Method and system for predicting and evaluating expansion capacity of blastocyst-stage embryo after unfreezing

Abstract

The invention provides a method and a system for predicting and evaluating the expansion capability of a blastocyst-stage embryo after unfreezing. The method comprises the following steps: acquiring multi-view video data of the blastocyst-stage embryo after unfreezing based on a preset time interval; based on the multi-view video data, constructing a high-resolution time sequence image; performing region division on each frame of image in the high-resolution time sequence image to obtain a blastocyst-stage embryo image frame set subjected to region division; performing feature calculation on the blastocyst-stage embryo image frame set subjected to region division to obtain a preprocessed blastocyst-stage embryo image frame set; inputting the preprocessed blastocyst-stage embryo image frame set into a preset prediction and evaluation model, and obtaining a prediction and evaluation result of the dilatation capability of the thawed blastocyst-stage embryo, the dynamic, fine region division and feature calculation quantization state of the blastocyst-stage embryo is comprehensively presented through multi-view and high-resolution collection, the dilatation capability of the prediction and evaluation model is predicted, and the dilatation capability prediction and evaluation result of the thawed blastocyst-stage embryo is obtained. Therefore, prediction and evaluation are more scientific and objective, and the accuracy of prediction and evaluation is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and specifically to a method and system for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing. Background Art

[0002] Assessing the expansion capacity of blastocyst-stage embryos after thawing is an important step in assisted reproductive technology, because the degree of blastocyst expansion is closely related to the developmental potential of the embryo and the ultimate pregnancy success rate.

[0003] At present, the method of predicting and evaluating the expansion ability of blastocyst embryos after thawing is to make a preliminary judgment on their survival status based on the morphological pictures of the blastocysts after thawing. However, there is currently no systematic and effective scientific method to estimate the subsequent expansion ability of blastocysts. Some reproductive centers usually carry out 1 to 2 hours of incubation after the blastocysts are thawed, and take pictures again before the blastocysts are transplanted. By comparing the morphological changes of the blastocysts in these two photos, we try to evaluate their expansion. However, due to the lack of scientific comparative scoring methods, only rough judgments can be made based on the two pictures before and after at this stage. This evaluation model is extremely subjective and difficult to be accurate and objective. The expansion ability of blastocysts after thawing has become a very critical factor in whether the blastocysts can be successfully implanted after thawing and transplantation. It plays an extremely important role in the entire blastocyst transplantation process and is directly related to the final effect and success rate of the transplantation. Therefore, how to evaluate and predict the expansion ability of blastocyst embryos after thawing has become a key issue that needs to be addressed. Summary of the invention

[0004] In order to overcome the above problems existing in the prior art, the present application provides a method and system for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing, which adopts the following technical solutions:

[0005] In a first aspect, the present application provides a method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing, comprising:

[0006] Acquire multi-view video data of blastocyst-stage embryos after thawing based on preset time intervals;

[0007] Construct high-resolution time series images based on multi-view video data;

[0008] Performing regional division on each frame of the high-resolution time series image to obtain a set of regionally divided blastocyst stage embryo image frames;

[0009] Performing feature calculation on the regionally divided blastocyst stage embryo image frame set to obtain a preprocessed blastocyst stage embryo image frame set;

[0010] The preprocessed blastocyst stage embryo image frame set is input into a preset prediction and evaluation model to obtain a prediction and evaluation result of the expansion ability of the blastocyst stage embryo after thawing.

[0011] Furthermore, constructing high-resolution time series images based on multi-view video data includes:

[0012] Key video frames are extracted from multi-view video data based on a preset step size, the extracted video frames are numbered and each key video frame is marked with a corresponding timestamp, and a high-resolution time series image is constructed based on the time sequence.

[0013] Furthermore, after extracting key video frames from the multi-view video data based on a preset step size, numbering the extracted video frames and marking corresponding timestamps for each key video frame, the method further includes: performing image normalization processing on each key video frame;

[0014] The image standardization process is performed on each key video frame, including: grayscale processing is performed on each key video frame to remove the interference of color information and highlight the morphological structure of the blastocyst stage embryo; random noise in the image summer process is removed from each grayscale key video frame by a noise reduction algorithm, and the image contrast of each key video frame after noise removal is enhanced by histogram equalization.

