Ultrasonic monitoring data conversion and identification method and equipment in hyper-ovulation process

Abstract

The invention provides an ultrasonic monitoring data conversion and identification method and equipment in a hyper-ovulation process, and the method comprises the steps: collecting two-dimensional and three-dimensional ultrasonic images in a follicle development process, marking the follicle two-dimensional images, and constructing an original follicle image data set; identifying and converting test set data in the data set, and completing statistics and analysis of test set prediction results; a two-dimensional ultrasonic method and a three-dimensional ultrasonic method are combined, a simplified scheme is defined according to the Wadell sphericity degree to calculate the sphericity degree of the follicles, and a training data set is expanded, so that the model can better detect the sphericity degree of the follicles; three-dimensional modeling is carried out, an ovary target is marked, and the target comprises an outer envelope; a cortical layer; according to the target space-time distance, calculating the migration path of each follicle in the development process according to the target space-time distance, and completing follicle development capability evaluation through the ovarian tissue mechanical distribution diagram. The present invention improves the oneness of follicle measurement methods.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of artificial intelligence technology, and in particular to a method and device for converting and identifying ultrasonic monitoring data during superovulation. Background Art

[0002] With the progress of assisted reproductive technology, the concept of individualized treatment has received more and more attention. Including individualized ovulation induction program, individualized ovulation induction drug dosage, individualized luteal support, individualized transplantation, etc. According to the basic characteristics and treatment response of different individuals, different clinical treatment methods are selected to continuously improve the success rate of IVF, which is the goal of assisted reproductive doctors. At present, the dosage of medication is mainly determined based on the size and number of antral follicles under B-ultrasound and the individual serum hormone level. The measurement of follicle size based on two-dimensional follicle structure is often inaccurate. In fact, when the follicles are too crowded, the follicles rarely present a perfect spherical shape. Through three-dimensional (3D) ultrasound measurement methods, more information about follicle morphology, including the sphericity of the follicles, can be obtained. However, the measurement of follicle sphericity and its impact on egg quality are technical defects that need to be overcome, mainly manifested in: most of the current methods for follicle assessment are ultrasonic measurements, focusing on the two-dimensional characteristics of follicles, and less attention is paid to the three-dimensional morphology of follicles, and there is a lack of research focusing on the details of follicles; there is no public image data set for the three-dimensional morphology of follicles, and clinical guidance is needed to complete the three-dimensional measurement of follicles and build a standard data set; the existing three-dimensional follicle technology has achieved unsatisfactory results in follicle assessment, and lacks attention to the three-dimensional morphology of follicles and the interaction of follicles. In different treatment cycles, the individual superovulation process is insensitive to the use of exogenous hormones, resulting in poor growth of follicles. Therefore, the development of a method and equipment for converting and identifying ultrasound monitoring data during superovulation can effectively overcome the defects in the above-mentioned related technologies, which has become a technical problem that needs to be solved urgently in the industry. Summary of the invention

[0003] In view of the above problems existing in the prior art, an embodiment of the present invention provides a method and device for converting and identifying ultrasound monitoring data during superovulation.

[0004] In the first aspect, an embodiment of the present invention provides a method for converting and identifying ultrasound monitoring data during superovulation, including: collecting two-dimensional and three-dimensional ultrasound images during follicular development, and annotating the two-dimensional images of follicles to construct an original follicle image data set; identifying and converting the test set data in the data set, and completing the statistics and analysis of the test set prediction results; combining two-dimensional and three-dimensional ultrasound methods, calculating the sphericity of the follicles according to the simplified definition of Wadell sphericity, expanding the training data set, so that the model can better detect the sphericity of the follicles; performing three-dimensional modeling, marking the targets of the ovaries, the targets include: the outer membrane; the cortex; the medulla layer and each follicle that develops during the ovulation cycle, calculating the migration path of each follicle during development according to the spatiotemporal distance between the targets, and completing the follicle development ability assessment through the ovarian tissue mechanics distribution map.

