Tire direction identification method and device, computer equipment and storage medium
Ultrasound images are analyzed through deep learning models, combined with fixed measurement parts and cross-cut scanning techniques, the problem of low operation dependence and automation in tire orientation judgment is solved, and efficient and accurate tire orientation recognition is achieved.
Patent Information
- Application Number
- CN202510314338.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-17
- Publication Date
- 2025-06-24
AI Technical Summary
The prior art has problems of low operation dependence, subjectivity and automation in the judgment of tire orientation, making it difficult to achieve efficient, objective and real-time evaluation.
The fetal orientation recognition method based on deep learning is adopted. By preprocessing ultrasound images and inputting a pre-trained fetal orientation deep learning model, the fetal orientation information is output, and the fixed measurement part and cross-cutting scanning method are combined to reduce the recognition deviation caused by inconsistent operation.
It effectively reduces the dependence on operator experience, improves the accuracy and automation of identification, and can judge the fetal orientation in real time and accurately during delivery, providing an objective basis for clinical decision-making.
Smart Images
Figure CN120198401A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical fields of medical imaging and artificial intelligence, and particularly relates to a method, device, computer device, and storage medium for fetal position recognition. Background Art
[0002] Fetal position refers to the relative position of the fetal head in the maternal pelvis, and the fetal position will continuously change throughout the entire delivery process. Accurately identifying the fetal position is of great significance for judging the progress of labor and selecting an appropriate delivery method. Currently, the judgment of fetal position in clinical practice mainly relies on transvaginal digital examination or ultrasound examination. However, these methods each have limitations and it is difficult to achieve efficient, objective, and real-time fetal position assessment. Traditional ultrasound examination requires the operator to undergo long-term training in order to master the placement position of the probe, the method of identifying ultrasound images, and make a comprehensive judgment in combination with human anatomical structures. However, due to the high requirements for the experience of users in ultrasound examination, it is subject to certain limitations in clinical practical applications. Therefore, the judgment of fetal position still mainly relies on digital examination. However, digital examination has certain subjectivity, and there may be differences in judgment among different operators. At the same time, digital examination may cause discomfort to pregnant women, limiting its wide application.
[0003] In response to the above problems, in the prior art, a multi-directional and dynamically adjustable scanning method is also adopted. This method cannot ensure the one-to-one correspondence between image features and the fetal head direction. At the same time, when the target feature is not recognized, it is necessary to supplement image features through scanning instructions, greatly increasing human intervention and having a low degree of automation.
[0004] In recent years, significant progress has been made in the application of artificial intelligence (AI) technology in the field of medical image analysis. AI can automatically analyze and perform pattern recognition on ultrasound images through deep learning technology, improving the accuracy and objectivity of fetal position determination. Therefore, it is necessary to propose a method, device, computer device, and storage medium for fetal position recognition. Summary of the Invention
[0005] The purpose of the embodiments of this application is to propose a method, device, computer device, and storage medium for fetal position recognition, and its main purpose is to improve the efficiency and accuracy of standard plane recognition and intrapartum parameter measurement based on a multi-task network.
[0006] To solve the above technical problems, the embodiments of this application provide a method for fetal position recognition, and adopt the following technical solutions:
[0007] A method for fetal position recognition includes the following steps:
[0008] S1. Select the upper edge of the pubic symphysis or the perineum for transverse scanning according to the fetal head presentation situation, and obtain a real-time ultrasound image including anatomical structure parameters or fetal head direction parameters;
[0009] S2. Preprocess the ultrasound image, where the preprocessing includes denoising, enhancement, and normalization processing;
[0010] S3. Input the preprocessed ultrasound image into a pre-trained fetal position deep learning model, and the fetal position deep learning model outputs corresponding fetal position information according to anatomical structure parameters or fetal head direction parameters.
[0011] Further, after step S3, it further includes:
[0012] S4. Display the fetal position information.
[0013] Further, step S4 further includes:
[0014] S5. Judge whether the labor process continues. If so, repeat steps S1 - S4.
[0015] Further, step S1 specifically includes:
[0016] S11. Judge the situation of fetal head presentation;
[0017] S12. When the fetal head presentation is greater than +1, select the upper edge of the pubic symphysis for transverse scanning to obtain an ultrasound image containing anatomical structure parameters, where the anatomical structure parameters include orbits, cerebral midline, cerebellum, and cerebral angles;
[0018] S13. When the fetal head presentation is less than +1, select the perineum for transverse scanning to obtain an ultrasound image containing at least fetal head direction parameters, where the fetal head direction parameters include cerebral midline and cerebral angles.
