Methods and oocyte quality analysis systems performed by computer systems

Abstract

This disclosure relates to a system and method for generally evaluating the viability of oocytes. In some embodiments, an image sequence is acquired relating to oocytes aspirated into a pressure tool. Based on the image sequence and the pressure applied to the oocytes, morphological and mechanical features relating to the oocytes can be derived. At least some of these features can then be input into a machine learning model to determine metrics indicating the quality of the oocytes, where certain metrics may indicate blastocyst formation. Optionally, the determined oocyte quality information can be presented to the user via an interactive user interface.

Description

Technical Field

[0001] The present disclosure relates to medical analysis using machine learning models, and more particularly to analyzing oocytes using sensor information and machine learning models.

Background Art

[0002] In recent years, progress has been made in infertility treatment. One example of treatment includes in-vitro-fertilization ("IVF"). IVF begins with an ovarian stimulation phase that stimulates egg production. Eggs (oocytes) can be collected from a patient and fertilized outside the body to form embryos. In an effort to select the most viable and / or most dominant embryos for implantation, multiple tests and analyses may be performed on the embryos. However, such tests and analyses have technical problems, and thus, an accurate method for performing this selection has technical challenges.

[0003] For example, different grading systems have been developed to support determining the viability of each embryo. These grading systems generally include manual annotation of embryo images or time-lapse videos. As can be understood, the selection process is prone to errors, for example, due to the subjective judgment of embryologists. Furthermore, there is currently no standard grading system that is widely adopted to select high-quality oocytes. Current automated technologies for embryo analysis are inaccurate and cannot utilize heterogeneous sensor information.

Summary of the Invention

[0004] A method is provided that is implemented by a system of one or more computers and an oocyte quality analysis system. The method provides a standard grading system that is widely adopted to select high-quality oocytes and utilizes different sensor information.

[0005] A method carried out by one or more computer systems includes the following steps: Multiple images are acquired to form an image sequence related to a time period, the image sequence depicting an oocyte and a portion of a tool that applies pressure to the oocyte, and the individual images are associated with individual pressure values ​​applied to the oocyte at each point in time of image capture. Based on the images and the pressure values, features related to the oocyte are determined, the features including morphological features that indicate measurements of the oocyte during the time period and mechanical features that indicate deformation of the oocyte during the time period. One or more metrics indicating oocyte quality are determined via a machine learning model based on inputs including at least a subset of the features, the particular metrics indicating blastocyst formation. The one or more metrics are configured to be presented via an interactive user interface.

[0006] In embodiments of this disclosure, the oocyte depicted in the image is of a mammal.

[0007] In embodiments of this disclosure, the image depicts the oocyte being aspirated into the portion of the tool.

[0008] In embodiments of this disclosure, the mechanical feature represents the elasticity of the oocyte, and the mechanical feature is determined based on the depth of aspiration of the oocyte to the portion of the tool.

[0009] In embodiments of this disclosure, the mechanical features represent the viscoelastic behavior of the oocyte.

[0010] In embodiments of this disclosure, the morphological features indicate at least the depth of aspiration of the oocytes to the portion of the tool relative to the corresponding image.

[0011] In embodiments of the present disclosure, one of the morphological features is derived by normalizing the aspiration depth of the oocyte to the portion of the tool in the corresponding image by the inner diameter of the tool.

[0012] In embodiments of the present disclosure, the morphological features include a first measurement relating to the cytoplasm of the oocyte and a second measurement relating to the zona pellucida of the oocyte.

[0013] In embodiments of the present disclosure, the portion of the tool is a pipette, which is configured to contact the oocyte and apply pressure during the time period.

[0014] In embodiments of the present disclosure, a first image associated with the start of the time period depicts the oocyte before pressure is applied, and a second image following the first image in the image sequence depicts the oocyte after pressure has been applied.

[0015] In embodiments of this disclosure, the pressure value is selected from the pressure range applied during the time period.

[0016] In the embodiments of this disclosure, the machine learning model is a support vector machine.

[0017] In embodiments of this disclosure, the machine learning model is a neural network.

[0018] In embodiments of the present disclosure, the method further comprises calculating a measured pressure applied to the oocyte based on the pressure value and geometric information relating to the portion of the tool, and determining the mechanical features by fitting a suction depth curve using a linear elastic model, wherein the fitting of the suction depth curve is performed based on the measured pressure and the suction depth of the oocyte to the portion of the tool for a corresponding image.

[0019] In embodiments of the present disclosure, the input further includes clinical information of a person relating to the oocyte, the clinical information including one or more of age, body mass index, developmental stage relating to the oocyte, and cryopreservation information.

[0020] A quality analysis system for oocytes, comprising one or more processors and a non-temporary computer storage medium that stores instructions that cause the one or more processors to perform the following processes when executed by the one or more processors, The aforementioned process is, The method involves obtaining multiple images that form an image sequence related to a time period, The aforementioned image sequence depicts oocytes and a portion of the tool used to apply pressure to the oocytes. Each image is associated with the individual pressure value applied to the oocyte at each point in time of image capture, and is obtained as follows: Determining characteristics related to the oocyte based on the aforementioned image and the aforementioned pressure value, The characteristics to be determined include morphological characteristics indicating the measurement of the oocyte during the time period and mechanical characteristics indicating the deformation of the oocyte during the time period. This includes determining one or more metrics indicating oocyte quality via a machine learning model based on an input that includes at least a subset of the aforementioned features, Certain metrics indicate blastocyst formation. A quality analysis system for oocytes, configured such that one or more of the aforementioned metrics are presented via an interactive user interface.

[0021] In embodiments of this disclosure, the image depicts the oocyte being aspirated into the portion of the tool.

[0022] In embodiments of this disclosure, the mechanical feature represents the elasticity of the oocyte, and the mechanical feature is determined based on the depth of aspiration of the oocyte to the portion of the tool.

[0023] In embodiments of this disclosure, the mechanical features represent the viscoelastic behavior of the oocyte.

[0024] In an embodiment of the present disclosure, the morphological feature at least indicates the aspiration depth of the oocyte into the part of the tool with respect to the corresponding image.

[0025] In an embodiment of the present disclosure, the morphological feature indicates a first measurement related to the cytoplasm of the oocyte and a second measurement related to the zona pellucida of the oocyte.

[0026] In an embodiment of the present disclosure, the pressure value is selected from a pressure range applied during the time period.

[0027] In an embodiment of the present disclosure, the instruction calculates a measured value of the pressure applied to the oocyte based on the pressure value and the geometric information related to the part of the tool, and further includes determining the mechanical feature by fitting an aspiration depth curve using a linear elastic model, where the fitting of the aspiration depth curve is performed based on the aspiration depth of the oocyte into the part of the tool with respect to the corresponding image and the measured value of the pressure.

[0028] In an embodiment of the present disclosure, the input further includes human clinical information related to the oocyte, and the clinical information includes one or more of age, body mass index, developmental stage related to the oocyte, and cryopreservation information.

[0029] A method performed by one or more computer systems, the method comprising the following steps: acquiring a plurality of image sequences depicting each oocyte, each image sequence depicting an individual oocyte and a portion of a tool that applies pressure to the oocyte, and acquiring pressure values ​​associated with the images contained in the image sequences; determining one or more metrics indicating the quality of each of the oocytes via a machine learning model, where a particular metric indicates blastocyst formation, and the input to the machine learning model for each oocyte includes morphological features indicating the measurement of the oocyte and mechanical features indicating the deformation of the oocyte during the time that pressure is applied to the oocyte; and outputting information indicating that one or more of the oocytes are selected based on the metrics.

[0030] In embodiments of this disclosure, the image depicts the oocyte being aspirated into the portion of the tool.

[0031] In embodiments of this disclosure, the mechanical feature represents the elasticity of the oocyte, and the mechanical feature is determined based on the depth of aspiration of the oocyte to the portion of the tool.

