representative and scattered values ​​of the shortest distance from the centroid to the edge are obtained as indicator values ​​related to stem cell morphology. The inventors have demonstrated that the therapeutic effect of transplantation can be estimated by using such indicator values ​​related to stem cell morphology.

[0020] The method for obtaining various indicator values ​​for each stem cell region in step (S13) is not limited, but the following are examples of methods for obtaining various indicator values. The area of ​​a stem cell can be obtained by measuring the number of pixels included in the stem cell region within the cell image. The perimeter of a stem cell can be obtained by measuring the number of pixels on the contour line of the stem cell region. The contour line of a stem cell region can be detected as the boundary line between the stem cell region and other regions (background within the cell image). The smoothness of the stem cell contour can be obtained by calculating the smoothness of the contour line of the stem cell region. The smoothness of the contour line can be calculated using a formula for calculating curvature, etc. The chord length of a stem cell can be obtained by measuring the longest distance among the distances between two points on the contour line of the stem cell region. The length of a stem cell in a predetermined direction can be obtained by measuring the length (number of pixels) of the stem cell region in a predetermined direction (e.g., horizontal or vertical) of the cell image. The shortest distance from the centroid of a stem cell to its edge can be obtained by finding the centroid of the stem cell region and calculating the shortest distance among the distances from that centroid to the contour line of the stem cell region. Therefore, the shortest distance from the center of gravity of a stem cell to its edge can also be expressed as the inscribed circle radius, and this expression may be used hereafter. However, the morphological indicators of stem cells are not limited to these examples and may include the circularity of the stem cell, or they may include ratios of the above two indicators, such as the ratio of chord length to length in a given direction, the ratio of perimeter to inscribed circle radius, or the ratio of contour smoothness to perimeter.

[0021] Furthermore, in step (S13), it is preferable to obtain indicator values ​​related to the brightness of stem cells in addition to the indicator values ​​related to cell morphology described above. Indicators related to the brightness of stem cells include the mean brightness, the standard deviation of brightness, and the coefficient of variation of brightness. The mean brightness can be obtained by calculating the average value of the brightness of each pixel within the stem cell region. The standard deviation of brightness can be obtained by calculating the standard deviation of the brightness of each pixel within the stem cell region. The coefficient of variation of brightness can be obtained by calculating the coefficient of variation of the brightness of each pixel within the stem cell region (the value obtained by dividing the mean brightness by the standard deviation of brightness). As will be described in detail later, the inventors have demonstrated that the therapeutic effect of transplantation can be estimated by using such indicator values ​​related to the brightness of stem cells. Individual stem cells are transparent and three-dimensional, and differences in the three-dimensional shape of each stem cell can be detected by differences in brightness using cell image imaging techniques. For example, the stem cell region of flat and thin stem cells is shown brightly overall, while the stem cell region of three-dimensional, rice-grain-shaped stem cells tends to be shown dark in the center and bright at the outline. For this reason, indicator values ​​related to the brightness of stem cells can also be called indicator values ​​related to the morphology of stem cells.

[0022] In the estimation step (S14), the therapeutic effect of transplanting the stem cell population obtained by cell culture is estimated before the transplant, based on multiple indicator values ​​obtained in step (S13). The therapeutic effect of the transplant estimated in step (S14) represents the degree of recovery judged from the patient's condition after stem cell transplantation. For example, the degree of recovery judged from the patient's condition after the start of rehabilitation following transplantation is estimated as the therapeutic effect of the transplant.

[0023] Therefore, the degree of recovery of a patient's specific motor function may be estimated as the therapeutic effect of transplantation. For example, if the diseased area of ​​the patient who is the target of stem cell transplantation is the brain, the therapeutic effect of transplantation estimated in step (S14) will at least indicate the degree of recovery of walking ability. Data indicating the degree of recovery of walking ability include the rate of increase in walking speed, the rate of increase in slide length, the rate of increase in cadence (steps per minute), and the rate of improvement in left-right symmetry. The inventors found that among the data obtained by continuing to observe the rehabilitation status of patients after stem cell transplantation, the rate of increase in walking cadence clearly indicates the degree of recovery of the patient, and considered that the rate of increase in cadence can be considered as the main parameter representing the therapeutic effect of transplantation.