[0005] Based on the contents of the above method embodiments, an ultrasonic monitoring data conversion and identification method during superovulation provided in an embodiment of the present invention, wherein the sphericity of follicles is calculated according to a simplified Wadell sphericity definition scheme, comprises: using an ultrasonic-based automatic volume counting system to automatically identify and measure the three-dimensional volume of follicles, displaying the plane of the maximum diameter of the ovary, and performing a three-dimensional volume scan after the image is stabilized to obtain the follicle volume and the diameter of a sphere equivalent to the follicle volume.

[0006] Based on the content of the above method embodiment, in the method for converting and identifying ultrasonic monitoring data during superovulation provided in the embodiment of the present invention, the diameter of the equivalent sphere includes:

[0007]

[0008] Where dv is the diameter of the equivalent sphere; V is the volume of the follicle; and π is the pi.

[0009] Based on the content of the above method embodiment, the ultrasonic monitoring data conversion and identification method in the superovulation process provided in the embodiment of the present invention, the sphericity of the follicle includes:

[0010]

[0011] Among them, Sphericity is the sphericity of the follicle; X is the maximum diameter of the follicle.

[0012] Based on the contents of the above method embodiments, the ultrasound monitoring data conversion and identification method in the superovulation process provided in the embodiments of the present invention, and the annotation of the two-dimensional image of the follicle, include: selecting the follicle picture as the data set, completing the ultrasound image annotation according to the guidance of the clinician, and allocating the ultrasound image data set in a ratio of 3:1:1 to obtain a training set, a validation set and a test set respectively.

[0013] Based on the content of the above method embodiment, the ultrasound monitoring data conversion and identification method during superovulation provided in the embodiment of the present invention, the test set data in the data set is identified and converted, including: using the YOLOv5 network to identify the two-dimensional follicle ultrasound image and the Wadell sphericity definition simplification scheme, the YOLOv5 network includes: an input component, a trunk component, a neck component and a prediction component.

[0014] On the basis of the contents of the above method embodiments, the ultrasound monitoring data conversion and identification method during superovulation provided in the embodiments of the present invention, the input component adopts an adaptive image scaling method to seamlessly adapt to follicle ultrasound images of different individual characteristics.

[0015] In the second aspect, an embodiment of the present invention provides an ultrasonic monitoring data conversion and identification device during superovulation, including: a first main module, used to collect two-dimensional and three-dimensional ultrasonic images during follicular development, and annotate the two-dimensional images of follicles to construct an original follicle image data set; a second main module, used to identify and convert the test set data in the data set, and complete the statistics and analysis of the test set prediction results; a third main module, used to combine two-dimensional and three-dimensional ultrasonic methods, calculate the sphericity of follicles according to the simplified definition scheme of Wadell sphericity, expand the training data set, and enable the model to better detect the sphericity of follicles; a fourth main module, used to implement three-dimensional modeling and mark ovarian targets, the targets include: outer membrane; cortical layer; medullary layer and each follicle developed during the ovulation cycle, the migration path of each follicle during development is calculated according to the spatiotemporal distance between the targets, and the follicle development ability is evaluated through the ovarian tissue mechanics distribution map.

[0016] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0017] At least one processor, at least one memory and a communication interface; wherein,

[0018] The processor, memory and communication interface communicate with each other;

[0019] The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the ultrasound monitoring data conversion and identification method during superovulation provided by any one of the various implementations of the first aspect.

[0020] In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable a computer to execute the method for converting and identifying ultrasonic monitoring data during superovulation provided by any one of the various implementation methods of the first aspect.

[0021] The ultrasonic monitoring data conversion and identification method and device during superovulation provided in the embodiment of the present invention have a big data foundation, collect a large number of follicular development images, and organize a number of clinical doctors to complete the data set collection, and complete the sphericity calculation model training based on the collected data; the AI ​​model used for follicular sphericity calculation is an end-to-end model, which can automatically complete the follicular sphericity calculation, and no human intervention is required during the detection process; the AI ​​model can output accurate follicular sphericity, supplement the details of follicular development, and improve the singleness of traditional follicle measurement methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0023] Figure 1 A schematic diagram of a process flow of a method for converting and identifying ultrasonic monitoring data during superovulation provided by an embodiment of the present invention;