[0015] Furthermore, the method of dividing each frame of the high-resolution time series images into regions to obtain a set of blastocyst stage embryo image frames divided into regions includes: dividing each frame of the high-resolution time series images into regions by using a U-Net model.

[0016] Furthermore, the U-Net model is used to divide each frame of the high-resolution time series image into regions, and the specific implementation steps are as follows:

[0017] The encoder of the U-Net model extracts the features of the blastocyst stage embryo image through the convolution layer. The convolution kernel in the convolution layer slides on the image, performs weighted summation on each local area, captures the local features in the blastocyst stage embryo image, and obtains the feature map of the blastocyst stage embryo image. As the network depth increases, the pooling layer downsamples the feature map. After each pooling, the size of the feature map becomes smaller, but the number of channels increases, and high-level semantic features are gradually extracted.

[0018] The decoder of the U-Net model uses transposed convolution to upsample the feature map and gradually restore the resolution of the image. Transposed convolution increases the size of the feature map by inserting zero values ​​between the elements of the input feature map and then performing a convolution operation. During the upsampling process, the decoder will make a jump connection from the feature map of the corresponding layer in the encoder to replenish the detail information lost during the downsampling process.

[0019] The feature map after the decoder restores the resolution is input into the classification layer. The classification layer determines the category of each pixel and assigns a most likely regional category label to each pixel, thereby completing the regional division of the blastocyst stage embryo image.

[0020] Furthermore, the feature calculation is performed on the regionally divided blastocyst stage embryo image frame set to obtain the preprocessed blastocyst stage embryo image frame set, including:

[0021] Based on the blastocyst stage embryo image frame set data set after region division, each frame of the image is read in sequence;

[0022] Obtaining the calculation results of regional morphological features and cell morphological features of each frame of image;

[0023] Arrange the various features calculated according to the order of image frames;

[0024] Associating the sorted feature data with the original blastocyst stage embryo image frame set after region division;

[0025] According to the results of the association, the feature data is integrated into the original image frame as additional information to generate a preprocessed blastocyst stage embryo image frame set.

[0026] In a second aspect, the present application also provides a system for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing, comprising:

[0027] A multi-view video data acquisition module, used to acquire multi-view video data of thawed blastocyst-stage embryos based on a preset time interval;

[0028] A high-resolution time series image construction module is used to construct high-resolution time series images based on multi-view video data;

[0029] A blastocyst stage embryo image frame set region division module is used to perform region division on each frame of the high-resolution time series image to obtain a region-divided blastocyst stage embryo image frame set;

[0030] A pre-processed blastocyst stage embryo image frame set module is used to perform feature calculation on the regionally divided blastocyst stage embryo image frame set to obtain a pre-processed blastocyst stage embryo image frame set;

[0031] The prediction and evaluation module is used to input the pre-processed blastocyst stage embryo image frame set into a preset prediction and evaluation model to obtain the prediction and evaluation results of the expansion ability of the blastocyst stage embryo after thawing.

[0032] In a third aspect, the present application provides an electronic device, including:

[0033] One or more processors; a memory; and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, which, when executed by the device, cause the device to perform the method as described in the first aspect.

[0034] In a fourth aspect, the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer executes the method described in the first aspect.

[0035] In a fifth aspect, the present application provides a computer program, which, when executed by a computer, is used to execute the method described in the first aspect.

[0036] In one possible design, the program in the fifth aspect may be stored in whole or in part on a storage medium packaged together with the processor, or may be stored in whole or in part on a memory not packaged together with the processor.

[0037] This application has the following beneficial effects:

[0038] This application obtains multi-view videos at preset time intervals, which can capture the dynamic changes of blastocyst embryos after thawing in all directions; based on the high-resolution time series images generated by multi-view videos, the subtle structure of blastocyst embryos is clearly presented; the high-resolution images are divided into regions frame by frame, and the blastocyst embryos are dissected into the key areas of the inner cell mass, trophoblast cells, and blastocyst cavity, going deep into the microscopic level of the embryo structure, providing a multi-level basis for evaluating the overall quality of blastocyst embryos, no longer limited to general overall observation; feature calculations are carried out on the basis of regional division, covering multi-dimensional indicators such as area, circumference, cell density, and cell morphological characteristics, and comprehensively quantifying the status of each part of the blastocyst embryo; the pre-processed image frame set is input into the preset prediction and evaluation model, and the preset prediction and evaluation model gives the expansion capacity prediction and evaluation results based on the input feature information, assisting doctors in selecting high-quality embryos and improving the success rate of assisted reproduction. This application comprehensively presents the dynamics of blastocyst embryos through multi-view and high-resolution acquisition, fine regional division and feature calculation quantification status, and the prediction and evaluation model predicts the expansion capacity, making the prediction and evaluation more scientific and objective, and improving the accuracy of the prediction and evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is an exemplary system architecture diagram to which the embodiments of the present application can be applied;

[0040] Figure 2 This is a flow chart of a method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing according to an embodiment of the present application;

[0041] Figure 3This is a flowchart of blastocyst embryo region division according to an embodiment of the present application;

[0042] Figure 4 This is a flow chart of preprocessing of blastocyst stage embryo image frame sets according to an embodiment of the present application;

[0043] Figure 5 A flowchart of obtaining prediction evaluation results according to an embodiment of the present application;

[0044] Figure 6 This is a system flow chart of an embodiment of the present application;

[0045] Figure 7 It is a schematic diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0047] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0048] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0049] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0050] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0051] Terminal devices 101, 102, 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Group Audio Layer 3), MP4 (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts Group Audio Layer 4) players, laptop computers and desktop computers, etc.

[0052] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0053] It should be noted that the method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing provided in the embodiments of the present application is generally executed by a server / terminal device, and accordingly, the system for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing is generally set in the server / terminal device.

[0054] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0055] Continue to refer Figure 2 , the figure shows a flow chart of a method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing of the present application, the method comprising the following steps:

[0056] Step 201, acquiring multi-view video data of thawed blastocyst stage embryos based on preset time intervals.

[0057] For example, after the blastocyst stage embryo is thawed, it is quickly transferred to a professional embryo cultivation device equipped with a high-definition, multi-angle camera. The embryo cultivation device needs to accurately simulate the physiological environmental conditions such as temperature, humidity, gas concentration, etc. in the uterus to ensure that the embryo can develop in a natural state.

[0058] Start the camera and continuously record the entire process of embryo cultivation and blastocyst incubation at preset time intervals (such as every 5 minutes), synchronously record video frames from multiple perspectives (at least 3 different directions to ensure that there are no visual blind spots), obtain original multi-perspective video data, and accumulate materials for a sufficient period of time (recommended to be no less than 24 hours) to provide rich resources for the subsequent construction of high-resolution time series images.

[0059] Step 202: construct high-resolution time series images based on multi-view video data.

[0060] In a possible implementation, constructing high-resolution time series images based on multi-view video data includes:

[0061] Key video frames are extracted from multi-view video data based on a preset step size, the extracted video frames are numbered and each key video frame is marked with a corresponding timestamp, and a high-resolution time series image is constructed based on the time sequence.

[0062] In a possible implementation, after extracting key video frames from multi-view video data based on a preset step size, numbering the extracted video frames and marking corresponding timestamps for each key video frame, image standardization processing is performed on each key video frame. The image standardization processing for each key video frame includes: graying each key video frame to remove the interference of color information and highlight the morphological structure of the embryo in the blastocyst stage; removing random noise in the image acquisition process from each grayed key video frame through a noise reduction algorithm, and enhancing the image contrast of each key video frame after noise removal through histogram equalization.

[0063] Step 203, performing region division on each frame of the high-resolution time series image to obtain a set of blastocyst stage embryo image frames divided into regions.

[0064] The blastocyst stage refers to the embryonic stage formed when the fertilized egg is cultured in vitro for 5-6 days. The embryo at this stage is composed of the inner cell mass, blastocyst cavity and trophectoderm, and is an important stage of embryonic development.

[0065] After the blastocyst with a higher degree of expansion of the blastocyst cavity is transplanted, the pregnancy rate is significantly higher than that of the blastocyst with a lower degree of expansion. Therefore, by evaluating the expansion ability of blastocyst-stage embryos, embryos that are more likely to achieve successful pregnancy can be screened out, thereby improving the success rate of assisted reproductive technology. The degree of expansion of blastocyst-stage embryos can be used as an important indicator for selecting high-quality blastocysts, helping doctors make more accurate transplantation decisions. By analyzing the expansion status of blastocyst-stage embryos, doctors can choose the best transplantation time point and strategy according to the specific situation, and then provide users with personalized plan formulation.