[0024] Even if the diseased area of ​​the patient who is the target of stem cell transplantation is not the brain, the therapeutic effect of the transplant estimated in step (S14) can be expressed as the degree of recovery of specific motor functions corresponding to that diseased area. For example, if the diseased area is the liver, the therapeutic effect of the transplant can be estimated not only from the results of blood tests to directly observe liver function, but also from the degree of recovery of motor ability and muscle strength. If the diseased area is the heart, the degree of recovery obtained from measurement data using the Bruce method, which measures motor function with an electrocardiogram, can be estimated as the therapeutic effect of the transplant. If the diseased area is cartilage, the degree of recovery of joint range of motion and walking ability can be estimated as the therapeutic effect of the transplant, and even if the diseased area is bone, the degree of recovery of motor abilities such as walking ability can be estimated as the therapeutic effect of the transplant.

[0025] Thus, the therapeutic effect of transplantation estimated in this embodiment is not merely an effect observed during hospitalization for transplantation, such as the engraftment rate in stem cell transplantation, and is judged based on comprehensive examination data, including examination data obtained during rehabilitation and evaluation data of activities of daily living (ADL). According to this embodiment, since such therapeutic effects of transplantation can be estimated during the stem cell culture period before transplantation, it is possible to obtain therapeutic effects more reliably with stem cell transplantation, and in turn, contribute to the development of regenerative medicine.

[0026] There are at least two methods for estimating the transplantation treatment effect in process (S14). One is a method (hereinafter referred to as the first method) that estimates the effect by using representative values ​​of the indicator values ​​for each indicator species related to the stem cell population for each cell image taken during each predetermined period of the culture period, and the other is a method (hereinafter referred to as the second method) that estimates the effect by applying a statistical analysis method to multiple indicator values ​​related to the stem cell population.

[0027] In the first method, in step (S13), either the representative value or the scattered value for each indicator species in each cell image, or both, are acquired as multiple indicator values, and the therapeutic effect of transplantation is estimated using either the representative value or the scattered value for each indicator species acquired in each cell image, or both. For example, the representative value for each indicator species is compared with a predetermined threshold, and the number of indicator species whose representative value exceeds the predetermined threshold is used to estimate whether the therapeutic effect of transplantation is high or not. As will be described later, the inventors have found that the mean and standard deviation of the area, perimeter, contour smoothness, chord length, horizontal length, and inscribed circle radius of stem cells are larger in the group of stem cells transplanted in patients with a high therapeutic effect than in the group of stem cells transplanted in patients with a low therapeutic effect. Furthermore, the mean value of the brightness of stem cells and the standard deviation of the coefficient of variation are also found to be larger in the group of stem cells transplanted in patients with a high therapeutic effect than in the group of stem cells transplanted in patients with a low therapeutic effect. On the other hand, it has been found that the mean luminance and the mean standard deviation of luminance are smaller in the stem cell group transplanted into patients with a high response rate to transplantation than in the stem cell group transplanted into patients with a low response rate to transplantation. Using this, for example, a threshold can be determined for each indicator species, and the representative value and threshold for each indicator species can be compared for each cell image. The number of indicator species that show a high response rate to transplantation can then be counted, and the response rate of transplantation can be estimated based on the number of such indicator species.

[0028] As an example of a second method, principal component analysis is used as a statistical analysis technique. In this case, step (S14) further involves deriving representative values ​​for each indicator type in predetermined image units in multiple cell images obtained in step (S11) based on multiple indicator values ​​obtained in step (S13), and applying principal component analysis to these derived representative values ​​for each indicator type. The therapeutic effect of transplantation is then estimated using at least the first principal component scores obtained by applying the principal component analysis.

[0029] For example, if in step (S11) multiple cell images are acquired at multiple predetermined times in the latter half of the culture period, a representative value for each indicator species is derived using the multiple cell images acquired in the latter half as a unit, and principal component analysis is performed on the derived representative values ​​for each indicator species. In step (S11), if multiple cell images are acquired in the first half of the culture period (referred to as the first half cell image group) and multiple cell images are acquired in the latter half (referred to as the second half cell image group), a representative value for each indicator species is derived using the first half cell image group as a unit, and a representative value for each indicator species is derived using the second half cell image group as a unit, and principal component analysis is applied to the representative values ​​for each indicator species of the first half cell group and the second half cell group, respectively, and principal component scores are calculated for the first half cell group and the second half cell group, respectively.