[0024] Figure 2 A schematic diagram of the structure of an ultrasonic monitoring data conversion and identification device during superovulation provided by an embodiment of the present invention;

[0025] Figure 3 A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention;

[0026] Figure 4 A schematic diagram of the 3D measurement effect of follicles provided in an embodiment of the present invention;

[0027] Figure 5 A schematic diagram of the effect of the YOLO model provided in an embodiment of the present invention on identifying two-dimensional follicles;

[0028] Figure 6 A schematic diagram of the calculation principle of follicle sphericity provided by an embodiment of the present invention;

[0029] Figure 7 A schematic diagram of the spatial mechanical evaluation effect of follicle development provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical scheme and advantages of the embodiment of the present invention clearer, the technical scheme in the embodiment of the present invention will be clearly and completely described in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention can be combined with each other arbitrarily to form a feasible technical solution. This combination is not subject to the constraints of the sequence of steps and / or the structural composition mode, but must be based on the ability of ordinary technicians in the field to achieve. When the combination of technical solutions is contradictory or cannot be achieved, it should be considered that the combination of such technical solutions does not exist and is not within the scope of protection required by the present invention. If there are step numbers in the following embodiments, they are only set for the convenience of explanation, and the order between the steps is not limited in any way. The execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0031] The embodiment of the present invention provides a method for converting and identifying ultrasound monitoring data during superovulation. Figure 1 The method includes: collecting two-dimensional and three-dimensional ultrasound images during follicular development, annotating the two-dimensional images of follicles, and constructing an original follicle image data set; identifying and transforming the test set data in the data set, and completing the statistics and analysis of the test set prediction results; combining two-dimensional and three-dimensional ultrasound methods, calculating the sphericity of follicles according to the simplified definition scheme of Wadell sphericity, expanding the training data set, so that the model can better detect the sphericity of follicles; performing three-dimensional modeling, marking the targets of the ovary, the targets include: the outer capsule; the cortex; the medulla layer and each follicle developed during the ovulation cycle, calculating the migration path of each follicle during development according to the spatiotemporal distance between the targets, and completing the evaluation of follicle development ability through the ovarian tissue mechanics distribution map.

[0032] Based on the contents of the above method embodiments, as an optional embodiment, the ultrasound monitoring data conversion and identification method during superovulation provided in the embodiments of the present invention, wherein the sphericity of the follicle is calculated according to the simplified Wadell sphericity definition scheme, comprises: using an ultrasound-based automatic volume counting system to automatically identify and measure the three-dimensional volume of the follicle, displaying the plane of the maximum diameter of the ovary, and performing a three-dimensional volume scan after the image is stabilized to obtain the follicle volume and the diameter of a sphere equivalent to the follicle volume.

[0033] Based on the content of the above method embodiment, as an optional embodiment, in the ultrasonic monitoring data conversion and identification method during superovulation provided in the embodiment of the present invention, the diameter of the equivalent sphere includes:

[0034]

[0035] Where dv is the diameter of the equivalent sphere; V is the volume of the follicle; and π is the pi.

[0036] Specifically, the maximum diameter, follicle volume and equivalent sphere provided by the two-dimensional and three-dimensional images are combined, and the follicle sphericity is obtained through these two types of image information, such as Figure 6 As shown. In three-dimensional ultrasound follicle monitoring, an ultrasound-based automatic volume counting system (SonoAVC) is used to automatically identify and measure the three-dimensional volume of follicles. The plane of the maximum diameter of the ovary is displayed, and after the image is stabilized, a three-dimensional volume scan is performed to obtain the follicle volume (V) and the diameter of the equivalent sphere of the volume (dv). The equivalent sphere diameter is shown in formula (1).

[0037] Based on the content of the above method embodiment, as an optional embodiment, the ultrasound monitoring data conversion and identification method in the superovulation process provided in the embodiment of the present invention, the sphericity of the follicle includes:

[0038]

[0039] Among them, Sphericity is the sphericity of the follicle; X is the maximum diameter of the follicle.