[0019] Further, median filtering or Gaussian filtering is used for denoising in step S2, and histogram equalization is used for enhancement.
[0020] Further, the training steps of the fetal position deep learning model in step S3 include:
[0021] S31. Obtain a set number of ultrasound images during the labor process according to the method in step S1, extract and label the corresponding fetal position information for each ultrasound image, where the fetal position information includes occiput posterior, right occiput posterior, left occiput posterior, right occiput transverse, left occiput transverse, occiput anterior, left occiput anterior, and right occiput anterior;
[0022] S32. Use a convolutional neural network or a deep learning model integrating self - attention mechanism to train the ultrasound image data set with a set number of samples obtained during the labor process by the method in S1;
[0023] S33. Use optimization algorithms such as stochastic gradient descent or Adam to train the deep learning model to obtain a fetal position deep learning model in which the key features of the orbits, midline of the brain, cerebellum, and cerebral angles in the ultrasound image correspond to the fetal position information;
[0024] S34. Iteratively optimize the model accuracy through the cross-entropy loss function and the validation set to obtain the final fetal position deep learning model.
[0025] Further, the step S32 specifically includes:
[0026] When the fetal orbital structure appears in the image, make a determination based on the symmetry of the orbits: if the left and right orbits are symmetric, it is determined as occiput posterior; if the orbit deflects to the left, it is determined as right occiput posterior; if the orbit deflects to the right, it is determined as left occiput posterior;
[0027] When the obvious midline of the brain and cerebellum structures are visible in the image, make a determination based on the clock position pointed by the cerebellum. When the cerebellum points to the 8 - 10 o'clock position, it is right occiput transverse; when the cerebellum points to the 2 - 4 o'clock position, it is left occiput transverse;
[0028] When the obvious cerebral angle structure appears in the image, make a determination based on the clock position it points to. When it points to the 6 o'clock position, it is occiput anterior; when it points to the 7 - 8 o'clock position, it is left occiput anterior; when it points to the 4 - 5 o'clock position, it is right occiput anterior.
[0029] To solve the above technical problems, the embodiment of the present application further provides a device for fetal position recognition, adopting the following technical solutions:
[0030] A device for fetal position recognition, comprising:
[0031] A transverse scanning module, configured to perform transverse scanning on the upper edge of the pubic symphysis or the perineum according to the fetal head presentation situation to obtain a real-time ultrasound image including anatomical structure parameters or fetal head direction parameters;
[0032] A preprocessing module, configured to preprocess the ultrasound image, and the preprocessing includes denoising, enhancement, and normalization processing;
[0033] A fetal position judgment module, configured to input the preprocessed ultrasound image into a pre-trained fetal position deep learning model, and the fetal position deep learning model outputs corresponding fetal position information according to the anatomical structure parameters or fetal head direction parameters.
[0034] To solve the above technical problems, the embodiment of the present application further provides a computer device, adopting the following technical solutions:
[0035] A computer device includes a memory and a processor. Computer-readable instructions are stored in the memory. When the processor executes the computer-readable instructions, the steps of the method for fetal position recognition described above are implemented.
[0036] To solve the above technical problems, an embodiment of the present application also provides a computer-readable storage medium, adopting the following technical solutions:
[0037] A computer-readable storage medium, characterized in that computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by a processor, the steps of the method for fetal position recognition described above are implemented.
[0038] Compared with the prior art, the embodiments of the present application mainly have the following beneficial effects: The present application effectively reduces the dependence on the operator's experience. At the same time, by using fixed measurement parts and transverse scanning techniques, the recognition deviation caused by inconsistent operations is greatly reduced. Combined with the deep learning model, which can learn and classify different fetal position characteristics, the present application can accurately judge the fetal position in real time during childbirth, providing an objective basis for clinical decision-making, with a higher degree of automation, reducing manual intervention, and ensuring a one-to-one correspondence between image features and the fetal head direction. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] To more clearly illustrate the solutions in the present application, the following will briefly introduce the drawings required for the description of the embodiments of the present application. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0040] Figure 1 It is an exemplary system architecture diagram to which the present application can be applied;
[0041] Figure 2 A flowchart of an embodiment of the method for fetal position recognition according to the present application;
[0042] Figure 3 It is a schematic diagram of the principle of the fetal position recognition device of the present application;
[0043] Figure 4 It is a schematic structural diagram of an embodiment of the computer device according to the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs; the terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above drawings are intended to cover non-exclusive inclusion. The terms "first", "second", etc. in the specification and claims of this application or the above drawings are used to distinguish different objects and not to describe a specific order.
[0045] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in connection with the embodiments can be included in at least one embodiment of this application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.