[0030] As will be described in detail later, the inventors have found that by applying principal component analysis to multiple index values ​​obtained from such cell images, it is possible to distinguish between stem cell groups transplanted to patients with a high transplantation response and stem cell groups transplanted to patients with a low transplantation response using the first principal component score. Therefore, in step (S14), the therapeutic effect of transplantation is estimated using at least the first principal component score obtained by applying principal component analysis. For example, as will be described later, the therapeutic effect of transplantation may be estimated using only the first principal component score, or the relationship between the first principal component score and the second principal component score, or the relationship between the first principal component score and the third principal component score, may be used to estimate whether the therapeutic effect of transplantation is high or low. Furthermore, as the second statistical analysis method, multivariate analysis other than principal component analysis may be used, or other statistical analysis methods such as regression analysis may be used.

[0031] In this embodiment, multiple cell images are acquired at multiple predetermined time points within the cell culture period, each image of a group of cultured stem cells. Multiple index values ​​related to the stem cell group are obtained for each cell image. These acquired index values ​​at least indicate the morphology of the stem cells at each time point within the culture period. Based on these multiple index values, the therapeutic effect after transplantation of the cultured stem cells to a patient can be estimated before the transplant. The inventors have successfully identified stem cell data (indicator species of stem cells) during the culture period that exert therapeutic effects in transplantation in regenerative medicine by comprehensively evaluating observational data (image data) of stem cell groups during the cell culture period, biochemical test data, physical test data, evaluation data of activities of daily living (ADL, FIM, etc.), or neuroscientific data after the transplantation of these stem cell groups to a patient. As a result, they have found that it is possible to estimate the therapeutic effect of transplantation before transplantation by using the index values ​​of the identified indicator species. Although there are methods to evaluate the cell culture process using the number of cells and proliferation rate during cell culture, the idea of ​​linking the morphology of individual stem cells with the therapeutic effect after transplantation to a patient to predict the therapeutic effect of transplantation is completely new. According to this embodiment, the therapeutic effect of transplantation, which is judged over a long period after transplantation, can be estimated during the stem cell culture period before transplantation, thus making it possible to obtain a more reliable therapeutic effect from stem cell transplantation.

[0032] [Implementing the Stem Cell Transplantation Therapy Efficacy Estimation Method (This Estimation Method)] At least some of the steps of this estimation method described above are executed by one or more information processing devices (computers) each equipped with one or more processors and one or more memories. For this reason, these information processing devices can also be called stem cell transplantation therapy efficacy estimation devices. Figure 3 is a diagram conceptually showing an example of the hardware configuration of an information processing device (stem cell transplantation therapy efficacy estimation device) 10 capable of executing this estimation method. The information processing device 10 is a so-called computer and has a CPU 11, memory 12, input / output interface (I / F) 13, communication unit 14, etc. The information processing device 10 may be a stationary PC (Personal Computer), or a portable terminal such as a portable PC, smartphone, or tablet.

[0033] The CPU 11 is a so-called processor, and in addition to a general CPU (Central Processing Unit), it may also include application-specific integrated circuits (ASICs), DSPs (Digital Signal Processors), GPUs (Graphics Processing Units), etc. The memory 12 is RAM (Random Access Memory), ROM (Read Only Memory), or auxiliary storage device (hard disk, etc.). The input / output I / F 13 can be connected to user interface devices such as the display device 15 and input device 16. The display device 15 is a device that displays a screen corresponding to drawing data processed by the CPU 11, etc., such as an LCD (Liquid Crystal Display) or CRT (Cathode Ray Tube) display. The input device 16 is a device that accepts user input such as a keyboard or mouse. The display device 15 and the input device 16 may be integrated and implemented as a touch panel. The communication unit 14 performs communication with other computers via a communication network and exchanges signals with other devices such as printers. Portable recording media may also be connected to the communication unit 14.

[0034] The hardware configuration of the information processing device 10 is not limited to the example in Figure 3. The information processing device 10 may include other hardware elements not shown. The number of each hardware element is also not limited to the example in Figure 3. For example, the information processing device 10 may have multiple CPUs 11. The information processing device 10 may also be implemented by multiple computers consisting of multiple enclosures.