[0040] Specifically, the formula is executed by comparing the minimum circumscribed sphere (obtained by the maximum diameter (X) of the follicle identified by the two-dimensional ultrasound image) with the equivalent sphere diameter to obtain the follicle sphericity as shown in formula (2). After three-dimensional modeling, the outer membrane of the ovary; the cortex; the medulla; and each follicle that develops during the ovulation cycle are accurately marked. Full-dimensional dynamic convolution (ODConv) is used to enhance feature learning capabilities, thereby providing greater flexibility in processing inputs of different sizes. ODConv greatly improves the feature extraction capability of the convolutional layer by learning the attention mechanism on the four dimensions of the convolution kernel space (position, channel, filter, and kernel). By gradually applying attention on these dimensions, the network can dynamically adjust the importance of features, thereby more effectively capturing changes in follicle position and achieving better overall performance in feature extraction. The migration path of each follicle during development is obtained by calculating the spatiotemporal distance between the above targets (the ovulation cycle is generally 8-12 days), and the ovarian tissue mechanical distribution map (such as Figure 7 as shown) to complete an assessment of their developmental capabilities.

[0041] Based on the content of the above method embodiment, as an optional embodiment, the ultrasound monitoring data conversion and identification method in the superovulation process provided in the embodiment of the present invention, and the annotation of the two-dimensional image of the follicle, includes: selecting the follicle picture as the data set, completing the ultrasound image annotation according to the guidance of the clinician, and allocating the ultrasound image data set in a ratio of 3:1:1 to obtain a training set, a validation set and a test set respectively.

[0042] Based on the content of the above method embodiment, as an optional embodiment, the ultrasound monitoring data conversion and identification method in the superovulation process provided in the embodiment of the present invention, the test set data in the data set is identified and converted, including: using the YOLOv5 network to identify the two-dimensional follicle ultrasound image and the Wadell sphericity definition simplification scheme, the YOLOv5 network includes: an input component, a trunk component, a neck component and a prediction component.

[0043] Specifically, for the problems faced during follicle detection, such as follicles squeezing each other and image background differences, it is believed that this is due to the lack of generalization and robustness of the model itself, and its poor adaptability to new environments. Data enhancement is a widely used technology in machine learning and an important form of regularization. It plays a pivotal role in improving the generalization and robustness of the model. However, when dealing with the spherical features of follicles, conventional data enhancement methods such as flipping and cropping may not be able to effectively change the image position and other characteristics of such features. Therefore, a data conversion algorithm is proposed based on the characteristics of follicle images.

[0044] First, regarding the problem of follicle squeezing, there is a clear difference between the images of follicles during ovulation stimulation and the ultrasound images of follicle development under natural conditions. In the images of ovulation stimulation, the squeezing of follicles usually causes the spherical features of the follicles to be completely lost, making it impossible to determine the spherical target of the follicles. However, due to the stereoscopic morphology of the follicles fitted in the three-dimensional ultrasound images, the follicles squeezed against each other in the image can still obtain relevant stereoscopic morphological features such as Figure 4 As shown in the figure. Considering the squeezing between follicles as a unique mechanical form, when facing traditional two-dimensional images, the spherical features are completely lost and the two-dimensional images cannot obtain the relevant morphology. On the contrary, when facing follicles squeezing each other, because three-dimensional ultrasound adapts to this unique mechanical form, its plane recognition ability is limited, and it is difficult to identify the maximum diameter of the follicle. Therefore, how to combine two-dimensional and three-dimensional ultrasound features is the key.

[0045] In this regard, in order to extract the maximum exposed diameter and position of the follicle in the two-dimensional ultrasound image, the YOLOv5 network is used as the basic network to identify the two-dimensional follicle image. Figure 5As shown in Figure 1. The framework consists of four different components: input, trunk, neck, and prediction. The trunk network is usually synonymous with the feature extraction network. It bears the important responsibility of extracting relevant information from the input image for use by the subsequent network. By extracting features from the input image containing the target, the corresponding feature map is obtained. In target detection, the trunk network is the basic part of the deep learning model, which is responsible for extracting features from the input image for subsequent tasks. By adopting the YOLOv5 network, the special interference caused by the squeezing of follicles is successfully simulated, thereby enriching the dataset and enabling the model to prioritize the overall characteristics of the follicles and extract the maximum diameter of the follicles.

