Egg evaluation method, egg evaluation device, and egg evaluation program

Abstract

An aspect of an ovum evaluation method according to the present invention is a method for evaluating an ovum, the method comprising: analysis steps (S21-S24) of using an image obtained by imaging a target ovum to analyze the state of deformation of the ovum and noninvasively determining an indicator value obtained by digitizing the dynamic characteristics thereof; and an evaluation step (S25) for evaluating the quality of the target ovum on the basis of the indicator value obtained in the analysis steps. Consequently, the quality of an ovum can be accurately evaluated according to a theoretical standard, on the basis of the results of performing noninvasive measurement or observation.

Description

[Technical Field]

[0001] The present invention relates to a method, an apparatus and a computer program for assessing egg quality. [Background technology]

[0002] Well-known clinical treatments for human assisted reproductive technology include in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICS). IVF involves inducing fertilization by coexisting eggs extracted from a patient with sperm (insemination), and then implanting the fertilized eggs into the mother (embryo transfer) after culturing them for a predetermined period of time. ICS, on the other hand, involves inserting a thin glass needle called an injection pipette into an egg extracted from a patient, and then injecting a single sperm into the egg using the injection pipette to achieve fertilization. ICS is performed by a specialized technician (commonly known as an "embryologist") under microscope observation.

[0003] In ICSI, fertilization is performed manually. To increase fertilization rates and post-embryo implantation rates, it is crucial to select and inseminate mature, high-quality eggs from the multiple eggs collected from the patient. The selection of high-quality eggs is also an important task for embryologists. Traditionally, evaluation of egg quality and selection based on the evaluation results have relied solely on the manual work of embryologists. This inevitably leads to variations in results due to differences in the experience and skill level of the embryologists. Furthermore, this work places a heavy burden on embryologists, making it difficult to increase efficiency (throughput). Furthermore, in many cases, egg quality is not evaluated according to clear, quantitative criteria, making it difficult to verify or check whether the evaluation was appropriate.

[0004] It is well known from the experience of embryologists that the quality of eggs is related to their hardness. For this reason, techniques have been proposed to measure the hardness of eggs in order to evaluate their quality.

[0005] For example, Patent Document 1 discloses a technique for quantitatively measuring the hardness, which is one of the mechanical characteristics of an egg, from the deformation response, strain state, or deformation state of the outer membrane based on the local deformation of the egg, which is observed in an image of the egg being clamped by a microprobe.

[0006] Non-Patent Document 1 shows that when a human embryo is sucked with a manipulator, if it is too hard or too soft, its developmental potential is reduced.

[0007] Non-patent document 2 discloses a technology for estimating the rigidity of a cell by attaching a micro-force sensor using a strain gauge to the tip of a two-fingered microhand and using the micro-force sensor to measure the reaction force generated in the end effector when the two-fingered microhand grasps the cell.

[0008] Non-patent document 3 describes the use of a micro-tactile sensor system with a strain gauge attached to the base of a needle to verify that the elasticity of the zona pellucida of an egg changes specifically at each stage of egg maturation, fertilization, and early embryonic development. [Prior art documents] [Patent documents]

[0009] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-54684 [Non-patent literature]

[0010] [Non-Patent Document 1] Livia Z. Yanez and 4 others, “Human oocyte developmental potential is predicted by mechanical properties within hours after fertilization”, Nature Communications, Vol.7, No.1, 2016, pp.1-12 [Non-patent document 2] "Force measurement in a microenvironment using a two-fingered microhand (Arai Takeo Laboratory, Osaka University)," [Online], [Retrieved January 22, 2021], Internet <url: http: www.arailab.sys.es.osaka-u.ac.jp index.php?option="com_content&view=article&id=75&Itemid=97&lang=ja"> [Non-patent document 3] Yoshinobu Murayama and 9 others, "Elasticity measurement of zona pellucida using a micro tactile sensor to evaluate embryo quality", Journal of Mammalian Ova Research, Vol.25, No.1, 2008, pp.8-16 [Non-patent document 4] Hiroaki Ito and 7 others, "Quantification of the influence of endotoxins on the mechanics of adult and neonatal red blood cells", The Journal of Physical Chemistry B, Vol. 119, No. 25, 2015, pp. 7837-7845 Summary of the Invention [Problem to be solved by the invention]

[0011] However, all of the above-mentioned conventional techniques use invasive techniques to obtain information related to the hardness and elasticity of eggs for the purpose of evaluating their quality. In ICSI, temporarily holding eggs during insemination and inserting an injection pipette into the eggs are essential steps in the treatment, but performing invasive manipulations on eggs for other procedures is undesirable due to the risk of damaging the eggs. Furthermore, the method described in Patent Document 1 requires the eggs to be captured and clamped in a dedicated chip, which requires extra work not found in normal treatment processes and raises concerns that the prolonged procedure time may increase damage to the eggs.

[0012] Furthermore, since an egg is generally spherical, if the contact position of the sensor sandwiching the egg deviates from the position facing the center of the egg, the force is applied obliquely, making it impossible to accurately measure elasticity, etc. Also, since the sensor does not always contact the same position on the egg, reproducibility is low when repeated measurements are made. Furthermore, since an egg is usually not a perfect sphere but a distorted sphere, it is quite difficult to contact the sensor on an axis passing through the center of the egg.

[0013] Meanwhile, in recent years, systems have been developed that use machine learning techniques such as deep learning to non-invasively assess egg quality, outputting an assessment result for the quality of an egg when an image of the egg is input. However, with these methods, the process of assessing egg quality is a black box, so to speak, and they cannot be said to be a theoretical assessment of egg quality according to clear, quantitative standards, and in that respect, there is not much difference from traditional assessments of egg quality that rely on the knowledge and experience of embryologists.

[0014] The present invention has been made to solve these problems, and its main purpose is to provide an egg evaluation method, an egg evaluation device, and an egg evaluation program that can evaluate egg quality accurately and in accordance with theoretical standards based on the results of non-invasive measurements and observations. [Means for solving the problem]

[0015] One aspect of the ovum evaluation method according to the present invention, which has been made to solve the above problems, is a method for evaluating an ovum, comprising: an analysis step of non-invasively determining an index value that quantifies the mechanical properties of a target ovum by analyzing the state of deformation of the target ovum using an image of the target ovum; an evaluation step of evaluating the quality of the target ovum based on the index value obtained by the analysis step; The following will be implemented.

[0016] Furthermore, one aspect of the ovum evaluation device according to the present invention, which has been made to solve the above-mentioned problems, is a device for carrying out the ovum evaluation method according to the above-mentioned aspect of the present invention, an analysis unit that receives an image obtained by photographing the target ovum and analyzes the state of deformation of the target ovum using the image to non-invasively obtain an index value that quantifies the mechanical properties of the target ovum; an evaluation unit that evaluates the quality of the target ovum based on the index value obtained by the analysis unit and outputs the evaluation result; Equipped with.