[0032] In embodiments of the present disclosure, the morphological features include a first measurement relating to the cytoplasm of the oocyte and a second measurement relating to the zona pellucida of the oocyte.

[0033] In embodiments of this disclosure, the information indicating the selection is output via an interactive user interface, and the interactive user interface is A graphical representation of one or more of the oocytes is presented, In response to user input to select a specific oocyte from one or more of the aforementioned oocytes, Detailed information relating to the aforementioned specific oocyte is presented.

[0034] In embodiments of the present disclosure, the information indicating the selection is output via an interactive user interface, which presents individual text descriptions that group together the individual oocytes of one or more of the oocytes.

[0035] A method is provided that is carried out by one or more computer systems and oocyte quality analysis systems, providing a standard grading system that is commonly employed to select high-quality oocytes and utilizing different sensor information. As a result, the method and oocyte quality analysis systems can be used to select the most viable and / or most dominant embryos for implantation and can be applied to infertility treatment. [Brief explanation of the drawing]

[0036] [Figure 1] The block diagram of an example oocyte analysis system, including a feature preprocessing engine and machine learning model, according to several embodiments of this disclosure is shown. [Figure 2A] Figure 1 shows the morphological characteristics of oocytes generated by the feature preprocessing engine in an example. [Figure 2B] Figure 1 shows the mechanical characteristics of oocytes generated by the feature preprocessing engine. [Figure 3A] A diagram of the user interface of the example, including the processing results of the example, is shown. [Figure 3B] Figure 1 shows a diagram of the user interface of an example oocyte analysis system that receives user input. [Figure 3C] A diagram of a user interface for another embodiment that displays the processing results is shown. [Figure 4] A flowchart of the process used in an example to determine an indicator of oocyte quality is shown. [Figure 5] A flowchart illustrating the process for selecting a subset of numerous oocytes based on quality indicators is shown. [Figure 6]The diagram shows a general architecture of an oocyte analysis system according to some embodiments of the present disclosure. [Modes for carrying out the invention]

[0037] This specification describes, for example, a technique for selecting high-quality eggs (e.g., oocytes) with an improved probability of blastocyst formation. As described later, the system may utilize different types of sensor information to analyze oocytes using machine learning techniques. The sensor information may include, for example, images of oocytes (unfertilized eggs) undergoing deformation due to the application of pressure, and pressure measurements indicating the pressure applied to the oocytes. This sensor information allows for an understanding of the morphological and mechanical properties of the oocytes as they undergo deformation, thereby enabling a more accurate assessment of the quality of each oocyte. In contrast, conventional techniques have relied on error-prone methods using a single type of information (e.g., images), such as manual adjustment models or grading systems. As described later, the disclosed technique utilizes different types of sensor information, along with a specially trained model, to more accurately and efficiently assess the quality or viability of eggs at an earlier stage.

[0038] To assess the viability of eggs, some conventional techniques rely on embryologists visually evaluating embryos (e.g., fertilized eggs). Some clinics record images of embryos, and embryologists may score embryos based on various grading systems and their visual evaluations. One of the major challenges in embryo selection is the high degree of effort, subjectivity, and variability that exists between embryologists of different skill levels and between grading systems of different performance. Specifically, embryologists often disagree, even among themselves or with each other, on which embryos have the best viability for transfer, even after spending considerable time visually evaluating them. Furthermore, it remains unclear which of the embryonic characteristics associated with a particular grading system ultimately predict the success rate of each embryo.

[0039] Other conventional techniques include automated techniques for selecting high-quality eggs. These techniques may be based on specific characteristics of fertilized eggs. Typically, these specific characteristics of fertilized eggs are obtained by analyzing videos capturing changes in the egg during growth and development (e.g., using microscopy and computer vision techniques). However, selecting eggs based on these characteristics (e.g., characteristics derived from observing the growth and development of the egg) may not always meet the demand.

[0040] On the other hand, the disclosed technology enables the analysis of egg viability at an earlier stage (for example, when the egg is still unfertilized, rather than when it has been fertilized). Therefore, the disclosed technology can eliminate the extra complexity associated with fertilizing eggs that may not later implant. By utilizing machine learning models for selecting oocytes based on features extracted from oocytes, the disclosed technology can provide a more objective and quantitative analysis of oocyte viability.

[0041] Furthermore, the disclosed technology utilizes machine learning techniques to analyze both the morphological and mechanical characteristics of oocytes. Morphological characteristics may include geometric information related to oocytes, such as the size or length of the zona pellucida, cytoplasm, polar body, and periuterine space, and the degree to which the oocyte is attracted to a pressure-applying tool (e.g., suction depth), as described below. Mechanical characteristics may include parameters determined or derived based on the deformation characteristics of the oocyte, as described below. For example, mechanical characteristics may include morphogenic dynamic parameters described in at least a specific model (e.g., the Zener model) or equivalent.

[0042] These machine learning techniques may utilize features to indicate one or more metrics that represent the quality of oocytes. Examples of metrics may include values ​​indicating the likelihood or probability of blastocyst formation. Examples of metrics may include indicators of oocyte "good" quality, the likelihood or probability of aneuploidy, and implantation rates.

[0043] More specifically, the system may acquire image sequences depicting oocytes deformed by mechanical stimulation. An example of stimulation may include the oocyte being aspirated into a part of a pressure tool (e.g., a pipette) that applies pressure to the oocyte. The system may process the image sequences using example computer vision techniques to derive the morphological and mechanical features described above. In some embodiments, normalization techniques may be used to accommodate different resolutions of the hardware (e.g., microscope camera) used to capture the images. Normalization techniques may be integrated as part of image processing techniques performed on the captured image sequences related to the oocytes. Normalization techniques are described in more detail below.

[0044] The extracted morphological and mechanical features derived from oocytes may then be used to train or perform inference using a machine learning model. The training process may include training a machine learning model using a subset of the morphological and mechanical features. The trained machine learning model may then be used to generate one or more metrics indicating oocyte quality, the particular metrics of which may indicate blastocyst formation from the oocyte.

[0045] Furthermore, and optionally, the metrics may be presented to the user or expert (e.g., an embryologist) via an interactive user interface. This allows for more efficient further analysis or evaluation of oocyte viability. Thus, based on embodiments of this disclosure, a more objective, automated, and time-efficient assessment of oocyte viability can be achieved.

[0046] As an example of the accuracy associated with the disclosed technology, a total of 185 oocytes were evaluated to assess blastocyst viability. The 185 samples were divided into 80% for training a machine learning model and 20% for testing the machine learning model. The machine learning used for viability assessment was a Support Vector Machine (SVM). The machine learning model was trained to predict whether a particular oocyte in the sample had a blastocyst or not. The prediction results were statistically analyzed and showed a precision of 73%, sensitivity of 85%, specificity of 59%, positive predictive value (PPV) of 77%, and negative predictive value (NPV) of 71%. Compared to the statistics of 45% precision, 59% sensitivity, 33% specificity, 43% PPV, and 49% NPV compiled based on predictions made by embryologists, the system and method of this disclosure achieve statistically better performance.

[0047] The aforementioned aspects of this disclosure and many associated advantages will become more readily apparent as they are better understood by referring to the following description in conjunction with the accompanying drawings.

[0048] Example of a block diagram

[0049] Figure 1 is a block diagram of an example oocyte analysis system 100 including a feature preprocessing engine 120 and a machine learning model 130 according to some embodiments of the present disclosure. As shown, the feature preprocessing engine 120 may receive an image sequence 102 of oocytes (mammalian oocytes such as human oocytes), pressure values ​​106, and clinical information 108 as input. The feature preprocessing engine 120 may then generate features 122 related to the oocytes as output. The features 122 can be input to a machine learning model 130, which outputs oocyte quality information 132. The oocyte quality information 132 can be presented to the user for further analysis.