[0046] Based on the content of the above method embodiment, as an optional embodiment, the ultrasound monitoring data conversion and identification method during superovulation provided in the embodiment of the present invention, the input component adopts an adaptive image scaling method to seamlessly adapt to the follicle ultrasound images of different individual characteristics.

[0047] Specifically, in the data collection and preprocessing stage, the present invention collects two-dimensional ultrasound images of follicle development during ovulation induction from the reproductive center, selects follicle images with clear and visible contours as the data set, completes the annotation of ultrasound images under the guidance of clinicians in the reproductive center, and distributes the ultrasound image data set in a ratio of 3:1:1 to obtain a training set, a validation set, and a test set. In the model training stage, the first step is feature extraction. The two-dimensional ultrasound image in the training set is used as input. After extracting feature information through the backbone network in a forward propagation manner, the corresponding feature map is obtained; then the feature map obtained from some layers of the backbone network is sent to the neck area for feature fusion, generally some feature pyramids are used for feature fusion; finally, the fused feature maps are sent to the detection head to classify and identify the target. The second step is model training. For the existing data set, we trained for 100 epochs with an initial learning rate of 0.01. The software environment used in the experiment includes Ubuntu 18.04, CUDA 10.2, Python 3.7, and Torch 1.8. In the model prediction stage, the ultrasound images in the test set are used as input to the training network, and the model outputs the effect of follicle prediction and gives the confidence level.

[0048] The method for converting and identifying ultrasonic monitoring data during superovulation provided in an embodiment of the present invention has a big data foundation, collects a large number of follicular development images, and organizes multiple clinicians to complete the data set collection, and completes the sphericity calculation model training based on the collected data; the AI ​​model used for follicular sphericity calculation is an end-to-end model, which can automatically complete the follicular sphericity calculation, and no human intervention is required during the detection process; the AI ​​model can output accurate follicular sphericity, supplements the details of follicular development, and improves the singleness of traditional follicle measurement methods.

[0049] The implementation basis of each embodiment of the present invention is to implement programmed processing through a device with a processor function. Therefore, in engineering practice, the technical solutions and functions of each embodiment of the present invention can be encapsulated into various modules. Based on this reality, on the basis of the above embodiments, an embodiment of the present invention provides an ultrasonic monitoring data conversion and identification device during superovulation, which is used to execute the ultrasonic monitoring data conversion and identification method during superovulation in the above method embodiment. See Figure 2 The device includes: a first main module, which is used to collect two-dimensional and three-dimensional ultrasound images during follicular development, annotate the two-dimensional images of follicles, and construct an original follicle image data set; a second main module, which is used to recognize and transform the test set data in the data set, and complete the statistics and analysis of the test set prediction results; a third main module, which is used to combine two-dimensional and three-dimensional ultrasound methods, calculate the sphericity of follicles according to the simplified definition scheme of Wadell sphericity, expand the training data set, and enable the model to better detect the sphericity of follicles; a fourth main module, which is used to implement three-dimensional modeling and mark the targets of the ovary, including: the outer membrane; the cortex; the medulla layer and each follicle developed during the ovulation promotion cycle, calculate the migration path of each follicle during development according to the spatiotemporal distance between the targets, and complete the evaluation of follicle development ability through the ovarian tissue mechanics distribution map.

[0050] The ultrasonic monitoring data conversion and identification device in the superovulation process provided by the embodiment of the present invention adopts Figure 2 Several modules in it have a big data foundation, collected a large number of follicular development images, and organized a number of clinicians to complete the data set collection. Based on the collected data, the sphericity calculation model training was completed; the AI ​​model used for follicular sphericity calculation is an end-to-end model, which can automatically complete the follicular sphericity calculation, and no human intervention is required during the detection process; the AI ​​model can output accurate follicular sphericity, supplement the details of follicular development, and improve the singleness of traditional follicle measurement methods.