[0017] In order to solve the above problems, one aspect of the ovum evaluation program according to the present invention is a program for evaluating ovum using a computer, which includes: On the computer, an analysis step of non-invasively determining an index value that quantifies the mechanical properties of a target ovum by analyzing the state of deformation of the target ovum using an image of the target ovum; an evaluation step of evaluating the quality of the target ovum based on the index value obtained by the analysis step; This is what causes the following to be executed.

[0018] Here, the "target ovum" refers to an ovum that is to be fertilized by in vitro fertilization or intracytoplasmic sperm injection (ICS), and furthermore, to be fertilized and then cultured for embryo transfer. Therefore, the tasks and operations necessarily performed on ovum in such assisted reproductive technology do not fall under the category of invasive in this specification and the present invention, even if they involve direct contact with the ovum, such as holding the ovum for fertilization in ICSI or inserting an injection pipette.

[0019] Furthermore, in the present invention, the "image" of the target ovum may be an image taken at a single point in time, i.e., a still image, or a moving image or time-lapse image consisting of a plurality of temporally consecutive images. Therefore, the "state of deformation of the target ovum" may be the state of the ovum at a certain point in time when it is deformed, or the behavior of the ovum during the process of its deformation.

[0020] Furthermore, in the present invention, the evaluation results from the "evaluation of egg quality" can include a result of whether the quality of a single egg is good or bad, a result of ranking the quality of multiple eggs, a multi-stage evaluation result corresponding to known grading used in evaluating fertilized eggs such as the Veeck classification or the Gardner classification, and even a judgment result regarding the suitability of embryo transfer or cryopreservation, etc. In the present invention, such evaluation results are output as a result of processing by a device (which may include a computer), and the evaluation results can correspond to the judgment results of a highly skilled and experienced embryologist.

[0021] Furthermore, the oocyte evaluation program of the above aspect of the present invention can be provided to the user by being stored on a computer-readable non-transitory recording medium, such as a CD-ROM, DVD-ROM, memory card, or USB memory (dongle). The program can also be provided to the user in the form of data transfer via a communication line such as the Internet. Furthermore, the program can be pre-installed in a computer that is part of the system (strictly speaking, a storage device that is part of the computer) when the user purchases the system. [Effects of the Invention]

[0022] The ovum evaluation method, device, and program of the above aspects of the present invention enable the quality of a target ovum to be evaluated non-invasively and accurately, based on index values ​​that quantify mechanical properties, i.e., with theoretical justification. This allows the ovum to be evaluated without damaging the ovum, and allows the ovum to be used effectively for its intended purpose after evaluation, for example. Furthermore, the present invention reduces the workload of embryologists and eliminates variations in quality evaluations that depend on the skill of the embryologist in charge, thereby improving the reliability of ovum quality evaluations. Furthermore, because the data obtained when evaluating ovum quality is retained as numerical values, the embryologist, etc., can easily and objectively verify later whether the evaluation was appropriate. [Brief explanation of the drawings]

[0023] [Figure 1] 1 is a flowchart showing an example of work and processing procedures in intracytoplasmic sperm injection using an embodiment of the ovum evaluation method according to the present invention. [Figure 2] 2 is a flowchart showing the detailed processing steps of the pre-fertilization oocyte evaluation process in the ICSI work and processing procedure shown in FIG. 1. [Figure 3] 1 is a schematic diagram illustrating an example of an ovum evaluation device for performing a pre-fertilization ovum evaluation process and a fertilized ovum evaluation process. [Figure 4] FIG. 10 is a diagram showing an example of an image of an ovum photographed by the ovum evaluation device. [Figure 5] FIG. 5 is a graph showing the results of measuring the luminance profile on the line U shown in FIG. 4. [Figure 6] FIG. 10 is an explanatory diagram of a process for averaging the luminance values ​​of each point on a line U. [Figure 7] A diagram showing the relationship between the angle φ and the radius r(φ) of the egg. [Figure 8] FIG. 10 is a diagram showing the mean square amplitude (MSA) of the radius calculated from the measured data and a fitting curve using a theoretical formula. [Figure 9] An explanatory diagram of a method for calculating the aspect ratio of a deformed ovum when an injection pipette is inserted. [Figure 10] A diagram showing the deformation of an oocyte when held by a holding pipette. [Figure 11] FIG. 10 is an explanatory diagram of how to determine the index value of oocyte deformation relaxation after the hole is opened by inserting the injection pipette. [Figure 12] FIG. 1 shows the time course of oocyte deformation relaxation after the hole is opened by the injection pipette insertion. [Figure 13] An explanatory diagram of how to calculate the internal area of ​​an ovum when inserting an injection pipette. DETAILED DESCRIPTION OF THE INVENTION

[0024] An embodiment of an ovum evaluation method and an ovum evaluation device according to the present invention will be described with reference to the accompanying drawings. Fig. 1 is a flowchart showing a series of work and processing procedures in ICSI using one embodiment of the ovum evaluation method according to the present invention. Fig. 2 is a flowchart showing detailed processing procedures for pre-fertilization ovum evaluation processing in the ICSI work and processing procedures shown in Fig. 1. Fig. 3 is a schematic diagram of one embodiment of the ovum evaluation device according to the present invention for carrying out the pre-fertilization ovum evaluation processing and the fertilized ovum evaluation processing shown in Fig. 1.

[0025] The ovum evaluation method and ovum evaluation device of this embodiment are primarily intended to evaluate the quality of ovum and select high-quality ovum when performing intracytoplasmic sperm injection (ICSI), a type of assisted reproductive technology. However, some of the techniques in the ovum evaluation method of this embodiment can also be used to evaluate and select ovum for in vitro fertilization rather than ICSI.

[0026] When performing ICSI, a doctor first collects eggs from a patient. Typically, multiple eggs are collected from the patient (Step S1).

[0027] Next, the embryologist places the collected eggs, either collectively or individually, in an egg evaluation device (described below) and performs a predetermined operation on the device. In response to this operation, the egg evaluation device performs a pre-fertilized egg evaluation process on the eggs as the first stage of egg evaluation (Step S2). Based on the evaluation results from the egg evaluation device, the embryologist selects one or more eggs for ICSI (Step S3).

[0028] The embryologist performs the insemination procedure while observing the selected ovum under a microscope. The state of the ovum during the insemination procedure is captured as a moving image by the ovum evaluation device (step S4).

[0029] Next, the embryologist performs a predetermined operation on the egg evaluation device. In response to this operation, the egg evaluation device performs a second-stage egg evaluation process, called fertilization-time egg evaluation, based on images taken during the insemination procedure (step S5). Based on the evaluation results from the egg evaluation device, the embryologist then selects one or more eggs (fertilized eggs) to be cultured (or cryopreserved) (step S6). The embryologist then cultures the selected fertilized eggs for a predetermined period of time, after which the doctor implants the resulting embryos into the patient (step S7). Alternatively, the embryologist may cryopreserve the fertilized eggs after the predetermined period of culture.