[0035] The information processing device 10 can perform at least some steps of this estimation method by having the CPU 11 execute a computer program stored in the memory 12. This computer program is installed, for example, from a portable recording medium such as a CD (Compact Disc) or memory card, or from another computer on a network, via the input / output interface 13 or communication unit 14, and stored in the memory 12.

[0036] The information processing device 10 (CPU 11) may perform all the steps illustrated in Figure 1. In this case, the information processing device 10 can be said to include an image acquisition means capable of performing step (S11), a region identification means capable of performing step (S12), an index acquisition means capable of performing step (S13), and an estimation means capable of performing step (S14). In this case, in step (S11), the CPU 11 can acquire cell images from an imaging device such as a microscope or digital camera or from another computer via communication. Cell images may also be read from a portable recording medium. In step (S12), the CPU 11 may automatically identify multiple stem cell regions from each cell image, or it may display the cell images on the display device 15 and then identify multiple stem cell regions based on the image region specified based on user input, or it may identify them in a combination of both. In step (S14), the CPU 11 can display estimated result information showing the estimated transplantation treatment effect on the display device 15, or it can transmit the estimated result information to another computer, or it can print it via a printing device. In this case, the information processing device 10 can be said to further include output processing means. Thus, the output format of the estimated result information showing the estimated transplant treatment effect is not limited.

[0037] Furthermore, the information processing device 10 (CPU 11) may allow human manual intervention between each process. For example, in process (S13), the CPU 11 may automatically acquire multiple types of indicator values ​​for each stem cell region, and perform processes such as summarizing the acquired indicator types for each stem cell region for each cell image, and calculating representative values ​​for each indicator type for each cell image, while involving user input. In this way, this estimation method can be performed using the information processing device (stem cell transplantation treatment effect estimation device) 10 as appropriate.

[0038] The following examples illustrate the above points in more detail. However, the following examples do not limit the scope of the above-mentioned points.

[0039] The inventors of this invention have compiled disease information, cellular image data during the stem cell culture period, total cell count, number of cells administered, biochemical test data after stem cell transplantation, clinical evaluation data, evaluation data of activities of daily living (ADL, FIM, etc.), and neuroscientific data measured by magnetoencephalography for seven male patients who underwent stem cell transplantation as a form of regenerative medicine for stroke (cerebrovascular disease (cerebral hemorrhage / cerebral infarction)). Some of this data is shown in Table 1 below. The stem cell transplantation performed on these seven patients was so-called autologous transplantation, in which mesenchymal stem cells established from the patient's own skull fragments were cultured and proliferated outside the body and then re-transplanted into the body.

[0040] In Table 1, "Stroke types" indicates the type of stroke, "CH" indicates cerebral hemorrhage, and "CI" indicates cerebral infarction. "Lesioned Area" and "Lesioned" indicate the site of injury, "Pu" indicates putaminal hemorrhage, "Th" indicates thalamic hemorrhage, "MCA" indicates the middle cerebral artery region, "L" indicates the left side, and "R" indicates the right side. Thus, "Stroke types," "Lesioned Area," and "Lesioned" are disease information. "Total cell count" indicates the total number of stem cells that can be cultured and transplanted, and "Number of doses" indicates the number of transplanted stem cells. "SIAS (Stroke Impairment Assessment Set)-M" and "FIM (Functional Independence Measure) (gait)" are quantified information using well-known methods for functional assessment of stroke patients, and are clinical evaluation data. "Walking speed increase rate," "slide length increase rate," "cadence increase rate," and "left-right symmetry improvement rate" indicate the degree of recovery of walking ability. "Cadence increase rate" indicates the rate of increase in steps per minute, and "left-right symmetry improvement rate" indicates the degree of recovery from poor left-right symmetry during walking caused by stroke.

[0041] According to the post-transplant patient data shown in Table 1, there was not much difference among the seven patients in the SIAS and FIM scores, which are commonly used to assess the function of stroke patients. However, in terms of the degree of recovery of walking ability, only patient number 4 showed remarkably good results. Patient number 4's "rate of increase in walking speed," "rate of increase in slide length," and "rate of increase in cadence" showed a higher degree of recovery compared to the other patients. In particular, "cadence" is known to be an age-independent value, and patient number 4's value was significantly better than that of the other patients. From this patient data, the inventors found that even in patient groups where there was no difference in the degree of recovery of walking ability, there was a difference in the degree of recovery of walking ability, and identified a new challenge: whether this difference could be estimated as a difference in the effectiveness of transplant treatment.