[0051] It should be noted that the device in the device embodiment provided by the present invention can be used to implement the method in the above method embodiment as well as the method in other method embodiments provided by the present invention. The only difference is that the corresponding functional modules are set, and the principle is basically the same as the principle of the above device embodiment provided by the present invention. As long as the technical personnel in the field refer to the specific technical solutions in other method embodiments on the basis of the above device embodiment, obtain the corresponding technical means and the technical solutions composed of these technical means by combining technical features, the device in the above device embodiment can be improved on the premise of ensuring the practicality of the technical solution, thereby obtaining the corresponding device class embodiment, which is used to implement the methods in other method class embodiments. For example:

[0052] Based on the contents of the above-mentioned device embodiment, as an optional embodiment, the ultrasound monitoring data conversion and identification device during superovulation provided in the embodiment of the present invention also includes: a first submodule, used to implement the calculation of the sphericity of follicles according to the simplified definition scheme of Wadell sphericity, including: using an ultrasound-based automatic volume counting system to automatically identify and measure the three-dimensional volume of follicles, displaying the plane of the maximum diameter of the ovary, and performing a three-dimensional volume scan after the image is stabilized to obtain the follicle volume and the diameter of a sphere equivalent to the follicle volume.

[0053] Based on the content of the above device embodiment, as an optional embodiment, the ultrasonic monitoring data conversion and identification device in the superovulation process provided in the embodiment of the present invention further includes: a second submodule, which is used to realize the diameter of the equivalent sphere, including:

[0054]

[0055] Where dv is the diameter of the equivalent sphere; V is the volume of the follicle; and π is the pi.

[0056] Based on the content of the above device embodiment, as an optional embodiment, the ultrasound monitoring data conversion and identification device in the superovulation process provided in the embodiment of the present invention further includes: a third submodule for realizing the sphericity of the follicle, including:

[0057]

[0058] Among them, Sphericity is the sphericity of the follicle; X is the maximum diameter of the follicle.

[0059] Based on the contents of the above-mentioned device embodiment, as an optional embodiment, the ultrasound monitoring data conversion and identification device in the superovulation process provided in the embodiment of the present invention also includes: a fourth submodule, used to implement the above and annotate the two-dimensional image of the follicle, including: selecting the follicle picture as the data set, completing the ultrasound image annotation according to the guidance of the clinician, and allocating the ultrasound image data set in a ratio of 3:1:1 to obtain a training set, a validation set and a test set respectively.

[0060] Based on the content of the above-mentioned device embodiment, as an optional embodiment, the ultrasound monitoring data conversion and identification device during superovulation provided in the embodiment of the present invention also includes: a fifth submodule, used to realize the identification and conversion of the test set data in the data set, including: using the YOLOv5 network to identify the two-dimensional follicle ultrasound image and the Wadell sphericity definition simplification scheme, the YOLOv5 network includes: an input component, a trunk component, a neck component and a prediction component.

[0061] Based on the content of the above-mentioned device embodiment, as an optional embodiment, the ultrasound monitoring data conversion and identification device during superovulation provided in the embodiment of the present invention also includes: a sixth sub-module, which is used to realize that the input component adopts an adaptive image scaling method to seamlessly adapt to follicle ultrasound images with different individual characteristics.

[0062] The method of the embodiment of the present invention is implemented by relying on electronic devices, so it is necessary to introduce the relevant electronic devices. Based on this purpose, the embodiment of the present invention provides an electronic device, such as Figure 3 As shown, the electronic device includes: at least one processor, a communication interface, at least one memory, and a communication bus, wherein at least one processor, the communication interface, and at least one memory communicate with each other through the communication bus. At least one processor can call the logic instructions in at least one memory to execute all or part of the steps of the method provided by the aforementioned various method embodiments.