[0030] As described above, in the egg evaluation method of this embodiment, the quality of eggs is evaluated using an egg evaluation device at two stages: before fertilization and at the time of fertilization, thereby ensuring the quality of the fertilized eggs to be implanted into patients. Both of these egg evaluation stages are performed before the fertilized eggs are placed in an incubator, allowing for the early selection of eggs with a high rate of successful fertilization. This allows for the efficient selection of fertilized eggs to be placed in an incubator with limited capacity, thereby reducing culture loss. Furthermore, the turnover rate of incubators can be increased, reducing the number of patients waiting for infertility treatment.

[0031] Next, the configuration of the ovum evaluation device shown in FIG. 3, which is used to evaluate the quality of ovum, will be described. This ovum evaluation device comprises a microscopic observation unit 1 including an imaging unit 10, an imaging control unit 2, a data processing unit 3, a main control unit 4, an input unit 5, and a display unit 6.

[0032] The data processing unit 3 includes, as functional blocks, an image data storage unit 30, a contour extraction unit 31, a radius calculation unit 32, a radius displacement calculation unit 33, a mechanical parameter calculation unit 34, a first evaluation unit 35, an aspect ratio calculation unit 36, and a second evaluation unit 37. As will be described in detail later, the first evaluation unit 35 performs classification or regression by machine learning, and for this purpose has a trained model storage unit 350 in which a trained model trained in advance using training data is stored.

[0033] The microscopic observation unit 1 may be either a bright-field microscope or a phase-contrast microscope. The imaging unit 10 may acquire either color or monochrome images. The imaging unit 10 may be a video camera capable of capturing video at a typical frame rate (60 frames / second), or a camera that performs time-lapse imaging at a predetermined time interval with an appropriately reduced frame rate.

[0034] At least some of the functions of the data processing unit 3 and the main control unit 4 can be implemented by using a computer such as a personal computer as a hardware resource, and executing control and processing software (programs) pre-installed on the computer to perform the operations of the above-mentioned functional blocks. In this case, the input unit 5 and the display unit 6 are a keyboard, a pointing device (such as a mouse), and a monitor attached to the personal computer, respectively. This computer program is one embodiment of the oocyte evaluation program according to the present invention.

[0035] Next, the work performed by the embryologist and the processing operations of the device when performing the processing of step S2 in FIG. 1, that is, the pre-fertilized ovum evaluation processing, using the ovum evaluation device will be described with reference to FIG.

[0036] The embryologist (or other person in charge) places the target ovum 100 to be evaluated on the stage of the microscopic observation unit 1 so that its polar body is not reflected in the image, and issues a command to start analysis via the input unit 5. Upon receiving this command, the photography control unit 2 operates the imaging unit 10 to capture a video of the ovum 100 for a predetermined time (e.g., approximately 30 seconds) (step S20). The video image data obtained by the imaging unit 10 is transferred to the data processing unit 3 and temporarily stored in the image data storage unit 30.

[0037] Next, the contour extraction unit 31 in the data processing unit 3 executes a process of extracting the contour of the ovum in each frame image of the moving image (step S21). Specifically, the following process is executed.

[0038] The contour extraction unit 31 first performs pre-processing on each frame image, performing noise reduction using a Gaussian filter, median filter, or the like. Figure 4 shows an example of a photographed image of an ovum. An ovum (egg cell) is covered by a cell membrane, and outside this membrane is a type of protective layer called the zona pellucida. The above-mentioned noise reduction process is intended to remove fine, salt-and-pepper noise that appears in the image near the zona pellucida and cell membrane.

[0039] The contour extraction unit 31 then obtains a brightness profile for each frame image, which indicates the change in brightness value of pixels (or pixels at predetermined intervals) along a straight line extending outward from the center point of the ovum. Fig. 5 shows the brightness profile on the straight line U shown in Fig. 4. Here, the center point of the ovum can be obtained by approximating the center coordinates of a circle using, for example, the least squares method (calculating the center of a circle from the contour coordinates).

[0040] To obtain a clear image with reduced noise, strong illumination light can be used, but such strong light can damage the oocyte. Therefore, it is necessary to reduce the illumination light when photographing an oocyte. As a result, noise (luminance noise) caused by photographing under dim lighting becomes noticeable in the luminance profile, and this noise can cause errors when extracting contours from the luminance profile. Therefore, to remove the luminance noise, the contour extraction unit 31 performs a smoothing process, replacing the luminance value of each point (pixel) on the luminance profile with the average luminance value of a total of five points, including two points in front of and two points behind the pixel.

[0041] 6A and 6B are explanatory diagrams of this smoothing process. As shown in FIG. 6A, if the luminance values ​​of five consecutive points on a line U are x i-2 , x i-1 , x i , x i+1 , x i+2 Then, the brightness value of the central point among the five points is [x i ] =(x i-2 +x i-1 +x i +x i+1 +x i+2 ) / 5. By performing the same calculation for each point, the brightness values ​​of the above five points are calculated as [x i-2 ], [x i-1 ], [x i ], [x i+1 ], [x i+2 This allows you to obtain a luminance profile that reduces the influence of luminance noise.

[0042] The contour extraction unit 31 then selects the maximum (or minimum) of the differential values ​​in the smoothed brightness profile along multiple straight lines extending radially in different directions from the center point of the ovum as a point on the contour. By finding points on the contour for each straight line drawn at a predetermined angular interval over a 360° angle range around the center point of the ovum and connecting these points, the contour of the target ovum can be extracted with high accuracy.

[0043] Non-Patent Document 4, written by some of the present inventors, discloses extracting the contours of red blood cells, which have a nearly circular shape, from an image and determining the radius (the distance from the center point to the contour). The present inventors investigated whether a similar method could be applied to ovum, but found that it would be difficult to apply as is. The reason for this is that, in the case of red blood cells, the shape of the brightness profile curve is smooth and the contour appears quite clearly, whereas in the case of ovum, a thick pellucida zone exists outside the cell membrane, which contains a lot of noise, especially near the cell membrane, making it difficult to extract the contour.

[0044] Therefore, the contour extraction unit 31 first calculates a luminance profile along each of 64 radial lines spaced at (360 / 64)° angles around the center point of the ovum, and roughly extracts the range in the luminance profile that corresponds to the contour. Then, while gradually narrowing the calculation range of the luminance profile in the radial direction to a range estimated to include the contour, the same calculation is repeated three times along 256 radial lines spaced at (360 / 256)° angles around the center point of the ovum, ultimately obtaining a highly accurate contour.