[0050] The image sequence 102 may include multiple images (e.g., images 104A-104N), each depicting an oocyte 110 together with a pressure tool 112. More specifically, images 104A-104N may form a video depicting the process of aspirating the oocyte 110 into the pressure tool 112. Image 104A may represent, for example, the first frame of the video, and image 104N may represent, for example, the last frame of the video. Image 104A depicts the state in which the oocyte 110 has not yet been aspirated into the pressure tool 112, while image 104N depicts the state in which at least a portion of the oocyte 110 has been aspirated into the pressure tool 112. In some embodiments, the image sequence 102 may have frame rates of 10Hz, 20Hz, 70Hz, or 3000Hz, and a total video length of 1 second, 2 seconds, or 10 seconds.

[0051] Furthermore, in some embodiments, the pressure tool 112 is a pipette having a diameter between specific thresholds (e.g., between 10 micrometers (μm), 20 μm, 40 μm, 60 μm, 70 μm, 100 μm, etc.). The pipette may apply negative pressure to the oocytes 110 (e.g., the pressure inside the pipette is lower than the pressure outside the pipette) to draw the oocytes 110 into the pipette without damaging the oocytes 110. The pressure in the examples may be between -0.01 and -0.5 psi. For example, the pipette may be in contact with the oocytes or in other ways. Image sequence 102 illustrates the drawing of oocytes 110 into the pressure tool 112, but in some embodiments, the pressure tool 112 may apply other forms of mechanical stimulation (e.g., positive pressure). In this way, different morphological responses of the oocytes 110 can be obtained. These morphological reactions can be used in the feature preprocessing engine 120 to analyze and / or extract different morphological features.

[0052] In Figure 1, the feature preprocessing engine 120 is shown to output a feature 122 associated with the oocyte 110 using an image sequence 102, pressure values ​​106, and clinical information 108. In some embodiments, all of this information may be used to determine the feature 122. In some embodiments, a subset of the information may be used.

[0053] Regarding the pressure value 106, the pressure value 106 may include multiple numerical values ​​indicating the magnitude of the force applied to the oocyte 110 during the timeframe or time period of images 104A to 104N. For example, the pressure value 106 may indicate that a first pressure (e.g., -0.3 psi) is applied to the oocyte 110 at the moment image 104A is taken (e.g., time or timestamp), and that the Nth pressure is applied to the oocyte 110 at the moment image 104N is taken. The pressure values ​​applied to the oocyte 110 may be the same or different in different images within the image sequence 102.

[0054] In some embodiments, the pressure applied to the oocyte 110 may increase over time, while in other embodiments, the applied pressure may decrease over time. Furthermore, the pressure value 106 may include a force applied to the oocyte 110, which can be generated by the pressure tool 112 and calculated based on the pressure applied to the oocyte 110. The force applied to the oocyte 110 can be used to derive the mechanical characteristics of the oocyte 110, which will be described in more detail later with respect to Figure 2B.

[0055] Clinical information 108 may include the age and body mass index (BMI) of the patient from whom the oocyte 110 originates. Furthermore, clinical information 108 may include information indicating whether the oocyte 110 has undergone cryopreservation (CP). Clinical information 108 may also indicate the number of available mature oocytes (MII) associated with the patient.

[0056] Based on at least a portion of the image sequence 102, pressure values ​​106, and clinical information 108, the feature preprocessing engine 120 may extract features 122 related to the oocyte 110. Features 122 may include morphological features (size or length of the zona pellucida, cytoplasm, polar body, or periuterine space) and mechanical features (e.g., elasticity and / or viscosity) of the oocyte 110. Certain features may be generated for each image in the image sequence 102, while other features may be determined based on all images or a subset of images. These features 122 are described in more detail below with reference to Figures 2A-2B.

[0057] Based on the features 122, the machine learning model 130 may generate oocyte quality information 132 for the oocytes 110. An example of the information 132 may include the likelihood of blastocyst formation. In some embodiments, the machine learning model 130 may be a support vector machine (SVM) trained to output the information 132. In other embodiments, the machine learning model 130 may be a deep learning model. For example, the deep learning model may include a recurrent neural network (RNN) trained to output the information 132. In this example, the RNN may be input with the features 122 as a sequence and output the information 132 for the sequence. The model may be a convolutional neural network or a fully-connected network. The machine learning model 130 may utilize all or a subset of the features 122 to determine the oocyte quality information 132. For example, the SVM may be trained to utilize a subset of the features 122.

[0058] In some embodiments, the oocyte quality information 132 may further indicate whether the oocyte 110 is capable of forming a “usable” blastocyst, where “usable” means that the blastocyst formed by the oocyte 110 is suitable for transplantation or implantation. Furthermore, or alternatively, the oocyte quality information 132 may also indicate whether the blastocyst formed by the oocyte 110 is “unusable” blastocyst, where “unusable” means that the blastocyst formed by the oocyte 110 is in such poor condition that it is not suitable for further in vitro fertilization treatment.

[0059] Block diagram - Morphological feature generation

[0060] Figure 2A illustrates the morphological features of an example oocyte generated by the feature preprocessing engine 120. The morphological features 210 may form part of the features 122 in Figure 1. The morphological features 210 may include the aspiration depth of the oocyte 110, the diameter of the pressure tool 112 (e.g., pipette), the size and / or shape of the cytoplasm and zona pellucida of the oocyte 110, etc. Optionally, these features may be determined for each input image.

[0061] The determination of the morphological features 210 of several examples of oocytes 110 can be illustrated by oocyte measurement 202N, which depicts an example measurement (e.g., bounding box A, distance B, distance C, distance D, distance E, bounding box F, and distance G). Oocyte measurement 202N corresponds to the measurement in image 104N, and oocyte measurement 202A corresponds to the measurement in image 104A.

[0062] For the illustrated oocyte measurement 202N, bounding box A may represent a region of interest (ROI) that is cropped for further image processing performed on the image sequence 102. Distance B may indicate the length and / or size of the upper zona pellucida of the oocyte 110. Distance C may indicate the length and / or size of the right side of the zona pellucida of the oocyte 110. Distance D may indicate the length and / or size of the lower side of the zona pellucida of the oocyte 110. Distance E may indicate the inner zona pellucida of the oocyte 110. Distance G may indicate the length of the inner diameter of the pressure tool 112. Bounding box F may represent a region of interest (ROI) that may be cropped to calculate the aspiration depth of the oocyte 110 into the pressure tool 112. The calculation of the aspiration depth is described in more detail below.

[0063] Different image processing and object recognition techniques may be used to obtain the morphological features 210 of the embodiment. In some embodiments, computer vision techniques may be used to derive the morphological features 210 of the embodiment. For example, edge detection techniques may be used to identify boundaries associated with different parts of an oocyte. In other embodiments, an image segmentation model based on a deep learning algorithm may be used to derive the morphological features 210 of the embodiment. For example, a segmentation mask may be generated for the features 210. As can be understood, the segmentation mask may be assigned a color or pixel value indicating whether a pixel forms part of a classified feature (e.g., the zona pellucida). The derivation of each of the morphological features of the embodiment will be discussed in more detail below.

[0064] A bounding box A can define an ROI cropped from any image in the image sequence 102 (e.g., image 104A or image 104N), and several morphological features 210, such as the inner boundary of the zona pellucida, the size of the pressure tool 112, and the aspiration depth of the oocyte 110, can be calculated within the ROI. In some embodiments, the bounding box A can be identified by the following exemplary technique. For example, the tip of the pressure tool 112 (e.g., a pipette) that contacts the oocyte 110 can be identified and / or positioned based on a pipette tip image which may be acquired in advance. Based on the position of the pipette tip, the bounding box A can be drawn to create an ROI around the pipette tip. Although the bounding box A is shown as a rectangle, other shapes can be used to define the ROI.