[0066] In a possible implementation, the region division of each frame of the high-resolution time series image to obtain a region-divided blastocyst stage embryo image frame set specifically includes:

[0067] This application uses the U-Net model to divide each frame of the high-resolution time series image into regions. Please refer to Figure 3 , the specific implementation steps are as follows:

[0068] Step 31, feature extraction and downsampling: The encoder of the U-Net model extracts the features of the blastocyst stage embryo image through the convolution layer. The convolution kernel in the convolution layer slides on the image, performs weighted summation on each local area, captures the local features in the blastocyst stage embryo image (such as grayscale changes in cell boundaries, texture differences between different regions, etc.), and obtains the feature map of the blastocyst stage embryo image; as the network depth increases, the pooling layer downsamples the feature map. After each pooling, the size of the feature map becomes smaller, but the number of channels increases, and high-level semantic features are gradually extracted. These semantic features can characterize the comprehensive characteristics of different regions of the blastocyst. Downsampling operations (such as maximum pooling) will reduce the resolution of the image while retaining important features, allowing the model to focus on more abstract and representative features.

[0069] Step 32, feature recovery and upsampling: The decoder of the U-Net model implements upsampling of the feature map through transposed convolution (deconvolution) to gradually restore the resolution of the image. Transposed convolution increases the size of the feature map by inserting zero values ​​between the elements of the input feature map and then performing a convolution operation. During the upsampling process, the decoder will perform a skip-connection on the feature map of the corresponding layer in the encoder to replenish the detail information lost during the downsampling process. Through cross-layer feature fusion, the U-Net model can better utilize features at different levels, combining high-level semantic features with low-level detail features to more accurately reconstruct and divide various regions in the blastocyst stage embryo image.

[0070] Step 33, regional category determination and division: The feature map after the decoder restores the resolution is input to the classification layer, and the classification layer performs category determination on each pixel. The U-Net model assigns a most likely regional category label to each pixel based on the learned feature patterns of different regions of the blastocyst (such as the inner cell mass, trophoblast cells, and blastocyst cavity), thereby completing the regional division of the blastocyst stage embryo image (for example, different regional categories are represented by different label matrices or colors). The determination of these labels is achieved by comparing the feature vectors of the pixels with the feature templates of each category learned by the U-Net model during the training process. For example, if the feature vector of a pixel best matches the feature template of the inner cell mass region, then the pixel will be classified as the inner cell mass region, thereby completing the regional division of the blastocyst stage embryo image.

[0071] Step 204 , performing feature calculation on the regionally divided blastocyst stage embryo image frame set to obtain a preprocessed blastocyst stage embryo image frame set.

[0072] In a possible implementation manner, the feature calculation is performed on the regionally divided blastocyst stage embryo image frame set to obtain the preprocessed blastocyst stage embryo image frame set. Please refer to Figure 4 , the specific contents include:

[0073] Step 41 , based on the blastocyst stage embryo image frame set data set after region division, read each frame of the image in sequence.

[0074] Step 42, obtaining the calculation results of the regional morphological features and the cell morphological features of each frame of image.

[0075] Step 43, sorting the various features obtained by calculation according to the order of the image frames; for example, creating a data table, each row represents a frame of image, each column represents a feature, and filling the corresponding feature values ​​into the table.

[0076] Step 44, associating the sorted feature data with the original set of blastocyst stage embryo image frames after region division; for example, by adding a reference to the feature data in the file header or metadata of the image frame, or creating an index file, each image frame can easily obtain its corresponding feature data.

[0077] Step 45, based on the result of the association, the feature data is integrated into the original image frame as additional information to generate a preprocessed blastocyst stage embryo image frame set. For example, the main feature values ​​can be displayed in the form of text at the edge or corner of the image frame, or the feature data can be embedded in the pixel value of the image in a certain encoding manner. Through association, in the subsequent analysis and processing process, the image content can be viewed intuitively and the corresponding feature data can be easily obtained.