[0042] On the other hand, the inventors acquired cell images by imaging the stem cell population being cultured for transplantation into each patient at multiple predetermined points during the culture period. Specifically, four cell images were acquired for each patient to whom the stem cell population was transplanted, each taken 12, 13, 14, and 15 days after the start of culture (hereinafter referred to as cell images from the first half of the culture), and four cell images were acquired for each patient taken 16, 20, 24, and 28 days after the start of culture (hereinafter referred to as cell images from the second half of the culture). Then, for the eight cell images corresponding to each patient (56 cell images for 7 patients), the region identification step (S12) in the estimation method described above was applied to identify the stem cell region, and the index acquisition step (S13) of the estimation method was executed to obtain multiple index values ​​for each cell image.

[0043] In this embodiment, in step (S13), for each stem cell region identified in each cell image, a plurality of types of index values related to the stem cells are respectively obtained, and the index values for each obtained stem cell region are grouped by cell image unit and index type unit, and the average value and standard deviation for each cell image and each index type are calculated. As a result, the average value and standard deviation for each index type with respect to each cell image are respectively obtained as a plurality of types of index values. The index types obtained in this embodiment are the area of the stem cell, the perimeter, the smoothness of the contour, the chord length, the horizontal length (horizontal length), the ratio of the horizontal length to the chord length, the circularity, the radius of the inscribed circle, the standard deviation of the luminance, and the average luminance. Specifically, the following 20 types of index values are obtained for each cell image. ・ Index value 1-1: Average value of the area ・ Index value 1-2: Average value of the perimeter ・ Index value 1-3: Average value of the smoothness of the contour ・ Index value 1-4: Average value of the chord length ・ Index value 1-5: Average value of the horizontal length ・ Index value 1-6: Average value of the ratio of the horizontal length to the chord length ・ Index value 1-7: Average value of the circularity ・ Index value 1-8: Average value of the radius of the inscribed circle ・ Index value 1-9: Average value of the standard deviation of the luminance ・ Index value 1-10: Average value of the average luminance ・ Index value 2-1: Standard deviation of the area ・ Index value 2-2: Standard deviation of the perimeter ・ Index value 2-3: Standard deviation of the smoothness of the contour ・ Index value 2-4: Standard deviation of the chord length ・ Index value 2-5: Standard deviation of the horizontal length ・ Index value 2-6: Standard deviation of the ratio of the horizontal length to the chord length ・ Index value 2-7: Standard deviation of the circularity ・ Index value 2-8: Standard deviation of the radius of the inscribed circle ・ Index value 2-9: Standard deviation of the standard deviation of the luminance ・ Index value 2-10: Standard deviation of the average luminance

[0044] Then, the inventors standardized the 20 types of index values obtained in this way for each of the 56 cell images for 7 patients by index type, arranged them as shown in FIGS. 4 and 5, and generated a heat map graph. For this standardization, the STANDARDIZE function with the original index value, the average value and standard deviation for each index type as arguments was used.

[0045] Figure 4 shows a heatmap graph of index values ​​obtained from cell images in the first half of culture, and Figure 5 shows a heatmap graph of index values ​​obtained from cell images in the second half of culture. In Figures 4 and 5, graphs are arranged horizontally for each patient's cell image, and vertically for patients 1 through 7. In Figure 4, "First Half 1" shows the cell image after 12 days, "First Half 2" shows the cell image after 13 days, "First Half 3" shows the cell image after 14 days, and "First Half 4" shows the cell image after 15 days. In Figure 5, "Second Half 1" shows the cell image after 16 days, "Second Half 2" shows the cell image after 20 days, "Second Half 3" shows the cell image after 24 days, and "Second Half 4" shows the cell image after 28 days. In each graph, the coloring is such that larger values ​​are closer to black, and smaller values ​​are closer to white.