[0065] Distances B, C, and D define the length or thickness of the zona pellucida on the upper, right, and lower sides, respectively. By averaging the lengths of these distances, morphological features indicating the average size of the zona pellucida of oocyte 110 may be obtained. Furthermore, distance E may be defined based on the position of the pipette tip obtained above when defining the boundary box A and the inner boundary of the zona pellucida of oocyte 110.

[0066] For example, distance E is useful for deriving the horizontal length of the cytoplasm of oocyte 110 (e.g., along the directions of arrows C and E). More specifically, the horizontal length of the cytoplasm of oocyte 110 can be derived by subtracting the sum of the lengths of arrows E and C from the horizontal length of oocyte 110 (e.g., defined by the right end of arrow C and the left end of arrow E). The horizontal length of the cytoplasm of oocyte 110 can also be used as part of the morphological feature 210 of the example. Similarly, the vertical length of the cytoplasm of oocyte 110 (e.g., along the directions of arrows B and D) can be derived by subtracting the sum of the lengths of arrows B and D from the vertical length of oocyte 110 (e.g., defined by the upper end of arrow B and the lower end of arrow D). The vertical length of the cytoplasm of oocyte 110 can also be used as part of feature 122.

[0067] The bounding box F can be used as an ROI for calculating the aspiration depth of the oocyte 110 (e.g., the length the oocyte 110 is drawn into the pressure tool 112 compared to the position of the oocyte 110 immediately before aspiration). The aspiration depth is useful for deriving mechanical features associated with the oocyte 110, which will be discussed in more detail below with respect to Figure 2B. In some embodiments, the aspiration depth can be derived by determining the deepest horizontal position (e.g., the end of the aspiration depth) where the oocyte 110 is drawn into the pressure tool 112. The distance between the end of the aspiration depth and the inner boundary of the zona pellucida of the oocyte 110 may then be calculated. The following are exemplary techniques for determining the aspiration depth.

[0068] The ROI (e.g., bounding box F) may be defined (e.g., identified) based on the position of the tip of the pressure tool 112 in contact with the oocyte 110. As illustrated in oocyte measurement 202N, the bounding box F has a rectangular shape; however, in other embodiments, the bounding box F may have a different shape (e.g., square or irregular). In some embodiments, the bounding box F extends to the leftmost portion of the image at a specific threshold distance; that is, the distance may be set to ensure that it includes the end or tip of the oocyte being aspirated into the tool 112.

[0069] A pixel curve may be derived by summing the pixels that are vertically aligned with each other within the bounding box F. More specifically, assuming that the bounding box F spans N pixels horizontally and M pixels vertically, the pixel intensities of pixels having the same horizontal position within the bounding box F can be summed. For example, M pixels at a particular horizontal position may be summed. Thus, N may be a positive integer depending on the resolution of the image sequence 102 and the horizontal length of the bounding box F. For example, if there are 50 horizontal positions, the value may be reached by summing the pixels that extend vertically within the box F for each position.

[0070] These values ​​may be used to generate a pixel curve having a sum on the first axis (e.g., the y-axis) and a number of horizontal positions on the second axis (e.g., the x-axis). The pixel curve may be smoothed using signal processing techniques such as the sliding window moving average method.

[0071] Subsequently, the first-order derivative may be determined on the smoothed pixel curve. This derivative may represent the summed intensity in the vertical direction. Next, the minimum value of the derivative of the pixel curve may be found, which indicates the endpoint of the aspiration depth of the oocyte 110 for a particular image (e.g., image 104A).

[0072] The exemplary techniques described above may be performed on other images in image sequence 102, thereby obtaining the end of the aspiration depth for each of images 104A to 104N. In some embodiments, the aspiration depth can be calculated based on the end of the aspiration depth and the inner boundary of the zona pellucida of the oocyte 110. By calculating the aspiration depth for each of images 104A to 104N in image sequence 102, the aspiration depth of the oocyte 110 over time can be plotted as a chart where the first axis represents time and the second axis represents the aspiration depth of the oocyte 110. In some embodiments, the aspiration depth can be further normalized based on the size of the zona pellucida of the oocyte 110. For example, the aspiration depth can be normalized by subtracting the average of distances B, C, and D from the calculated aspiration depth.

[0073] While the above describes an example of determining the suction depth, other techniques may be used and are included in this disclosure, as can be understood. For example, deep learning techniques may be used to determine the boundaries or tips of oocytes within the tool. Alternatively, edge detection techniques (e.g., edge detection kernels) may be used to identify the edges of oocytes within the tool.

[0074] Furthermore, and optionally, the feature preprocessing engine 120 may not need to analyze all images in the image sequence 102. In some embodiments, only a portion of the image sequence 102 is analyzed by selecting a starting frame and performing the analysis on the starting frame and frames after the starting frame. The starting frame may be a frame that captures an image of the oocyte 110 immediately before it is aspirated into the pressure tool 112. For example, the system may identify frames related to a time or timestamp earlier than the timestamp related to the application of negative pressure. In another embodiment, the system may analyze image frames and identify frames earlier than frames with an aspiration depth greater than a threshold (e.g., zero, a small value). Advantageously, the analysis is more time-efficient and consumes less computational power.

[0075] Furthermore, and optionally, the preprocessing engine 120 can reduce (e.g., shrink) the number of frames to be analyzed by removing consecutive frames in which the oocyte 110 remains stationary or motionless. For example, the preprocessing engine 120 may decide that only 0.5 seconds of the video depicted by image sequence 102 show the movement of the oocyte 110, and the remaining 1.5 seconds of the video of image sequence 102 show the oocyte 110 remaining motionless or relatively static. The preprocessing engine 120 may then extract 0.5 seconds of the video for further analysis. Advantageously, the analysis time can be reduced by shrinking the video length for analysis.

[0076] The feature preprocessing engine 120 can further determine the inner diameter of the pressure tool 112, which is illustrated by distance G in the oocyte measurement 202N. In some embodiments, the length of the inner diameter of the pressure tool 112 can be obtained by calculating the number of pixels aligned vertically along the inner diameter of the pressure tool 112. The inner diameter of the pressure tool 112 can be used as a normalization factor to improve the interoperability of the oocyte analysis system 100 between different video capture platforms which may have different hardware specifications (e.g., resolution of captured images).

[0077] Normalization may also relate to the z-height of the oocyte within the object containing the oocyte while the image is being taken. That is, the oocyte is placed in water or another liquid, and its height may vary (e.g., closer to the bottom or top of the water). Therefore, this height adjustment may alter the dimensions of the oocyte compared to another oocyte, or compared to the same oocyte in different images within sequence 102. For example, the oocyte may have slight variations in height between image sequences 102.

[0078] More specifically, the temporal change in the oocyte aspiration depth may be normalized by the ratio between the inner diameter of the pressure tool 112 and the pixel length of a particular image sequence. Using computer vision techniques, the inner diameter of the pressure tool 112 can be identified, and then the length of the inner diameter can be calculated by counting the number of pixels covered by the inner diameter along the axis of the pressure tool 112. For example, the inner diameter of the pressure tool 112 may extend over 1,000 pixels, and then the aspiration depth can be divided by 1,000 for normalization. Normalizing the oocyte aspiration depth by the inner diameter of the pressure tool 112 allows different frames of oocytes acquired under different imaging settings to be placed on the same basis for the purpose of calculating morphological features associated with the oocytes. In this way, the interoperability and reproducibility of the currently disclosed system and method can be improved.

[0079] Additional morphological features related to the oocyte 110 can also be obtained by the feature preprocessing engine 120. For example, features may include the size and / or length of the polar body, cytoplasm, and / or perivitelline space (e.g., Perivitelline Space (PVS), the space between the zona pellucida and the cytoplasm) of the oocyte 110. Both image segmentation models based on deep learning algorithms and computer vision techniques can be employed to determine these additional morphological features. Advantageously, the available additional morphological features may be potentially useful for training and / or testing the machine learning model 130 to improve the accuracy, sensitivity, specificity, NPV, and PPV of the results produced by the machine learning model 130.