[0078] In a possible implementation, the regional morphological feature calculation results and the cell morphological feature calculation results of each frame image are obtained, wherein the regional morphological feature calculation includes the regional area calculation, the regional perimeter calculation, and the regional cell density calculation; the cell morphological feature calculation includes the circularity calculation, the elongation calculation, and the nucleus to cytoplasm ratio calculation.

[0079] For example, regional area calculation: for each region, traverse all pixels in the image, and by determining the regional label to which the pixel belongs (if the region is distinguished based on color, the pixel color value can be compared; if it is based on a label matrix, the label value can be directly read), the number of pixels belonging to the region is counted and recorded as N. Using the known spatial resolution of the image, assuming that the actual area corresponding to each pixel is S (in square microns), the actual area of ​​the region A = N × S. Repeat this step for each region in the blastocyst (inner cell mass, trophoblast cells, and blastocyst cavity) to obtain their actual areas in each frame of the image. (Determine the spatial resolution information of the image, and find or record the spatial resolution-related parameters of the image, such as the actual physical size corresponding to each pixel (the unit may be microns / pixel). This information can convert the calculation results at the pixel level into feature values ​​at the actual physical level.)

[0080] Calculation of region perimeter: Edge pixel detection: For each region, use an edge detection algorithm (such as Canny edge detection) to determine the boundary pixels of the region. The Canny edge detection algorithm first smoothes the image through Gaussian filtering to reduce the impact of noise on edge detection, and then calculates the gradient amplitude and direction of the image, and extracts clear edge pixels through steps such as non-maximum suppression and double threshold detection. Count the number of boundary pixels M. Since each pixel has a corresponding actual length in physical space (assuming it is L, in microns), the actual perimeter of the region P = M × L. Similarly, calculate each region separately to obtain their actual perimeters in each frame of the image.

[0081] Regional cell density calculation: For each region, the number of cells contained therein is counted and recorded as C. Combined with the actual area A of the region calculated previously, the cell density D = C / A (in units of cells / square micron) can be obtained to obtain the cell density of each region in each frame of the image.

[0082] For example, circularity calculation (for a single cell): For a single identified cell, the outline pixel points of the cell are determined by a contour tracing algorithm (such as a method based on boundary tracing). Based on the cell outline pixel points, the area a (in square micrometers) and perimeter p (in micrometers) of the cell are calculated in a similar way to the area and perimeter calculation. The circularity is calculated using the formula Calculate the circularity of the cells. The closer the circularity value is to 1, the closer the shape of the cell is to a circle. Repeat this step for each identified cell to obtain their circularity in each frame image.

[0083] Elongation calculation (for a single cell): For the outline of a single cell, the length of the major axis and minor axis of the cell is determined by fitting the minimum circumscribed rectangle. Methods such as principal component analysis (PCA) can be used to find the main directions of the cell outline points to determine the major axis and minor axis. Calculate the length ratio of the major axis to the minor axis, that is, elongation E = major axis length / minor axis length. The elongation is greater than 1, and the larger the value, the more slender the cell. Through this step, the elongation of each cell in each frame of the image can be obtained.

[0084] Calculation of the ratio of nucleus to cytoplasm: If the nucleus and cytoplasm can be distinguished by special staining or other methods, their areas are calculated separately, assuming that the area of ​​the nucleus is A. n The cytoplasmic area is A c (Units are all square micrometers). Calculate the area ratio of the cell nucleus to the cytoplasm P = A n / A c This ratio can reflect the functional state and differentiation of cells, and can obtain the ratio of the nucleus to the cytoplasm of each cell in each frame of the image.

[0085] Step 205, inputting the pre-processed blastocyst stage embryo image frame set into a preset prediction and evaluation model to obtain a prediction and evaluation result of the expansion ability of the blastocyst stage embryo after thawing.

[0086] In a possible implementation, the pre-processed blastocyst stage embryo image frame set is input into a preset prediction and evaluation model to obtain a prediction and evaluation result of the expansion ability of the blastocyst stage embryo after thawing. Please refer to Figure 5 , the specific contents include:

[0087] Step 51 , sequentially read the pre-processed blastocyst stage embryo image frame set, perform scaling processing on each frame of the blastocyst stage embryo image frame set, and obtain a blastocyst stage embryo image frame set with a consistent size.