[0045] The contour extraction unit 31 similarly extracts the contour of the ovum for all frame images constituting a video sequence spanning a predetermined time. Next, the radius calculation unit 32 calculates the radius (strictly speaking, the distance from the center point to the contour) in each direction around the center point of the ovum at predetermined angular intervals Δθ for each frame image (step S22). That is, as shown in FIG. 7, the radius r(φ) is calculated for a plurality of predetermined angles φ. Here, as an example, Δθ=1° and the radius r(φ) (φ=0 to 360°) in 360 directions is calculated for each frame image. Note that the center point when calculating the radius should be the position of the center of gravity calculated from the contour.

[0046] The shape of an ovum, a type of cell, changes over time due to fluctuations. The radial displacement calculation unit 33 calculates the radial displacement that reflects the shape fluctuations of the ovum based on information on a huge number of radii for each image frame (step S23). To achieve this, the method disclosed in Non-Patent Document 4 is used. Specifically, the radial displacement calculation unit 33 calculates the mean square amplitude (MSA), which is a function of wave number q, by performing a Fourier transform using the following equation (1):

number

[0047] Equation (1) can be understood as calculating the difference between the radius and the time average of the radius for each direction, adding up the differences for all directions, and then Fourier transforming the time change of the sum as a function of wavenumber. Figure 8 plots the mean square amplitude values ​​calculated using the above procedure based on the measured data.

[0048] On the other hand, as disclosed in Non-Patent Document 4, the following formula (2) is known as a theoretical formula expressing the elasticity of cells.

number

[0049] In equation (2), q x is the continuous wave number corresponding to the experimental q. Also, L is the length of one dimension of the cell (here, the egg). Also, k B is the Boltzmann constant, and T is the absolute temperature during the experiment. The remaining γ, σ, and κ are unknown mechanical parameters: γ is the spring constant, σ is the surface tension, and κ is the bending elasticity. The mechanical parameter calculation unit 34 calculates the three unknown mechanical parameters by fitting equation (2) to the mean square amplitude calculated from the measured data as described above (step S24). In Figure 8, the curves obtained after this fitting are shown by solid lines. These three mechanical parameters are index values ​​that reflect the mechanical properties of the ovum, i.e., its hardness and softness. In other words, the mechanical properties of the ovum are quantified by this.

[0050] Next, the first evaluation unit 35 obtains information about the quality of the egg from the values ​​of the egg mechanical parameters calculated as described above (step S25). The egg evaluation device of this embodiment uses a machine learning technique for this quality evaluation. That is, the first evaluation unit 35 determines the quality of the egg by performing classification using a trained model based on a predetermined machine learning algorithm that inputs the values ​​of the three mechanical parameters and outputs an evaluation result in which a good-quality egg is assigned a value of "1" and a poor-quality egg is assigned a value of "0."

[0051] The trained model is created in advance (for example, before the manufacturer ships the device) as follows: That is, mechanical parameters are determined for one or more eggs collected from each of a large number of patients according to the procedure described above. Meanwhile, an experienced embryologist evaluates the quality of each egg during the normal fertilization and subsequent culture process and leaves an evaluation result. The evaluation result by this embryologist can be binary information indicating whether the quality is good or not, such as whether it is suitable for implantation or cryopreservation. It may also be multivalued information based on known multi-stage evaluation methods such as the Veeck classification or the Gardner classification. Of course, during the normal fertilization and culture process, eggs that are clearly inferior may be discarded at some stage. In such cases, the evaluation result can be determined to be poor quality without the embryologist's judgment.

[0052] In this way, a large amount of training data is prepared, each set consisting of measured mechanical parameters (i.e., input data) and evaluation results (i.e., ground truth data) for a large number of eggs, and by training this training data in a neural network or the like, a trained model, i.e., an egg quality classifier, can be created. This trained model is stored in the trained model storage unit 350. The task of creating a trained model based on the training data described above may be performed by the user, but is typically performed by the manufacturer of the device or the manufacturer that provides the software.

[0053] When evaluating an unknown egg in the first evaluation unit 35, the mechanical parameter values ​​obtained for the unknown egg are input into the trained model, and the output is a result indicating whether the egg is good or bad. This evaluation result is output from the display unit 6 via the main control unit 4. If it is simply desired to obtain binary information, such as whether or not the egg is suitable for implantation, as the evaluation result of the egg quality, a binary classification machine learning algorithm can be used in the first evaluation unit 35. Furthermore, even if the correct answer data is binary, the probability that the quality is good can be calculated as a numerical value by using a machine learning algorithm for regression analysis, such as logistic regression. Therefore, in this case, the quality of the egg can be determined by, for example, comparing the calculated probability with a threshold value.

[0054] Furthermore, by creating a trained model based on a machine learning algorithm of regression analysis using objective numerical values ​​such as a fertilization rate or an implantation rate as correct answer data, the first evaluation unit 35 can obtain the evaluation result of the egg quality as a numerical value such as a fertilization rate or an implantation rate. In this case, the first evaluation unit 35 can also determine whether the numerical value of the probability output as the evaluation result is equal to or greater than a predetermined threshold value, thereby obtaining a result of whether the egg quality is good or bad and displaying the result on the display unit 6. Of course, such a numerical value of the probability may be displayed together with the quality determination result, or alternatively, the numerical value of the probability alone may be displayed.

[0055] Furthermore, in general, in infertility treatment, multiple eggs are collected from a patient, but ultimately it is necessary to narrow down the selection to one egg. Therefore, if the probability that the eggs are of good quality is obtained as a numerical value as described above, the first evaluation unit 35 can rank the multiple eggs collected from one patient based on that numerical value and display the ranking results on the display unit 6. After reviewing this, an embryologist (doctor) can refer to the ranking to select eggs to be used for ICSI from the multiple eggs or select eggs suitable for implantation into the patient's body.

[0056] Furthermore, multi-stage evaluation methods such as the Veeck classification and Gardner classification described above are well-known as methods used by embryologists to evaluate the quality of fertilized eggs. The Veeck classification is an index used to evaluate the quality of early embryos on the second to fourth day of culture after fertilization, and is a five-stage classification. On the other hand, the Gardner classification is an index used to evaluate the state of blastocysts on the fifth to sixth day of culture, and is a six-stage classification. These are methods for evaluating fertilized eggs, but a trained model can be created by training using training data in which the grading results of the Veeck classification or Gardner classification are used as correct answer data, and this trained model can be used to obtain grading results of the Veeck classification or Gardner classification from the mechanical parameter values ​​of pre-fertilized eggs.

[0057] Here, it is natural that various known algorithms that can be used for classification and regression as machine learning, such as logistic regression, support vector machines, k-nearest neighbors, random forests, linear regression, regularization, neural networks, etc., can be used as appropriate.

[0058] Furthermore, when evaluating egg quality, not only the mechanical parameters of the egg are used as evaluation indices, but also various other information about the egg, such as egg size, may be added as evaluation indices. Furthermore, instead of the egg itself, information specific to the patient, such as the patient's age, past artificial insemination results, medical history, etc. may also be added as evaluation indices.