[0046] Figure 4 shows that the graph for patient #4 is darker in color than the graphs for the other patients in the cell images for "First Half 1" and "First Half 3". Patient #3's graph is also darker in color only in the cell image for "First Half 3", but not in the images from other periods. Figure 5 shows that in all cell images from "Second Half 1" to "Second Half 4", the graph for patient #4 is darker in color than the graphs for the other patients. This shows that by obtaining at least multiple cell images taken at multiple periods in the latter half of the culture period and using the 20 indicator values ​​obtained from them, the morphology of the stem cell group transplanted to patient #4, who had a high degree of recovery of walking ability, can be distinguished from the stem cell groups transplanted to the other patients. In other words, it has been demonstrated that the difference in the degree of recovery of walking ability after stem cell transplantation can be estimated as a difference in the effectiveness of the transplantation treatment. Furthermore, it can be said that the effectiveness of the transplantation treatment can be estimated more reliably by using the 20 indicator values ​​obtained from multiple cell images taken at multiple periods in the first half of the culture period.

[0047] From such results, the average value and standard deviation for each target species are calculated in advance from a population of a large number of cell images, the index values obtained from the cell images to be estimated are standardized by the average value and standard deviation of the population, and the standardized values for each target species are compared with a predetermined threshold value. Then, by using this comparison result, it is possible to estimate the transplantation treatment effect of the cell image to be estimated.

[0048] Furthermore, the present inventors verified whether the transplantation treatment effect could be estimated in the same manner even when the above 20 index values were reduced. Specifically, excluding the average value (index value 1-6) and standard deviation (index value 2-6) of the ratio of the horizontal length to the chord length, the average value (index value 1-7) and standard deviation (index value 2-7) of the circularity, the average value (index value 1-9) and standard deviation (index value 2-9) of the luminance standard deviation, and the average value (index value 1-10) and standard deviation (index value 2-10) of the average luminance, the following 12 index values were used. - Index value 1-1: Average value of area - Index value 1-2: Average value of perimeter - Index value 1-3: Average value of smoothness of contour - Index value 1-4: Average value of chord length - Index value 1-5: Average value of horizontal length - Index value 1-8: Average value of inscribed circle radius - Index value 2-1: Standard deviation of area - Index value 2-2: Standard deviation of perimeter - Index value 2-3: Standard deviation of smoothness of contour - Index value 2-4: Standard deviation of chord length - Index value 2-5: Standard deviation of horizontal length - Index value 2-8: Standard deviation of inscribed circle radius

[0049] Then, the present inventors standardized the 12 index values thus obtained for each of the 56 cell images for 7 patients by target species, arranged them as shown in FIGS. 6 and 7, and generated a heat map graph. The method of this standardization is the same as described above.

[0050] Figure 6 shows a heatmap graph of index values ​​obtained from cell images in the first half of the culture period, and Figure 7 shows a heatmap graph of index values ​​obtained from cell images in the second half of the culture period. As can be seen from Figures 6 and 7, by using multiple cell images taken at multiple time points in the culture period, whether in the first or second half, patient No. 4 can be distinguished from the other patients. Therefore, it is possible to estimate the effect of transplantation treatment in a similar manner by using 12 index values, excluding index values ​​related to roundness and brightness, rather than being limited to the use of the 20 index values ​​mentioned above.

[0051] Furthermore, the inventors not only investigated the method of using the index values ​​directly, but also the method of applying principal component analysis to the 20 index values ​​mentioned above. Specifically, the 20 index values ​​obtained for each cell image were grouped by the patient to be transplanted (the group of stem cells to be transplanted), and the average value for each index was calculated at that level. Principal component analysis was then applied to the average values ​​of the 20 index values ​​calculated for each of the seven patients. Eigenvalue analysis of the correlation matrix was used in this principal component analysis. As a result, the first, second, and third principal component scores were calculated for each patient. This method was also performed similarly for 12 index values.

[0052] In principal component analysis for 20 indicator values, the cumulative contribution rates of the first, second, and third principal components were 0.6739, 0.8620, and 0.9577, respectively. In principal component analysis for 12 indicator values, the cumulative contribution rates of the first, second, and third principal components were 0.8690, 0.9800, and 0.9967, respectively. This indicates that in all cases, the first to third principal components can explain more than 95% of the information in the data.