[0088] Step 52, normalize each frame of the blastocyst stage embryo image frame set with consistent size to a preset interval.

[0089] Step 53, the normalized blastocyst stage embryo image frame set is input into a preset prediction and evaluation model according to the time series. The preset prediction and evaluation model processes the time series information through the long short-term memory network, and finally outputs the probability value corresponding to each category through the fully connected layer and the softmax function, and the category with the largest probability value is used as the prediction and evaluation result of the expansion capacity of the blastocyst stage embryo after thawing, where the blastocyst expansion capacity is divided into three categories: high, medium, and low. For example, assuming that the output probability vector is [0.5, 0.3, 0.2], this means that the probability that the blastocyst stage embryo belongs to the high, medium, and low expansion capacity categories is 0.5, 0.3, and 0.2, respectively.

[0090] This application uses a long short-term memory network to capture the dynamic trend of blastocyst embryo morphology over time.

[0091] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0092] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0093] Continue to refer Figure 6 The system for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing described in this embodiment includes:

[0094] A multi-view video data acquisition module 601 is used to acquire multi-view video data of thawed blastocyst-stage embryos based on a preset time interval;

[0095] A high-resolution time series image construction module 602 is used to construct a high-resolution time series image based on multi-view video data;

[0096] The blastocyst stage embryo image frame set region division module 603 is used to divide each frame of the high-resolution time series image into regions to obtain a blastocyst stage embryo image frame set with region division;

[0097] A pre-processed blastocyst stage embryo image frame set module 604 is used to perform feature calculation on the regionally divided blastocyst stage embryo image frame set to obtain a pre-processed blastocyst stage embryo image frame set;

[0098] The prediction and evaluation module 605 is used to input the pre-processed blastocyst stage embryo image frame set into a preset prediction and evaluation model to obtain a prediction and evaluation result of the expansion ability of the blastocyst stage embryo after thawing.

[0099] To solve the above technical problems, the present application also provides a computer device. Figure 7 , Figure 7 This is a basic structural block diagram of the computer device in this embodiment.

[0100] The computer device 7 includes a memory 7a, a processor 7b, and a network interface 7c that are interconnected through a system bus. It should be noted that the figure only shows a computer device 7 with components 7a-7c, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0101] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0102] The memory 7a includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 7a can be an internal storage unit of the computer device 7, such as a hard disk or memory of the computer device 7. In other embodiments, the memory 7a can also be an external storage device of the computer device 7, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 7. Of course, the memory 7a can also include both the internal storage unit of the computer device 7 and its external storage device. In this embodiment, the memory 7a is generally used to store the operating system and various application software installed on the computer device 7, such as the program code of the expansion capacity prediction and evaluation method after thawing of blastocyst stage embryos, etc. In addition, the memory 7a can also be used to temporarily store various types of data that have been output or are to be output.

[0103] The processor 7b may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 7b is generally used to control the overall operation of the computer device 7. In this embodiment, the processor 7b is used to run the program code or process data stored in the memory 7a, such as running the program code of the method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing.

[0104] The network interface 7c may include a wireless network interface or a wired network interface. The network interface 7c is generally used to establish a communication connection between the computer device 7 and other electronic devices.

[0105] The present application also provides another embodiment, namely, providing a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a program of a method for predicting and evaluating the expansion capacity of a blastocyst stage embryo after thawing, and the method for predicting and evaluating the expansion capacity of a blastocyst stage embryo after thawing can be executed by at least one processor, so that the at least one processor executes the steps of the method for predicting and evaluating the expansion capacity of a blastocyst stage embryo after thawing as described above.

[0106] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0107] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing, characterized in that: include: Acquire multi-view video data of blastocyst-stage embryos after thawing based on preset time intervals; Construct high-resolution time series images based on multi-view video data; Performing regional division on each frame of the high-resolution time series image to obtain a set of regionally divided blastocyst stage embryo image frames; Performing feature calculation on the regionally divided blastocyst stage embryo image frame set to obtain a preprocessed blastocyst stage embryo image frame set; The preprocessed blastocyst stage embryo image frame set is input into a preset prediction and evaluation model to obtain a prediction and evaluation result of the expansion ability of the blastocyst stage embryo after thawing.