[0059] In particular, since it is known that the patient's age has a significant impact on egg quality, it is highly appropriate to use information about the patient's age as one of the evaluation indicators. In this case, it is possible to treat age as an input to a trained model equivalent to a mechanical parameter, but instead, the results obtained by machine learning, such as a threshold for determining which eggs are of good quality among the results of ranking multiple eggs, or a threshold for determining the quality as good if the probability representing good quality is above a certain percentage, may be changed depending on the patient's age.

[0060] The inventors experimentally verified whether a trained model developed to evaluate egg quality from the perspective of whether it is suitable for implantation into a mother can accurately evaluate eggs collected from actual patients. In this verification, the quality of eggs collected from 14 patients, multiple eggs per patient, was evaluated based on the measured mechanical parameters. As a result, it was confirmed that the evaluation results using the trained model were nearly equivalent to the egg quality evaluations made by experienced embryologists.

[0061] As described above, once the quality assessment of the pre-fertilized eggs is complete, the embryologist refers to the assessment results and selects one or more eggs of good quality from the multiple eggs collected from one patient, discarding the others.The embryologist then performs the ICSI procedure on the selected eggs of good quality.

[0062] The ICSI procedure is exactly the same as conventional procedures, in which an embryologist inserts an injection pipette into an egg and injects sperm while observing the egg through a microscope. Here, this ICSI procedure is performed under observation by the microscopic observation unit 1 of the egg evaluation device shown in Figure 3. The imaging unit 10 captures the state of the egg during this procedure, particularly the state of the egg from the moment just before the injection pipette is inserted into the egg until a predetermined time has passed after the injection pipette is removed. The obtained video data is transferred to the data processing unit 3 and temporarily stored in the image data storage unit 30.

[0063] Thereafter, the embryologist (or other person in charge) performs a predetermined operation on the input unit 5. In response to this, the data processing unit 3 executes an egg evaluation process at the time of fertilization as a second-stage evaluation of egg quality based on the stored video image data. This egg evaluation will be described in detail with reference to Figure 9. As already mentioned, the operation of inserting the injection pipette into the egg is an essential operation that is indispensable for ICSI, and does not involve any operation that involves contact or invasion of the egg other than the usual operations, so this egg evaluation can also be said to be a non-invasive evaluation. Figure 9 is an image showing the state of an egg when an injection pipette is inserted into the egg for ICSI.

[0064] As shown in Figure 9, when injecting sperm into an egg, the egg is held in place by a holding pipette. The tip of the injection pipette penetrates the zona pellucida of the held egg, then penetrates the egg's cell membrane and is pushed deep into the egg cytoplasm. Almost simultaneously, the sperm is injected into the egg cytoplasm, after which the injection pipette is quickly withdrawn from the egg. As the tip of the injection pipette penetrates the zona pellucida, the tip of the injection pipette presses the zona pellucida, causing it to temporarily indent inward. As the tip of the injection pipette advances further, a hole opens in part of the zona pellucida. Specifically, as the injection pipette moves, the zona pellucida undergoes significant deformation; once the hole opens in the zona, the deformation relaxes and the zona returns to its original state. The same is true for the egg's cell membrane; as the injection pipette moves, the cell membrane undergoes significant inward deformation, causing it to indent inward; once the hole opens in the cell membrane, the deformation relaxes and the zona returns to its original state.

[0065] In the data processing unit 3, the aspect ratio calculation unit 36 ​​detects from the stored video images an image in which the zona pellucida is in the most deformed state, just before the deformation begins to relax after the large deformation of the zona pellucida that occurs when the injection pipette is inserted, as described above. The aspect ratio calculation unit 36 ​​calculates the length of the zona pellucida in the short axis direction and the length of the long axis direction from the image, and obtains the ratio of these lengths as the aspect ratio. The length in the short axis direction is the width of the zona pellucida at a position passing through approximately the center of the ovum along the direction of the injection pipette advancement and retraction (horizontal in Figure 9). On the other hand, the length in the long axis direction is the width of the zona pellucida at a position passing through approximately the center of the ovum along the direction perpendicular to the direction of the injection pipette advancement and retraction (vertical in Figure 9).

[0066] The aspect ratio of the zona pellucida, which is part of an egg, is affected by the mechanical properties of the zona pellucida, specifically its elasticity. In other words, the aspect ratio is a useful indicator for measuring the mechanical properties of an egg, and since the mechanical properties of an egg affect its quality, the quality of the egg can be evaluated based on the aspect ratio. The second evaluation unit 37 receives the aspect ratio value and compares it with a predetermined threshold to determine the quality of the egg. The result of the egg quality evaluation is then displayed on the display unit 6 via the main control unit 4. Of course, here, as with the first evaluation unit 35, the quality of the egg may also be evaluated based on indicators other than the aspect ratio, such as personal information about the patient, such as their age.

[0067] Alternatively, the quality of an oocyte may be determined by calculating the aspect ratio of the cell membrane rather than the zona pellucida, i.e., the aspect ratio at the most maximally deformed state immediately before the cell membrane begins to relax after significant deformation upon insertion of the injection pipette.Furthermore, the quality of an oocyte may be determined by using both the aspect ratio of the zona pellucida and the aspect ratio of the cell membrane.

[0068] For example, even if the quality of an egg is judged to be good in the pre-fertilization evaluation, if the quality is judged to be poor in the egg evaluation at the time of insemination, the fertilization rate and implantation rate may be poor. Therefore, by excluding fertilized eggs that are judged to be of poor quality at the time of insemination from the subjects for culture, it is possible to reduce the loss of fertilized eggs in culture and make effective use of the incubator.

[0069] In the above explanation, the aspect ratio of the zona pellucida and / or cell membrane at maximum deformation was used to evaluate the oocyte at the time of insemination. However, other information that reflects the mechanical properties of the cell membrane or zona pellucida can also be used to evaluate the quality of the oocyte, as described below.

[0070] FIG. 10 shows an example of an image showing the state of an ovum held by a holding pipette. In the example of FIG. 10(A), the cell membrane is deformed so that it clearly protrudes due to suction by the holding pipette. In contrast, in the example of FIG. 10(B), the cell membrane barely protrudes despite suction by the holding pipette. This difference in deformation reflects differences in the elasticity of the ovum. Therefore, the second evaluation unit 37 can determine the quality of the ovum by determining the presence or absence of deformation, for example, through image processing. Alternatively, the quality of the ovum may be determined by quantifying the amount of deformation (amount of protrusion) and comparing the quantitative value with a threshold value.

[0071] The second evaluation unit 37 can also determine the time required for the cell membrane deformation to relax (return to its original state) from the moment the injection pipette penetrates the cell membrane from the video images acquired during the insemination procedure (hereinafter referred to as the "relaxation time"), and judge the quality of the ovum based on this relaxation time. Figure 11 is a diagram showing the location at which changes in the cell membrane are observed when determining an index value for the relaxation of ovum deformation after the hole is opened by the injection pipette insertion. Figure 12 is a diagram showing an example of the temporal change in cell membrane deformation (relaxation of ovum deformation) after the injection pipette penetrates.