[0053] Figure 8 is a graph showing the results of principal component analysis of 20 indicator values ​​obtained from cell images in the latter half of the culture period, and Figure 9 is a graph showing the results of principal component analysis of 12 indicator values ​​obtained from cell images in the latter half of the culture period. Figures 8(A) and 9(A) show the relationship between the first principal component score (PC1) and the second principal component score (PC2) calculated for seven patients in two dimensions, Figures 8(B) and 9(B) show the relationship between the first principal component score (PC1) and the third principal component score (PC3) calculated for seven patients in two dimensions, and Figures 8(C) and 9(C) show the relationship between the second principal component score (PC2) and the third principal component score (PC3) calculated for seven patients in two dimensions.

[0054] In Figures 8(A) and 9(A), and Figures 8(B) and 9(B), the position of the plot corresponding to patient number 4 is significantly different from that of the other patients. Thus, as can be seen from Figures 8 and 9, by examining the relationship between the first principal component score and the second principal component score, the relationship between the first principal component score and the third principal component score, or by looking only at the first principal component score, it is possible to distinguish the stem cell group transplanted to patient number 4 from the stem cell groups transplanted to the other patients. In this way, it has been demonstrated that by examining the relationship between the first principal component score obtained by applying principal component analysis to the 12 and 20 index values ​​mentioned above, and the second or third principal component score, it is possible to estimate the therapeutic effect of transplantation that reflects the difference in the degree of recovery of walking ability, similar to the method described above. It should be noted that while an example using multiple cell images from the latter half of the culture is shown here, it has been demonstrated that the therapeutic effect of transplantation can be estimated in the same manner even when performing principal component analysis using multiple cell images from both the first and second halves of the culture.

[0055] Based on these results, principal component analysis can be used to calculate the first, second, and third principal component scores of the index values ​​obtained from the cell image to be estimated. By determining whether the first principal component score is below a predetermined threshold (in the above example, -6), or by considering the plot positions of the first and second principal component scores and the first and third principal component scores, it is possible to estimate the transplantation treatment effect of the cell image to be estimated.

[0056] The above-mentioned content can also be specified as follows: (Note 1) A method for estimating the therapeutic effect of stem cell transplantation, comprising: an image acquisition step of acquiring multiple cell images of the cultured stem cell group at multiple predetermined time points within the period of cell culture for stem cell transplantation in regenerative medicine; a region identification step of identifying individual stem cell regions in each of the multiple cell images; an index acquisition step of acquiring multiple types of index values ​​related to the stem cell group for each of the multiple cell images based on the identified stem cell regions; and an estimation step of estimating the therapeutic effect of transplanting the stem cell group obtained by the cell culture before transplantation based on the acquired multiple types of index values, wherein the multiple types of index values ​​acquired in the index acquisition step include multiple types of index values ​​related to the morphology of stem cells. (Note 2) A method for estimating the therapeutic effect of stem cell transplantation according to Note 1, wherein the multiple types of index values ​​related to the morphology of stem cells include representative and scattered values ​​of the area and perimeter of the stem cell for each cell image, and further include representative and scattered values ​​for each cell image in any of the smoothness of the stem cell contour, chord length, length in a predetermined direction, and the shortest distance from the centroid to the edge. (Note 3) The method for estimating the therapeutic effect of stem cell transplantation according to Note 1 or 2, wherein the multiple types of index values ​​obtained in the index acquisition step further include index values ​​relating to the brightness of stem cells at each of the multiple predetermined time periods. (Note 4) The method for estimating the therapeutic effect of stem cell transplantation according to any one of Notes 1 to 3, wherein the therapeutic effect of transplantation estimated in the estimation step indicates at least the degree of recovery of walking ability when the disease site of the patient who is the target of stem cell transplantation is the brain. (Note 5) The method for estimating the therapeutic effect of stem cell transplantation according to any one of Notes 1 to 4, wherein the multiple cell images obtained in the image acquisition step are images taken at each of the multiple predetermined time periods in the latter half of the cell culture period, or images taken at each of the multiple predetermined time periods in the first half and the multiple predetermined time periods in the latter half of the cell culture period.(Note 6) The method for estimating the therapeutic effect of stem cell transplantation according to Note 5, wherein the multiple index values ​​for each of the multiple cell images obtained in the index acquisition step are either representative values ​​or scattered values ​​for each index type in each cell image, or both, and the estimation step estimates the therapeutic effect of transplantation using either representative values ​​or scattered values ​​for each index type in each cell image obtained in the index acquisition step, or both. (Note 7) The method for estimating the therapeutic effect of stem cell transplantation according to any one of Notes 1 to 5, wherein the estimation step includes: deriving representative values ​​for each index type in predetermined image units in the multiple cell images obtained in the image acquisition step based on the multiple index values ​​obtained in the index acquisition step; and applying principal component analysis to the derived representative values ​​for each index type, and estimating the therapeutic effect of transplantation using at least the first principal component scores obtained by applying the principal component analysis. (Appendix 8) A stem cell transplantation treatment efficacy estimation device comprising at least one processor and one or more memories, wherein a computer program stored in the memory is executed by the processor, thereby enabling the estimation method for the efficacy of stem cell transplantation treatment described in any one of Appendix 1 to 7.