[0063] In addition, the logic instructions in the at least one memory mentioned above can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each method embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0064] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0065] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0066] The flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. Based on this understanding, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and sometimes in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0067] It should be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "includes..." do not exclude the existence of other identical elements in the process, method, article or device including the elements. Any "predetermined threshold", "preset threshold" and other similar expressions that do not indicate specific values ​​can be determined by a person of ordinary skill in the art through simple experiments or corresponding debugging.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for converting and identifying ultrasonic monitoring data during superovulation, characterized in that: include: Collect 2D and 3D ultrasound images of follicles during development, annotate the 2D images of follicles, and construct a dataset of original follicle images; identify and transform the test set data in the dataset, and complete the statistics and analysis of the test set prediction results; combine 2D and 3D ultrasound methods, calculate the sphericity of follicles according to the simplified definition of Wadell sphericity, expand the training dataset, and enable the model to better detect the sphericity of follicles; Three-dimensional modeling is performed to mark the targets of the ovary, including the outer membrane, cortical layer, medullary layer, and each follicle that develops during the ovulation cycle. The migration path of each follicle during development is calculated based on the temporal and spatial distances between the targets, and the follicle development capacity is evaluated through the ovarian tissue mechanics distribution map.

2. The method for converting and identifying ultrasonic monitoring data during superovulation according to claim 1, characterized in that: The method of calculating the sphericity of follicles according to the simplified Wadell sphericity definition scheme includes: using an ultrasonic-based automatic volume counting system to automatically identify and measure the three-dimensional volume of follicles, displaying the plane of the maximum diameter of the ovary, and performing a three-dimensional volume scan after the image is stabilized to obtain the follicle volume and the diameter of a sphere equivalent to the follicle volume.

3. The method for converting and identifying ultrasonic monitoring data during superovulation according to claim 2, characterized in that: The diameter of the equivalent sphere includes: Where dv is the diameter of the equivalent sphere; V is the volume of the follicle; and π is the pi.

4. The method for converting and identifying ultrasonic monitoring data during superovulation according to claim 3, characterized in that: The sphericity of the follicle includes: Among them, Sphericity is the sphericity of the follicle; X is the maximum diameter of the follicle.

5. The method for converting and identifying ultrasonic monitoring data during superovulation according to claim 4, characterized in that: The method of labeling the two-dimensional follicle image includes: selecting the follicle image as a data set, labeling the ultrasound image according to the guidance of a clinician, and allocating the ultrasound image data set in a ratio of 3:1:1 to obtain a training set, a validation set, and a test set, respectively.

6. The method for converting and identifying ultrasonic monitoring data during superovulation according to claim 5, characterized in that: The identification and transformation of the test set data in the data set includes: using the YOLOv5 network to identify the two-dimensional follicle ultrasound image and the Wadell sphericity definition simplification scheme, and the YOLOv5 network includes: an input component, a trunk component, a neck component and a prediction component.

7. The method for converting and identifying ultrasonic monitoring data during superovulation according to claim 6, characterized in that: The input component adopts an adaptive image scaling method to seamlessly adapt to follicle ultrasound images with different individual characteristics.

8. A device for converting and identifying ultrasonic monitoring data during superovulation, characterized in that: include: The first main module is used to collect two-dimensional and three-dimensional ultrasound images of follicle development, annotate the two-dimensional images of follicles, and construct a data set of original follicle images; The second main module is used to identify and transform the test set data in the data set, and complete the statistics and analysis of the test set prediction results; the third main module is used to combine the two-dimensional and three-dimensional ultrasound methods, calculate the sphericity of the follicle according to the simplified definition of Wadell sphericity, expand the training data set, and enable the model to better detect the sphericity of the follicle; The fourth main module is used to realize three-dimensional modeling and mark the targets of the ovary, which include: the outer membrane; the cortical layer; the medullary layer and each follicle that develops during the ovulation cycle. The migration path of each follicle during development is calculated based on the temporal and spatial distances between the targets, and the follicle development ability is evaluated through the ovarian tissue mechanics distribution map.

9. An electronic device, characterized in that: include: At least one processor, at least one memory and a communication interface; wherein, The processor, memory and communication interface communicate with each other; The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, which cause a computer to execute the method of any one of claims 1 to 7.

Images