[0072] Figure 11 is an image showing the state just before the tip of the injection pipette penetrates the cell membrane. Here, we focus on three sizes, x1, x2, and x3, shown in the image as deformation of the cell membrane. x1 is the distance from the tip of the injection pipette to the position of the cell membrane facing forward just before the cell membrane opens. x2 is the depth of the depression in the cell membrane formed by being pushed by the injection pipette. x3 is the diameter of the edge of the depression in the cell membrane formed by being pushed by the injection pipette. All of these values ​​reach their maximum just before the tip of the injection pipette penetrates the cell membrane, and decrease with the passage of time after penetration.

[0073] Therefore, the data processing unit 3 calculates x1, x2, and x3 from the frame images at predetermined time intervals included in the moving image after the tip of the injection pipette penetrates the cell membrane. These measured values ​​change, for example, as shown in the plot in FIG. 12. Therefore, the second evaluation unit 37 calculates the relaxation time τ by fitting the theoretical formula (3) to the measured values. i Calculate. (x i / R0)=A i ·exp(-t / τ i ) …(3) where i = 1 to 3. R0 is the diameter of the ovum, and A i is the deformation strength. The greater the elasticity, the shorter the relaxation time. Therefore, the three relaxation times τ i These can be used as evaluation indices, and the quality of eggs can be evaluated by appropriately combining them. Of course, the above-mentioned machine learning method can be used to estimate the three relaxation times τ i may be input to derive the evaluation result of the ovum.

[0074] Alternatively, the quality of an ovum may be evaluated using the relaxation time of at least one of the sizes x1, x2, and x3 shown in Fig. 11. Alternatively, the quality of an ovum may be evaluated using, as an evaluation index, at least one of the sizes x1, x2, and x3 at the time when the cell membrane is maximally deformed.

[0075] Furthermore, instead of the size (length) of a certain region, the area of ​​the region surrounded by the zona pellucida or the area of ​​the region surrounded by the cell membrane can also be used. Figure 13 shows the region surrounded by the zona pellucida (A) and the region surrounded by the cell membrane (B) on an image of an ovum. The areas of the above regions when the deformation of the zona pellucida or cell membrane is at its maximum just before the tip of the injection pipette penetrates the cell membrane, or the change in the area of ​​the above regions as the deformation relaxes after the tip of the injection pipette penetrates the cell membrane, also reflect the elasticity of the ovum. Therefore, it is possible to evaluate the quality of an ovum by using such information as an evaluation index.

[0076] Of course, it is possible to improve the accuracy and reliability of the evaluation of egg quality by using the above-mentioned multiple evaluation indexes in combination, or by using the evaluation results obtained from each evaluation index in combination.

[0077] Furthermore, since the evaluation of egg quality by the first evaluation unit 35 and the evaluation of egg quality by the second evaluation unit 37 are completely independent evaluations, it is clear that only one of them may be performed. Of course, using both in combination in ICSI is extremely convenient, as it allows the eggs to be narrowed down to be fertilized and also the eggs to be cultured (fertilized eggs). It goes without saying that the evaluation of egg quality by the first evaluation unit 35 is also useful when selecting eggs to be subjected to in vitro fertilization rather than ICSI.

[0078] Furthermore, while the ovum evaluation device in the above-described embodiment was capable of performing both ovum evaluation before fertilization and ovum evaluation at the time of insemination, these functions can also be performed by separate devices. In the above description, the ovum evaluation device outputs the evaluation or judgment results of the set ovum quality, and the embryologist selects ovums based on these evaluation results. However, it is also possible to incorporate a function for selecting ovums based on the ovum quality evaluation results into the ovum evaluation device. In this case, only ovums judged to be of the best quality may be selected, or conversely, only ovums estimated to be clearly inferior may be automatically discarded and the others may be retained. In other words, the final selection and selection of ovums may be left to the discretion of the embryologist, or ovum selection and selection may be performed automatically without such judgment.

[0079] Furthermore, the above-described embodiment and the various modifications described above are merely examples of the present invention, and it is natural that any appropriate modifications, alterations, additions, etc. made within the spirit of the present invention will also be encompassed within the scope of the claims of the present application.

[0080] [Various aspects] It will be apparent to those skilled in the art that the above-described exemplary embodiments are examples of the following aspects.

[0081] (First Aspect) One aspect of the ovum evaluation method according to the present invention is a method for evaluating an ovum, comprising: an analysis step of non-invasively determining an index value that quantifies the mechanical properties of a target ovum by analyzing the state of deformation of the target ovum using an image of the target ovum; an evaluation step of evaluating the quality of the target ovum based on the index value obtained by the analysis step; The following will be implemented.

[0082] Furthermore, one aspect of the ovum evaluation device according to the present invention is an analysis unit that receives an image obtained by photographing the target ovum and analyzes the state of deformation of the target ovum using the image to non-invasively obtain an index value that quantifies the mechanical properties of the target ovum; an evaluation unit that evaluates the quality of the target ovum based on the index value obtained by the analysis unit; Equipped with.

[0083] Furthermore, one aspect of the ovum evaluation program according to the present invention is a program for evaluating ovum using a computer, the program including: an analysis step of non-invasively determining an index value that quantifies the mechanical properties of a target ovum by analyzing the state of deformation of the target ovum using an image of the target ovum; an evaluation step of evaluating the quality of the target ovum based on the index value obtained by the analysis step; This is what causes the following to be executed.

[0084] According to the ovum evaluation method, ovum evaluation device, and ovum evaluation program of the first aspect, the quality of a target ovum can be evaluated non-invasively and accurately, based on index values ​​that quantify mechanical properties, i.e., with theoretical justification. This prevents damage to the ovum when evaluating its quality, allowing the ovum to be used effectively for its intended purpose after evaluation, for example. Furthermore, the workload of embryologists is reduced, and variations in quality evaluations that depend on the skill of the embryologist in charge are eliminated, improving the reliability of egg quality evaluations. Furthermore, because the data from the evaluation of ovum quality is retained as numerical values, embryologists can easily verify later whether the quality evaluation was appropriate.

[0085] (Second Aspect) In the ovum evaluation method of the first aspect, the target ovum is an ovum before fertilization, and the image is a moving image or a time-lapse image of the target ovum; The analysis step may include a contour extraction step of extracting the contour of a target egg from each of a plurality of images constituting a moving image or a time-lapse image, a displacement amount acquisition step of determining the amount of displacement of the contour of the target egg, and a mechanical parameter calculation step of calculating the mechanical parameters of the target egg as the index value from the amount of displacement of the contour of the target egg.