[0057] This application claims priority based on Japanese Patent Application No. 2024-229440, filed on 25 December 2024, and incorporates all disclosures thereof herein.

[0058] 10. Information processing device (stem cell transplantation treatment efficacy estimation device), 11. CPU, 12. Memory, 13. Input / Output I / F, 14. Communication unit, 15. Display device, 16. Input device

Claims

1. A method for estimating the therapeutic effect of stem cell transplantation, comprising: an image acquisition step of acquiring multiple cell images of the cultured stem cell group at multiple predetermined time points within the period of cell culture for stem cell transplantation in regenerative medicine; a region identification step of identifying individual stem cell regions in each of the multiple cell images; an index acquisition step of acquiring multiple types of index values ​​related to the stem cell group for each of the multiple cell images based on the identified stem cell regions; and an estimation step of estimating the therapeutic effect of transplanting the stem cell group obtained by the cell culture before transplantation based on the acquired multiple types of index values, wherein the multiple types of index values ​​acquired in the index acquisition step include multiple types of index values ​​related to the morphology of stem cells.

2. The method for estimating the effect of stem cell transplantation therapy according to claim 1, wherein the multiple index values ​​relating to the morphology of stem cells include representative and scattered values ​​of the area and perimeter of the stem cells for each cell image, and further include representative and scattered values ​​for each cell image of any multiple of the following: smoothness of the stem cell contour, chord length, length in a predetermined direction, and shortest distance from the center of gravity to the edge.

3. The method for estimating the effect of stem cell transplantation therapy according to claim 2, wherein the plurality of index values ​​obtained in the index acquisition step further include index values ​​relating to the brightness of stem cells at each of the plurality of predetermined time periods.

4. The method for estimating the therapeutic effect of stem cell transplantation according to claim 1, wherein the therapeutic effect of transplantation estimated in the estimation step indicates at least the degree of recovery of walking ability when the disease site of the patient who is the target of stem cell transplantation is the brain.

5. The method for estimating the efficacy of stem cell transplantation therapy according to any one of claims 1 to 4, wherein the plurality of cell images acquired in the image acquisition step are images taken at multiple predetermined times in the latter half of the cell culture period, or images taken at multiple predetermined times in the first half and multiple predetermined times in the latter half of the cell culture period, respectively.

6. The method for estimating the therapeutic effect of stem cell transplantation according to claim 5, wherein the multiple index values ​​for each of the multiple cell images obtained in the index acquisition step are either representative values ​​or scattered values ​​for each index type in each cell image, or both, and the estimation step estimates the therapeutic effect of transplantation using either representative values ​​or scattered values ​​for each index type in each cell image obtained in the index acquisition step, or both.

7. The estimation step comprises: deriving a representative value for each index type in predetermined image units in the plurality of cell images acquired in the image acquisition step based on the plurality of index values ​​acquired in the index acquisition step; and applying principal component analysis to the derived representative value for each index type, wherein the therapeutic effect of transplantation is estimated using at least the first principal component score obtained by applying the principal component analysis, according to any one of claims 1 to 4.

8. A stem cell transplantation therapy efficacy estimation device comprising at least one processor and one or more memories, wherein a computer program stored in the memories is executed by the processor, thereby enabling the estimation method for the efficacy of stem cell transplantation therapy described in any one of claims 1 to 4.

Images