[0080] Block Diagram - Mechanical Feature Generation

[0081] Figure 2B illustrates the mechanical features of oocytes generated by the feature preprocessing engine 120 in Figure 1. As shown in Figure 2B, the feature preprocessing engine 120 receives the image sequence 102 and pressure values ​​106 as input and generates example mechanical features 254 based on the example elastic model 252. The example mechanical features 254 may show deformation or movement of the oocytes 110 during the image sequence 102.

[0082] The pressure value 106 may include the force applied to the oocyte 110 (e.g., via the pressure tool 112 described above). Specifically, the force applied to the oocyte 110 may be calculated based on the following equation (A). Therefore, the force may be based on the pressure value and geometric information related to the pipette.

[0083]

number

[0084] The feature pretreatment engine 120 may use the force described above in combination with the suction depth shown in Figure 2A to determine feature 254.

[0085] In some embodiments, features exhibiting the viscoelastic behavior of oocytes 110 can be obtained using linear elastic models, modified linear elastic models, and standard linear solid models. Examples of exemplary models may include Zener models, modified Zener models, and the like. For example, the mechanical characteristics of oocytes 110 may be obtained by fitting the suction depth of oocytes 110 over time using equation (B) below. For example, the models described above (e.g., a spring-damper model) may be used with two springs and two dampers (e.g., dashpots) to model the mechanical properties, as shown in Figure 2B. The models may be fitted by minimizing the sum of squared errors between the measured suction depth and the modeled suction depth using exemplary techniques (e.g., the Broyden-Fletcher-Goldfarb-Shanno algorithm, as an example). Equation (B) includes the parameters k0, k1, τ, η0, and η1 of the exemplary elastic model 252. These parameters may be determined by fitting the suction depth. In equation (B), F0 can represent the force applied to the oocyte 110. For example, F0 can represent the force associated with a particular suction depth (hereinafter referred to as "depth"). In this example, different suction depths associated with different times (e.g., images) may be used along with their respective associated forces to determine the parameters. Parameters k0 and k1 model the solid-like behavior of the oocyte 110, while η0 and η1 model the liquid-like behavior of the oocyte 110. Thus, the example elastic model 252 can take into account both types of behavior associated with the oocyte 110.

[0086]

number

[0087] Parameters k0 and k1 can describe the “instantaneous elongation” that the oocyte 110 experiences when a force is applied to it. This instantaneous elongation corresponds to or is proportional to 1 / (k0+k1) and can be seen as a measure of “slack” in the elastic elements of the oocyte 110, or as the amount of force that can be applied to the oocyte 110 before significant resistance is shown. Parameter k1 can be seen as a general measure of stiffness and may represent how tightly proteins in the cytoplasm or zona pellucida are bound. Parameter η1 can be seen as a measure of how continuously the zona pellucida deforms in response to the applied force calculated according to equation (A). As in the linear elastic solid model, after the spring elements have fully elongated, η1 is responsible for the molecular-level shape change that continues to elongate the oocyte 110. Parameter τ represents how quickly (e.g., velocity) the oocyte 110 deforms (e.g., enters the pressure tool 112) after the initial instantaneous elongation. η0 can be seen as a measure of cytoplasmic viscosity or fluid viscosity in the space between the zona pellucida and the inner cells of the oocyte 110 (e.g., PVS).

[0088] In addition to the use of negative pressure, other non-limiting examples of causing movement or deformation of the oocyte 110 may be used and are included within the scope of this disclosure. For example, positive pressure on the oocyte 110 may be used to expel the oocyte 110 or to deform, retain, or perturb different parts of the oocyte 110. Illustrative parts may include the zona pellucida, cytoplasm, or parts of the oocyte 110 near its surface. Other forms of force (e.g., light pressure) may also be applied to the oocyte 110.

[0089] In some embodiments, the pressure or force applied to the oocyte 110 is appropriately adjusted by the pressure tool 112 to avoid undesirable effects on the oocyte 110. For example, if the pressure applied to the oocyte 110 is too high, it may damage the structure of the oocyte 110 and reduce its viability. In some embodiments, the pressure applied to the oocyte 110 is between -0.01 and -0.5 psi (or 0.01 to 0.5 psi if positive pressure is applied). In some embodiments, the pressure applied to the oocyte 110 by the pressure tool 112 is adjusted based on the number of days elapsed since fertilization (e.g., day 1, day 2, or day 3). In some embodiments, the inner diameter of the pressure tool 112 is between 40 and 70 μm, and the applied pressure may be adjusted based on the inner diameter of the pressure tool 112 to generate an appropriate level of force applied to the oocyte 110.

[0090] Example User Interface

[0091] Referring to Figures 3A-3C, an example application of the oocyte analysis system 100 shown in Figure 1 will be explained.

[0092] Figure 3A illustrates an example user interface 300A that includes the processing results of the example oocyte analysis system 100 shown in Figure 1. As shown in Figure 3A, the quality information 132 of the oocytes generated by the oocyte analysis system 100 may be presented via the user interface 300A. The user interface 300A can be presented as a web page accessible by a browser or as a user interface for an application.

[0093] As illustrated, the user interface 300A receives oocyte quality information 132 from the system 100 described in this disclosure. The information 132 may reflect the quality metrics of each of a number of oocytes extracted from a patient. In the illustrated example, the user interface 300A presents summary information based on the analysis of five oocytes. The user interface 300A indicates that "oocyte 3" has the highest likelihood of blastocyst formation. This indication may be based on blastocyst formation determined by a machine learning model (e.g., model 130).

[0094] The user interface 300A may respond to user input related to the display of detailed analysis. In some embodiments, the detailed analysis may include all or a subset of the features described above. The detailed analysis may include graphical representations, such as images from image sequences related to oocytes 3. In some embodiments, image sequences may be presented as a video or animation in the user interface 300A.

[0095] In some embodiments, the oocyte quality information 132 can be linked to or compared with evaluation results obtained from preimplantation genetic testing (PGT) or implantation performed on the same oocyte sample. For example, for oocytes indicated by the oocyte quality information 132 as capable of blastocyst formation (e.g., "good" oocytes), preimplantation genetic testing for aneuploidy (PGT-A) can be performed to derive the euploidy rate of oocytes selected by the oocyte analysis system 100. As another example, for oocytes indicated by the oocyte quality information 132 as incapable of blastocyst formation (e.g., "poor" oocytes), preimplantation genetic testing for aneuploidy (PGT-A) can be similarly performed to derive the aneuploidy rate of oocytes selected by the oocyte analysis system 100.

[0096] In another example, oocytes indicated by oocyte quality information 132 as capable of blastocyst formation (e.g., "good" oocytes with a likelihood exceeding a threshold) can be further evaluated after transplantation. For example, this may be used to measure the predictive ability of the oocyte analysis system 100 to predict oocyte viability. The probability of embryo implantation can be obtained based on the "good" oocytes indicated by the oocyte analysis system 100 to evaluate the predictiveness of the oocyte analysis system 100. Advantageously, by linking different evaluation stages for embryo quality, the oocyte analysis system 100 can be improved based on evaluation results from PGT-A and implantation. For example, a low probability of embryo implantation may suggest that the parameters of the machine learning model 130 need to be adjusted by using different subsets of features 122 to train the machine learning model 130.

[0097] Figure 3B illustrates an embodiment user interface 300B that enables the embodiment oocyte analysis system 100 of Figure 1 to receive user input such as patient information. As shown in Figure 3B, the user interface 300B can enable the user to add, edit, and save patient information to the oocyte analysis system 100. As shown, the patient information displayed on the user interface 300B is assigned a patient ID (e.g., 1234567890). The patient ID can be used to associate a specific oocyte with a specific patient. Patient information such as height, weight, and the date the patient information was created is displayed below the patient ID. Additional patient information can be added by pressing the "+ Add" button at the top center of the user interface 300B.