[0086] Similarly, in the ovum evaluation device of the first aspect, the target ovum is an ovum before fertilization, and the image is a moving image or a time-lapse image of the target ovum, The analysis unit may include a contour extraction processing unit that extracts the contour of a target egg from each of a plurality of images that constitute a moving image or a time-lapse image, a displacement amount acquisition processing unit that determines the amount of displacement of the contour of the target egg, and a mechanical parameter calculation processing unit that calculates the mechanical parameters of the target egg as the index value from the amount of displacement of the contour of the target egg.

[0087] The shape of an egg, a type of cell, changes over time due to fluctuations. The manner in which the shape of an egg changes due to this fluctuation is affected by elasticity, which reflects the hardness of the egg. For example, in the above-mentioned egg evaluation method, the contour extraction step and displacement amount acquisition step obtain the change in the shape of the egg as information on the displacement amount of the contour. Then, in the mechanical parameter calculation step, the mechanical parameters of the egg are calculated from the information on the displacement amount of the contour. As a result, the egg evaluation method and egg evaluation device of the second aspect can accurately evaluate the quality of a target egg based on the mechanical parameters of the egg.

[0088] (Third Aspect) In the ovum evaluation method of the second aspect, the mechanical parameter calculation step may involve determining the mechanical parameters by performing parameter fitting using a theoretical equation that represents the elasticity of the cell.

[0089] Similarly, in the egg evaluation device of the second aspect, the mechanical parameter calculation processing unit can calculate the mechanical parameters by performing parameter fitting using a theoretical formula representing the elasticity of cells to actual measured values ​​representing the amount of displacement of the contour of the target egg.

[0090] Generally, to understand the mechanical properties of an object, such as its elasticity, it is common practice to apply some kind of force to the object and measure the object's rate of deformation in response to that force. In contrast, the egg evaluation method and egg evaluation device of the third aspect make it possible to accurately obtain information about the elasticity of an egg without applying any kind of force to the egg, even indirectly, or restricting its movement. Furthermore, to observe shape changes due to fluctuations, it is sufficient to photograph the exterior for at most about one minute, and usually for a shorter period of time, thereby virtually eliminating any substantial damage to the egg.

[0091] (Fourth Aspect) In the egg evaluation method of any of the first to third aspects, the evaluation step may evaluate the quality of the egg using training data including the results of an embryologist's evaluation of the egg at at least one stage from fertilization to implantation, or a classifier obtained by machine learning using training data in which information on the probability of success or failure at one of the stages is known.

[0092] Similarly, in any of the egg evaluation devices of the first to third aspects, the evaluation unit can evaluate the quality of the egg using training data including the evaluation results of the egg by an embryologist at at least one stage from fertilization to implantation, or a classifier created in advance by machine learning using training data in which information on the probability of success or failure at one of the stages is known.

[0093] The machine learning algorithm used is not particularly limited. The egg evaluation method and egg evaluation device of the fourth aspect can accurately evaluate eggs before culture, reflecting the judgment results of a highly skilled embryologist on the suitability of fertilized egg transplantation. Furthermore, the egg evaluation method and egg evaluation device of the fourth aspect can accurately evaluate egg quality, reflecting past success / failure results such as fertilization rate and implantation rate.

[0094] (Fifth Aspect) In the egg evaluation method of any of the first to fourth aspects, the evaluation step may evaluate the quality of the egg using patient-specific information including age in addition to the index value.

[0095] Similarly, in any of the egg evaluation devices of the first to fourth aspects, the evaluation unit can evaluate the quality of the egg using patient-specific information including pre-entered age in addition to the index value.

[0096] It is well known that fertilization and implantation rates decrease as a patient ages. While the effects of aging are often seen in decreased egg elasticity, other factors that have yet to be fully elucidated may also be involved. In contrast, the egg evaluation method and egg evaluation device of the fifth aspect incorporate patient-specific information, such as age, into the evaluation, enabling a more accurate evaluation of egg quality and a higher chance of pregnancy.

[0097] (Sixth Aspect) In the egg evaluation method of any of the first to fifth aspects, the evaluation step may involve obtaining information regarding the quality of each of multiple eggs collected from a single patient, and using this information to rank the multiple eggs.

[0098] Similarly, in the ovum evaluation device according to any of the first to fifth aspects, the evaluation unit can rank the plurality of ova using information about ova quality obtained for each of the plurality of ova.

[0099] According to the sixth aspect of the ovum evaluation method and ovum evaluation device, it is possible to perform ICSI on one ovum estimated to have the highest probability of conception among multiple ovum collected from a patient, or on a small number of ovum estimated to have a relatively high possibility of conception. Furthermore, when there are multiple fertilized ovum, it is also possible to select and implant the one estimated to have the highest probability of conception, and cryopreservation of the rest is possible.

[0100] (Seventh Aspect) In the ovum evaluation method of the first aspect, the image is an image taken when a needle is inserted into the target ovum for the purpose of ICSI, The analysis step may be to obtain an index value that reflects the degree of deformation of the zona pellucida or cell membrane at the maximum deformation caused by needle insertion, or an index value that reflects the degree of relaxation of the deformation of the zona pellucida or cell membrane caused by needle insertion.

[0101] Similarly, in the ovum evaluation device of the first aspect, the image is an image taken when a needle is inserted into the target ovum for the purpose of ICSI, The analysis unit can calculate an index value that reflects the degree of deformation of the zona pellucida or cell membrane at the maximum deformation caused by needle insertion, or an index value that reflects the degree of relaxation of the deformation of the zona pellucida or cell membrane caused by needle insertion.

[0102] In the seventh aspect of the ovum evaluation method and ovum evaluation device, the quality of the ovum is evaluated from an index value calculated based on an image taken when ICSI is performed. As described above, this evaluation of ovum quality can also be said to be performed non-invasively, in that no special invasive measurements or observations are performed to evaluate ovum quality.

[0103] According to the seventh aspect of the ovum evaluation method and ovum evaluation device, a quantitative evaluation using specific numerical values ​​can be performed instead of the subjective qualitative evaluation, such as whether the ovum is elastic or not, that has been conventionally performed by embryologists when performing ICSI. This improves the accuracy and reliability of the evaluation of ovum quality. Furthermore, the data obtained when evaluating ovum quality is saved as numerical values, making it easier to verify the accuracy of the evaluation.

[0104] (Eighth Aspect) In the egg evaluation method of the seventh aspect, the index value reflecting the degree of deformation at the maximum deformation can be the ratio of the widths in two directions, the needle insertion direction and the direction perpendicular thereto, at the maximum deformation of the zona pellucida or cell membrane of the target egg.

[0105] Similarly, in the seventh aspect of the egg evaluation device, the index value reflecting the degree of deformation at the maximum deformation can be the ratio of the widths in two directions, the needle insertion direction and the direction perpendicular to that, of the zona pellucida or cell membrane of the target egg at the maximum deformation.