2. The method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing according to claim 1, characterized in that: The method of constructing high-resolution time series images based on multi-view video data includes: Key video frames are extracted from multi-view video data based on a preset step size, the extracted video frames are numbered and each key video frame is marked with a corresponding timestamp, and a high-resolution time series image is constructed based on the time sequence.

3. The method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing according to claim 2, characterized in that: After extracting key video frames from the multi-view video data based on a preset step size, numbering the extracted video frames and marking corresponding timestamps for each key video frame, the method further includes: performing image standardization processing on each key video frame; The image standardization process is performed on each key video frame, including: grayscale processing is performed on each key video frame to remove the interference of color information and highlight the morphological structure of the blastocyst stage embryo; random noise in the image summer process is removed from each grayscale key video frame by a noise reduction algorithm, and the image contrast of each key video frame after noise removal is enhanced by histogram equalization.

4. The method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing according to claim 1, characterized in that: The method of dividing each frame of the high-resolution time series images into regions to obtain a set of blastocyst stage embryo image frames divided into regions includes: dividing each frame of the high-resolution time series images into regions by using a U-Net model.

5. The method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing according to claim 4, characterized in that: The U-Net model is used to divide each frame of the high-resolution time series image into regions, and the specific implementation steps are as follows: The encoder of the U-Net model extracts the features of the blastocyst stage embryo image through the convolution layer. The convolution kernel in the convolution layer slides on the image, performs weighted summation on each local area, captures the local features in the blastocyst stage embryo image, and obtains the feature map of the blastocyst stage embryo image. As the network depth increases, the pooling layer downsamples the feature map. After each pooling, the size of the feature map becomes smaller, but the number of channels increases, and high-level semantic features are gradually extracted. The decoder of the U-Net model uses transposed convolution to upsample the feature map and gradually restore the resolution of the image. Transposed convolution increases the size of the feature map by inserting zero values ​​between the elements of the input feature map and then performing a convolution operation. During the upsampling process, the decoder will make a jump connection from the feature map of the corresponding layer in the encoder to replenish the detail information lost during the downsampling process. The feature map after the decoder restores the resolution is input into the classification layer. The classification layer determines the category of each pixel and assigns a most likely regional category label to each pixel, thereby completing the regional division of the blastocyst stage embryo image.

6. The method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing according to claim 1, characterized in that: The step of performing feature calculation on the regionally divided blastocyst stage embryo image frame set to obtain the preprocessed blastocyst stage embryo image frame set includes: Based on the blastocyst stage embryo image frame set data set after region division, each frame of the image is read in sequence; Obtaining the calculation results of regional morphological features and cell morphological features of each frame of image; Arrange the various features calculated according to the order of image frames; Associating the sorted feature data with the original blastocyst stage embryo image frame set after region division; According to the results of the association, the feature data is integrated into the original image frame as additional information to generate a preprocessed blastocyst stage embryo image frame set.

7. A system for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing, used to implement the method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing according to claims 1-6, characterized in that: include: A multi-view video data acquisition module, used to acquire multi-view video data of thawed blastocyst-stage embryos based on a preset time interval; A high-resolution time series image construction module is used to construct high-resolution time series images based on multi-view video data; A blastocyst stage embryo image frame set region division module is used to perform region division on each frame of the high-resolution time series image to obtain a region-divided blastocyst stage embryo image frame set; A pre-processed blastocyst stage embryo image frame set module is used to perform feature calculation on the regionally divided blastocyst stage embryo image frame set to obtain a pre-processed blastocyst stage embryo image frame set; The prediction and evaluation module is used to input the pre-processed blastocyst stage embryo image frame set into a preset prediction and evaluation model to obtain the prediction and evaluation results of the expansion ability of the blastocyst stage embryo after thawing.

8. An electronic device, characterized in that: include: one or more processors; Memory; and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, which, when executed by the device, enable the device to perform the steps of the method for predicting and evaluating the expansion capacity of blastocyst stage embryos after thawing as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed on a computer, enables the computer to execute the steps of the method for predicting and evaluating the expansion capacity of blastocyst-stage embryos after thawing as described in any one of claims 1 to 6.

Images