[0098] Although not shown in Figure 3B, the user interface 300B can facilitate other interactions with the embodiment oocyte analysis system 100. For example, other patient information such as BMI and / or age described above can also be edited and associated with a specific patient by operating the user interface 300B. The added patient information may then be used by the oocyte analysis system 100 for quality analysis of one or more specific oocytes. Furthermore, the user interface 300B can facilitate the retrieval of patient information for a specific patient based on other recorded patient information such as patient ID or MII. In addition, the user interface 300B can enable the user to edit information related to a specific oocyte, such as the date the oocyte was received and information about the oocyte donor (e.g., from whom the specific oocyte was taken), such as the donor's date of birth, donor's height and / or donor's weight.

[0099] The user interface 300B may also receive user input that causes the Example Oocyte Analysis System 100 to analyze a specific oocyte and generate oocyte quality information 132 for that specific oocyte. Furthermore, the user interface 300B may warn the user to check or readjust the position of the camera used to capture an image sequence 102 for quality analysis of one or more oocytes. In some embodiments, the oocyte quality information 132 generated by the Example Oocyte Analysis System 100 may be presented to the user as described below.

[0100] Figure 3C illustrates an example user interface 300C that presents the processing results of the example oocyte analysis system 100 shown in Figure 1. As shown in Figure 3C, the user interface 300C may present different oocyte quality information 132 based on the analysis performed by the oocyte analysis system 100. As shown in Figure 3C, the oocyte quality information 132 is presented as a score indicating the quality of the oocyte. In addition, a text description summarizing the quality may be presented. For example, the text may be based on quality metrics, selected from pre-stored text expressions or words (e.g., normal, good, etc.), or determined using a language model (e.g., a large-scale language model).

[0101] The left side shows the analysis results for an oocyte with a good score (e.g., 95), which may mean that there is a very high probability that a "usable" blastocyst will be formed from that oocyte. On the other hand, the right side shows the analysis results for an oocyte with a poor score (e.g., 14), which may mean that there is a very low probability that a "usable" blastocyst will be formed from that oocyte. The user interface 300C in the center shows the analysis results for an oocyte with a "normal" score (e.g., 60), which may mean that there is a higher probability than a lower probability that that oocyte will form a usable blastocyst.

[0102] After receiving the analysis results for a specific oocyte, the user interface 300C may allow the user to view the analysis results for other oocytes, or may prompt the user to analyze the quality of oocytes that have not yet been analyzed by the Example Oocyte Analysis System 100. Specifically, the user may view the analysis results for another oocyte by pressing the "Next Oocyte" button, or view the analysis results for previously analyzed oocytes by pressing the "Back" button.

[0103] Example Flowchart

[0104] Figure 4 is a flowchart of an example process 400 for determining metrics indicating oocyte quality. All or at least part of process 400 may be performed, for example, by the oocyte analysis system 100 shown in Figure 1. As shown above, process 400 may provide a way to determine the viability of oocytes 110 (e.g., whether blastocyst formation occurs) without relying on manual and subjective evaluation by an embryologist. Therefore, process 400 may be used to achieve a more objective, time-efficient, and automated acquisition of oocyte quality 110.

[0105] In block 402, the system acquires images that form an image sequence of oocytes. As described above, the images may be captured by a microscope camera and depict a series of events showing deformation or movement of oocytes resulting from force applied to them by a tool (e.g., a pipette).

[0106] In block 404, the system acquires a pressure value related to the oocyte being aspirated into the tool. For example, the pressure value may include the pressure or force applied to the oocyte 110 during the process of it being aspirated into the pressure tool. In some embodiments, the pressure value may be kept constant throughout the entire aspiration process. In some embodiments, the pressure value may change in a particular way (e.g., a low pressure is followed by a high pressure force).

[0107] In block 406, the system determines morphological features associated with the oocyte based on the acquired images that form an image sequence. As an example, morphological features associated with the oocyte include the aspiration depth of the oocyte, the size and / or length of the cytoplasm, the size and / or length of the zona pellucida, and the diameter of the tool used to aspirate the oocyte.

[0108] As illustrated in Figure 2A above, morphological features can be acquired using computer vision and / or machine learning techniques. In some embodiments, one or more neural networks may be trained to output segmentation masks specific to particular morphological features. Thus, size or length associated with a portion of an oocyte may be identified. In some embodiments, segmentation may involve two stages, namely two-dimensional (2D) modeling followed by three-dimensional (3D) modeling. In the 2D modeling stage, features associated with a 2D image are extracted. In the 3D modeling stage, the features extracted from the 2D image may then be concatenated to form time-series data, each of which contains features extracted from one 2D image within a 2D image sequence. A machine learning model (e.g., a deep learning model) may then classify the features generated by the 3D modeling stage, for example, as shown in block 410.

[0109] In block 408, the system determines mechanical characteristics related to the oocyte. For example, the system determines parameters that indicate the deformation or movement of oocyte 110 in the image sequence. In this example, the parameters may be related to the elastic model shown in Figure 2B.

[0110] In block 410, the system uses a machine learning model to determine metrics that indicate the quality of oocytes. The system provides features as input to the machine learning model, such as concatenated features determined for an image or a sequence of features determined for a corresponding image. In some embodiments, the machine learning model 130 may be a support vector machine. In some embodiments, the model may be a deep learning model (e.g., a neural network). In some embodiments, a subset of features may be provided. For example, one feature, two features, three features, ten features, etc., may be provided.

[0111] The metrics may contain information indicating at least oocyte blastocyst formation. Furthermore, one or more metrics may indicate oocyte-related aneuploidy and / or implantation.

[0112] In some embodiments, one or more metrics may indicate whether an oocyte 110 forms a “good” blastocyst, where a “good” blastocyst may mean that the associated Gardner embryo / blastocyst grading is greater than 3CC. Furthermore, optionally, in addition to using embodiment morphological and embodiment mechanical features to determine metrics indicating oocyte quality, machine learning models may further use patient clinical information to determine oocyte quality information. As mentioned in the description with respect to Figure 1, clinical information may include age, patient BMI, and / or other clinical information such as CP and MII related to the oocyte (e.g., developmental stage related to the oocyte).

[0113] After determining metrics indicating the quality of oocytes, process 400 may return to block 402 to determine quality information for another oocyte.

[0114] Figure 5 is a flowchart of an example process 500 for selecting a subset of multiple oocytes based on quality metrics. All or at least part of process 500 may be performed by, for example, the oocyte analysis system 100 shown in Figure 1. Process 500 may provide a way to determine the viability of multiple oocytes without relying on time-consuming subjective evaluations by embryologists. Therefore, process 500 can be used to obtain a more objective, time-efficient, and automated quality assessment of multiple oocytes.

[0115] In block 502, the system acquires image sequences and pressure values ​​related to a large number of oocytes. As described herein, the image sequences may depict oocytes deformed by pressure applied by a pressure tool or by pressure.

[0116] In block 504, the system acquires metrics indicating the quality of individual oocytes in a large number of oocytes. The determination of these metrics is described in more detail above and may indicate the potential for blastocyst formation for each oocyte.

[0117] In block 506, the system selects a subset of oocytes from a large number of oocytes. For example, the system may identify a top threshold number of oocytes based on their respective blastocyst formation potential. In another example, the system may aggregate or otherwise combine metrics from each oocyte. In this example, the system may select a top threshold number of oocytes based on the aggregated or combined metrics. As described above, the metrics may include blastocyst formation, as well as metrics related to successful outcomes in the later stages or chromosomal abnormalities (e.g., PGT-A, euploidy).