[0106] According to the ovum evaluation method and ovum evaluation device of the eighth aspect, the quality of the ovum can be evaluated with high accuracy based on an index value that accurately reflects the degree of elasticity of the ovum during intracytoplasmic sperm injection.

[0107] (Ninth Aspect) In the ovum evaluation method according to any one of the first to eighth aspects, a selection step of selecting ovum using the evaluation results in the evaluation step can be further carried out.

[0108] According to the ninth aspect of the egg evaluation method, the burden of egg selection work on embryologists is reduced. Furthermore, eggs can be selected based on evaluation results that are backed by theory rather than subjective evaluation, improving the reliability of selection.

[0109] (10th aspect) In the ovum evaluation method according to any one of the first to sixth aspects, the quality of an ovum is evaluated by the analysis step and the evaluation step for an ovum before fertilization, A second analysis step of non-invasively determining an index value that quantifies the mechanical properties of an ovum selected based on the results of the evaluation by analyzing the state of deformation of the ovum using an image taken when a needle is inserted into the ovum for the purpose of ICSI; A second evaluation step of evaluating the quality of the oocyte subjected to ICSI based on the index value obtained by the second analysis step; The above may be further implemented.

[0110] Similarly, in the ovum evaluation device according to any one of the first to sixth aspects, a second analysis unit that non-invasively obtains an index value that quantifies the mechanical properties of an ovum selected based on the results of the evaluation of ovum quality by the analysis unit and the evaluation unit by analyzing the state of deformation of the ovum using an image taken when a needle is inserted into the ovum for the purpose of ICSI; and A second evaluation unit that evaluates the quality of the ovum subjected to ICSI based on the index value obtained by the second analysis unit; The device may further include:

[0111] Specifically, in the egg evaluation method and egg evaluation device of the tenth aspect, for example, egg quality can be evaluated using the egg evaluation method of the second aspect, eggs can be selected based on the results, and then the quality of the selected eggs can be evaluated again using the egg evaluation method of the seventh aspect when performing ICSI on the selected eggs. By evaluating egg quality in this way at two stages, before fertilization and at the time of fertilization, the accuracy and reliability of the evaluation can be improved, increasing the possibility of pregnancy. Furthermore, loss of fertilized eggs during culture and cryopreservation can be reduced, reducing the cost of assisted reproductive technology. [Explanation of symbols]

[0112] 1...Microscopic observation section 10...imaging unit 100...egg 2...Shooting control unit 3...Data processing section 30...Image data storage unit 31...Contour extraction unit 32...Radius calculation section 33...Radial displacement calculation unit 34...Mechanical parameter calculation section 35...First evaluation section 350...Model memory section 36...Aspect ratio calculation unit 37...Second evaluation section 4...Main control unit 5...Input section 6...Display section< / url:>

Claims

1. 1. A method for evaluating an egg, comprising: an analysis step of non-invasively determining an index value representing the mechanical properties of the target egg by analyzing the state of deformation of the zona pellucida of the target egg caused by the needle insertion using an image of the target egg when the needle is inserted into the target egg during ICSI; an evaluation step of evaluating the quality of the target ovum based on the index value obtained by the analysis step; wherein the index value is the aspect ratio of the zona pellucida.

2. The egg evaluation method of claim 1, wherein the evaluation step evaluates the quality of the egg using training data including the results of an embryologist's evaluation of the egg at at least one stage from fertilization to implantation, or a classifier obtained by machine learning using training data in which information on the probability of success or failure at one of the stages is known.

3. The ovum evaluation method according to claim 1 , wherein the evaluation step evaluates the quality of the ovum using patient-specific information including age in addition to the index value.

4. The egg evaluation method according to claim 1, wherein the evaluation step obtains information regarding egg quality for each of a plurality of eggs collected from a single patient, and ranks the plurality of eggs using the information.

5. The ovum evaluation method according to claim 1 , wherein the analysis step determines an index value that reflects the degree of maximum deformation of the zona pellucida caused by the insertion of the needle.

6. The egg evaluation method described in claim 5, wherein the index value reflecting the degree of deformation at the maximum deformation is the ratio of the widths of the zona pellucida of the target egg in two directions, the direction of needle insertion and the direction perpendicular thereto, at the maximum deformation.

7. The ovum evaluation method according to claim 1 , further comprising a selection step of selecting ovum using the evaluation results in the evaluation step.

8. 2. The egg evaluation method according to claim 1, further comprising a pre-fertilization evaluation step of evaluating the quality of the egg based on an image of the egg before fertilization, wherein the quality of the egg is evaluated by the analysis step and the evaluation step for eggs selected based on the results of the evaluation by the pre-fertilization evaluation step.

9. an analysis unit that receives an image of the target ovum when a needle is inserted into the target ovum during ICSI, and uses the image to analyze the state of deformation of the zona pellucida of the target ovum caused by the insertion of the needle, thereby non-invasively determining an index value that represents the mechanical properties of the ovum; an evaluation unit that evaluates the quality of the target ovum based on the index value obtained by the analysis unit; wherein the index value is an aspect ratio of the zona pellucida.

10. The egg evaluation device of claim 9, wherein the evaluation unit evaluates the quality of the egg using a classifier created in advance by machine learning using training data including the evaluation results of the egg by an embryologist at at least one stage from fertilization to implantation, or training data in which information on the probability of success or failure at one of the stages is known.

11. The ovum evaluation device according to claim 9 , wherein the evaluation unit evaluates the quality of the ovum using patient-specific information including pre-entered age in addition to the index value.

12. The ovum evaluation device according to claim 9 , wherein the evaluation unit ranks the plurality of ova using information about ova quality obtained for each of the plurality of ova.

13. The image is an image taken when a needle is inserted into the target ovum for the purpose of ICSI, The ovum evaluation device according to claim 9 , wherein the analysis unit calculates an index value that reflects the degree of maximum deformation of the zona pellucida caused by the insertion of the needle.

14. The egg evaluation device described in claim 13, wherein the index value reflecting the degree of deformation at the maximum deformation is the ratio of the widths of the zona pellucida of the target egg in two directions, the needle insertion direction and the direction perpendicular thereto, at the maximum deformation of the zona pellucida of the target egg.

15. The egg evaluation device according to claim 9, further comprising a pre-fertilization evaluation unit that evaluates the quality of the egg based on an image of the egg before fertilization, and for eggs selected based on the results of the evaluation by the pre-fertilization evaluation unit, evaluation of the quality of the egg is carried out by the analysis unit and the evaluation unit.

16. A program for evaluating an egg using a computer, the program comprising: an analysis step of non-invasively determining an index value representing the mechanical properties of the target egg by analyzing the state of deformation of the zona pellucida of the target egg caused by the needle insertion using an image of the target egg when the needle is inserted into the target egg during ICSI; an evaluation step of evaluating the quality of the target ovum based on the index value obtained by the analysis step; wherein the index value is the aspect ratio of the zona pellucida.

Images