[0118] In block 508, the system outputs and / or presents information related to a selected subset of oocytes. This information may be presented through a graphical user interface (GUI), such as the user interface 300A shown in Figure 3A. The information may indicate whether a particular oocyte has the potential to form a blastocyst, as depicted in Figures 3A and 3C. Furthermore, other information related to the selected subset of oocytes can be presented to the user. For example, morphological characteristics, mechanical characteristics related to the oocytes, and / or clinical information of the patient from whom the oocytes were obtained can be presented. By performing process 500 using the oocyte analysis system 100, viability information related to a large number of oocytes can be obtained in a time-efficient manner.

[0119] Example System

[0120] Figure 6 depicts a general architecture of an embodiment system. The system may be used in some embodiments to perform the functions described herein. In some embodiments, the system may be an oocyte analysis system 100, which includes an arrangement of computer hardware and software configured to carry out aspects of this disclosure. The oocyte analysis system 100 may include more (or fewer) elements than those shown in Figure 6. However, not all of these elements need to be shown in order to provide an implementable disclosure.

[0121] As illustrated, the oocyte analysis system 100 includes a processor 602, a pressure tool 604 (e.g., pressure tool 112 in Figure 1), a network interface 606, an image sensor 608 (e.g., one or more microscope cameras used to capture the image sequence 102 in Figure 1), and a data store 610, all of which may communicate with each other via a communication bus 612. The pressure tool 112 may not be included in some embodiments, and the system 100 may represent a backend processing system. The network interface 606 may provide connectivity to one or more network or computing systems, which may enable the oocyte analysis system 100 to receive and transmit information and instructions from other computing systems, interfaces (e.g., user interface 300A in Figure 3A) or services. In some embodiments, the oocyte analysis system 100 may be configured to handle requests from other devices or modules, such as requests to analyze the quality of oocytes. The data store 610 may, for example, be any non-temporary computer-readable data store, which in various embodiments may store any or all of the elements depicted in Figure 6 as to be loaded into memory 614.

[0122] The processor 602 may also communicate with the memory 614. The memory 614 may contain computer program instructions (which in some embodiments may be grouped as modules or components) that the processor 602 may execute to implement one or more embodiments. The memory 614 generally includes RAM, ROM, and / or other persistent, auxiliary, or non-temporary computer-readable media. The memory 614 may store an operating system 616 that provides computer program instructions for use by the processor 602 in the general management and operation of the oocyte analysis system 100. The memory 614 may further store specific computer-executable instructions and other information (which may be referred to herein as “modules” or “engines”) for implementing aspects of the disclosure. For example, the memory 614 may include a feature preprocessing engine 632 and a machine learning model 634 that may implement aspects of the disclosure as described above. The memory 614 may further store a user interface module 618 that may enable the presentation of information to the user interface 300A in Figure 3A. In addition, memory 614 can store function libraries 620 (for example, to store parameters for different types of machine learning models) and features 630 that may be extracted by the feature preprocessing engine 632. All modules or elements loaded into memory 614 may also be stored in datastore 610 when various operations are performed.

[0123] Many of the components shown in Figure 6 are optional, and it will be recognized that embodiments of the oocyte analysis system 100 may or may not combine components. Furthermore, the components do not need to be separate or discrete. Components may also be reorganized. In some embodiments, components illustrated as part of the oocyte analysis system 100 may be included in additional or alternative other computing devices, so that some aspects of this disclosure are performed by the oocyte analysis system 100, while others are performed by other computing devices.

[0124] All methods and tasks described herein may be performed by a computer system and may be fully automated. In some cases, the computer system may include multiple separate computers or computing devices (e.g., physical servers, workstations, storage arrays, cloud computing resources, etc.) that communicate and interact over a network to perform the functions described. Each of such computing devices typically includes a processor (or more processors) that executes program instructions or modules stored in memory or other non-temporary computer-readable storage media or devices (e.g., solid-state storage devices, disk drives, etc.). The various functions disclosed herein may be embodied in such program instructions or implemented in application-specific circuits (e.g., ASICs or FPGAs) of the computer system. If the computer system includes multiple computing devices, these devices may, but do not necessarily, be located in the same place. The results of the disclosed methods and tasks may be permanently stored by converting physical storage devices, such as solid-state memory chips or magnetic disks, into different states. In some embodiments, the computer system may be a cloud-based computing system in which its processing resources are shared by multiple separate entities or other users.

[0125] The processes described in this disclosure or illustrated in the figures may be initiated on demand when started by a user or system administrator, or in response to any other event, such as in response to an event, on a predetermined schedule or a dynamically determined schedule. When such a process is initiated, a set of executable program instructions stored on one or more non-temporary computer-readable media (e.g., hard drives, flash memory, removable media) may be loaded into the memory (e.g., RAM) of a server or other computing device. The executable instructions may then be executed by the hardware-based computer processor of the computing device. In some embodiments, such a process or part thereof may be executed in series or in parallel on multiple computing devices and / or multiple processors.

[0126] Depending on the embodiment, any particular operation, event, or function of any process or algorithm described herein may be executed in a different sequence, and may be added, integrated, or omitted entirely (for example, not all described operations or events are necessary for the implementation of the algorithm). Furthermore, in certain embodiments, operations or events may be executed not sequentially, but concurrently, for example, through multithreading, interrupt handling, or multiple processors or processor cores or other parallel architectures.

[0127] Various exemplary logic blocks, modules, routines, and algorithmic steps described in relation to embodiments disclosed herein may be implemented as electronic hardware (e.g., ASIC or FPGA devices), computer software running on computer hardware, or a combination of both. Furthermore, various exemplary logic blocks and modules described in relation to embodiments disclosed herein may be implemented or executed by machines such as processor devices, digital signal processors ("DSPs"), application-specific integrated circuits ("ASICs"), field-programmable gate arrays ("FPGAs") or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. The processor device may be a microprocessor, but alternatively, the processor device may be a controller, microcontroller, or state machine, a combination thereof, or similar. The processor device may include electrical circuits configured to process computer-executable instructions. In another embodiment, the processor device includes an FPGA or other programmable device that performs logical operations without processing computer-executable instructions. Processor devices may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration. Although this specification primarily describes digital technologies, processor devices may also include primarily analog components. For example, some or all of the rendering techniques described herein may be implemented in analog circuits or mixed analog and digital circuits.The computing environment may include any type of computer system, including but not limited to microprocessor-based computer systems, mainframe computers, digital signal processors, portable computing devices, device controllers, or in-device computing engines.

[0128] Elements of methods, processes, routines, or algorithms described in connection with embodiments disclosed herein may be embodied in hardware directly, in software modules executed by a processor device, or a combination of both. Software modules may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of non-temporary computer-readable storage medium. As an example, a storage medium may be coupled to a processor device so that the processor device can read information from and write information to the storage medium. Alternatively, the storage medium may be integrated with the processor device. The processor device and storage medium may reside within an ASIC. The ASIC may reside within a user terminal. Alternatively, the processor device and storage medium may exist as separate components within a user terminal.

[0129] The conditional language used herein, such as “possible,” “may,” “may,” “may,” and “for example,” is generally intended to convey that a particular embodiment includes a particular feature, element, or step, but other embodiments do not, unless otherwise stated or understood in the context in which it is used. Therefore, such conditional language is not generally intended to imply that a feature, element, or step is somehow required for one or more embodiments, or that one or more embodiments necessarily include logic for determining whether these features, elements, or steps are included or performed in any particular embodiment, with or without other inputs or prompts. Terms such as “equip,” “include,” and “have” are synonyms and are used in a comprehensive, open-ended manner, not excluding additional elements, features, actions, behaviors, etc. Furthermore, the term “or” is used in its comprehensive (not exclusive) sense, so when used to connect a list of elements, for example, “or” means one, some, or all of the elements in the list.

[0130] Disjunctive language, such as the phrase "at least one of X, Y, or Z," indicates, as is understood in the context in which it is commonly used, that an item, term, etc., may be X, Y, or Z, or any combination thereof (e.g., X, Y, or Z), unless otherwise specified. Therefore, such disjunctive language is not generally intended, nor should it be interpreted, to imply that a particular embodiment requires the presence of at least one X, at least one Y, and at least one Z, respectively.

[0131] While the above detailed description has shown, described, and pointed out novel features applicable to various embodiments, it can be understood that various omissions, substitutions, and modifications in the form and details of the illustrated apparatus or algorithm can be made without departing from the spirit of this disclosure. As is recognizable, certain embodiments described herein can be embodied in forms that do not provide all the features and benefits described herein, and some features can be used or implemented separately from others. All modifications that fall within the meaning and equivalence of the claims are encompassed within that scope.

Claims

1. A method carried out by one or more computer systems, The method involves obtaining multiple images that form an image sequence related to a time period, The aforementioned image sequence depicts oocytes and a portion of the tool used to apply pressure to the oocytes. Each image is associated with the individual pressure value applied to the oocyte at each point in time of image capture, and is obtained as follows: Determining characteristics related to the oocyte based on the aforementioned image and the aforementioned pressure value, The characteristics to be determined include morphological characteristics indicating the measurement of the oocyte during the time period and mechanical characteristics indicating the deformation of the oocyte during the time period. This includes determining one or more metrics indicating oocyte quality via a machine learning model based on an input that includes at least a subset of the aforementioned features, Certain metrics indicate blastocyst formation. A method comprising one or more metrics configured to be presented via an interactive user interface.

2. The method according to claim 1, wherein the oocyte depicted in the image is of a mammal.

3. The method according to claim 1, wherein the image depicts the oocyte being aspirated into the part of the tool.

4. The aforementioned mechanical characteristics indicate the elasticity of the oocyte. The method according to claim 1, wherein the mechanical characteristics are determined based on the depth of aspiration of the oocytes to the portion of the tool.

5. The method according to claim 4, wherein the mechanical features indicate the viscoelastic behavior of the oocyte.

6. The method according to claim 1, wherein the morphological features indicate at least the depth of aspiration of the oocytes to the portion of the tool relative to the corresponding image.

7. The method according to claim 1, wherein one of the morphological features is derived by normalizing the aspiration depth of the oocyte to the portion of the tool in the corresponding image by the inner diameter of the tool.

8. The method according to claim 1, wherein the morphological features are determined by a first measurement relating to the cytoplasm of the oocyte and a second measurement relating to the zona pellucida of the oocyte.

9. The aforementioned part of the tool is a pipette, The method according to claim 1, wherein the pipette is configured to contact the oocyte and apply pressure during the aforementioned time period.

10. The first image relating to the start of the aforementioned time period depicts the oocyte before pressure is applied, The method according to claim 1, wherein in the image sequence, a second image following the first image depicts the oocyte to which pressure has been applied.

11. The method according to claim 1, wherein the pressure value is selected from the pressure range applied during the time period.

12. The method according to claim 1, wherein the machine learning model is a support vector machine.

13. The method according to claim 1, wherein the machine learning model is a neural network.

14. Based on the aforementioned pressure value and the geometric information related to the aforementioned part of the tool, the measured pressure applied to the oocyte is calculated. The method further includes determining the mechanical features by fitting the suction depth curve using a linear elastic model, The method according to claim 1, wherein the fitting of the aspiration depth curve is performed based on measurements of the aspiration depth and pressure of the oocytes to the portion of the tool relative to the corresponding image.

15. The input further includes clinical information of the person associated with the oocyte, The method according to claim 1, wherein the clinical information includes one or more of age, body mass index, developmental stage related to the oocyte, and cryopreservation information.

16. A quality analysis system for oocytes, comprising one or more processors and a non-temporary computer storage medium that stores instructions that cause the one or more processors to perform the following processes when executed by the one or more processors, The aforementioned process is, The method involves obtaining multiple images that form an image sequence related to a time period, The aforementioned image sequence depicts oocytes and a portion of the tool used to apply pressure to the oocytes. Each image is associated with the individual pressure value applied to the oocyte at each point in time of image capture, and is obtained as follows: Determining characteristics related to the oocyte based on the aforementioned image and the aforementioned pressure value, The characteristics to be determined include morphological characteristics indicating the measurement of the oocyte during the time period and mechanical characteristics indicating the deformation of the oocyte during the time period. This includes determining one or more metrics indicating oocyte quality via a machine learning model based on an input that includes at least a subset of the aforementioned features, Certain metrics indicate blastocyst formation. A quality analysis system for oocytes, configured such that one or more of the aforementioned metrics are presented via an interactive user interface.

17. The oocyte quality analysis system according to claim 16, wherein the image depicts the oocyte being aspirated into the part of the tool.

18. The aforementioned mechanical characteristics indicate the elasticity of the oocyte. The oocyte quality analysis system according to claim 16, wherein the mechanical characteristics are determined based on the aspiration depth of the oocytes to the portion of the tool.

19. The oocyte quality analysis system according to claim 18, wherein the mechanical characteristics indicate the viscoelastic behavior of the oocyte.

20. The oocyte quality analysis system according to claim 16, wherein the morphological features indicate at least the depth of aspiration of the oocyte to the portion of the tool relative to the corresponding image.

21. The oocyte quality analysis system according to claim 16, wherein the morphological characteristics include a first measurement relating to the cytoplasm of the oocyte and a second measurement relating to the zona pellucida of the oocyte.

22. The oocyte quality analysis system according to claim 16, wherein the pressure value is selected from the pressure range applied during the aforementioned time period.

23. The aforementioned instruction is, Based on the aforementioned pressure value and the geometric information related to the aforementioned part of the tool, the measured pressure applied to the oocyte is calculated. The method further includes determining the mechanical features by fitting the suction depth curve using a linear elastic model, The oocyte quality analysis system according to claim 16, wherein the fitting of the aspiration depth curve is performed based on measurements of the aspiration depth and pressure of the oocytes to the portion of the tool relative to the corresponding image.

24. The input further includes clinical information of the person associated with the oocyte, The oocyte quality analysis system according to claim 16, wherein the clinical information includes one or more of age, body mass index, developmental stage related to the oocyte, and cryopreservation information.

25. A method carried out by one or more computer systems, The aforementioned method, This involves obtaining multiple image sequences that depict each oocyte, Each image sequence depicts individual oocytes and parts of the tool used to apply pressure to the oocytes. The pressure values ​​associated with the images included in the aforementioned image sequence are obtained, and the acquisition of these values ​​is performed. Determining one or more metrics indicating the quality of each of the oocytes through a machine learning model, Certain metrics indicate blastocyst formation. The input to the machine learning model for each oocyte is determined to include morphological features that represent measurements of the oocyte and mechanical features that represent deformation of the oocyte during the time pressure is applied to the oocyte. A method comprising outputting information indicating that one or more of the oocytes are selected based on the metrics.

26. The method according to claim 25, wherein the image shows the oocyte being aspirated into the part of the tool.

27. The aforementioned mechanical characteristics indicate the elasticity of the oocyte. The method according to claim 25, wherein the mechanical characteristics are determined based on the depth of aspiration of the oocytes to the portion of the tool.

28. The method according to claim 25, wherein the morphological features are determined by a first measurement relating to the cytoplasm of the oocyte and a second measurement relating to the zona pellucida of the oocyte.

29. The information indicating the selection is output via an interactive user interface. The aforementioned interactive user interface is A graphical representation of one or more of the oocytes is presented, In response to user input to select a specific oocyte from one or more of the aforementioned oocytes, The method according to claim 25, which provides detailed information relating to the specific oocyte.

30. The information indicating the selection is output via an interactive user interface. The method according to claim 25, wherein the interactive user interface presents individual text descriptions that group together one or more of the oocytes.

Images