[0046] To enable those skilled in the technical field to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.
[0047] As Figure 1 shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.
[0048] Users can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications may be installed on the terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.
[0049] The terminal devices 101, 102, and 103 can be various electronic devices with a display screen and supporting web browsing, including but not limited to smartphones, tablets, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, and desktop computers, etc.
[0050] The server 105 can be a server that provides various services, such as a background server that supports the pages displayed on the terminal devices 101, 102, and 103.
[0051] It should be noted that the method for fetal position recognition provided by the embodiments of the present application is generally executed by the server. Correspondingly, the fetal position recognition device is generally set in the terminal device.
[0052] It should be understood that Figure 1 the numbers of the terminal devices, network, and server in
[0053] Continuing to refer to Figure 2 , a flowchart of an embodiment of the method for fetal position recognition according to the present application is shown. The method for fetal position recognition includes the following steps:
[0054] Step S1: Select the upper edge of the pubic symphysis or the perineum for transverse scanning according to the fetal head presentation situation to obtain a real-time ultrasound image including anatomical structure parameters or fetal head direction parameters;
[0055] In this embodiment, the step S1 specifically includes:
[0056] Step S11: Judge the fetal head presentation situation.
[0057] Step S12: If the fetal head presentation is greater than +1 (the fetal head position is higher), select the upper edge of the pubic symphysis for transverse scanning to obtain an ultrasound image including anatomical structure parameters, and the anatomical structure parameters include the eye socket, cerebral midline, cerebellum, and cerebral angle.
[0058] Step S13: If the fetal head presentation is less than +1, select the perineum for transverse scanning to obtain an ultrasound image including at least fetal head direction parameters, and the fetal head direction parameters include the cerebral midline and cerebral angle.
[0059] Among them, the upper edge of the symphysis pubis and the perineum are distributed along the same longitudinal axis. By using the same transverse probe scanning method, it can be ensured that when the images present the same anatomical features, the corresponding fetal head directions are consistent, thus reducing the complexity introduced by "the same image, different scanning positions" in the traditional method.
[0060] Step S2: Preprocess the ultrasonic image, and the preprocessing includes denoising, enhancement, and normalization processing;
[0061] In step S2, a conventional two-dimensional ultrasonic device is used to perform real-time transverse scanning on the selected measurement site to obtain a sequence of ultrasonic images of the fetal head and surrounding structures. Median filtering or Gaussian filtering is used for denoising, and histogram equalization is used for enhancement to obtain a clearer and more standardized image input.
[0062] Step S3: Input the preprocessed ultrasonic image into a pre-trained fetal position deep learning model, and the fetal position deep learning model outputs the corresponding fetal position information according to the anatomical structure parameters or fetal head direction parameters.
[0063] The training steps of the fetal position deep learning model in step S3 specifically include:
[0064] Step S31: Obtain a set number of ultrasonic images during the labor process according to the method in step S1, extract and label the fetal position information corresponding to each ultrasonic image. The fetal position information includes occiput posterior, right occiput posterior, left occiput posterior, right occiput transverse, left occiput transverse, occiput anterior, left occiput anterior, and right occiput anterior. That is, judge each frame of the ultrasonic image video one by one and assign the corresponding fetal position label to the fetal position category corresponding to each frame, including occiput posterior, right occiput posterior, left occiput posterior, right occiput transverse, left occiput transverse, occiput anterior, left occiput anterior, and right occiput;
[0065] Step S32: Use a convolutional neural network or a deep learning model integrating self-attention mechanism to train the model on the ultrasonic image dataset with a set number of samples obtained during the labor process by the method in S1;
[0066] Step S33: Use optimization algorithms such as stochastic gradient descent or Adam to train the deep learning model to obtain a fetal position deep learning model in which the key features of the orbits, cerebral midline, cerebellum, and cerebral angles in the ultrasonic image correspond to the fetal position information;
[0067] Step S34: Iteratively optimize the model accuracy through the cross-entropy loss function and the validation set to obtain the final fetal position deep learning model.
[0068] Step S4: Display the fetal position information, which can assist medical staff in judging and making decisions on the position of the fetal head during the labor process.
[0069] Step S5: Determine whether the labor process continues. If so, repeat Steps S1 - S4 to ensure continuous monitoring and evaluation of the entire delivery process.
[0070] It should be understood that although the steps in the flowchart of the accompanying drawings are shown sequentially according to the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless otherwise clearly stated in this document, there is no strict order restriction for the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub - steps or multiple stages. These sub - steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub - steps or stages of other steps.
[0071] This application effectively reduces the dependence on the operator's experience. At the same time, by using fixed measurement sites and transverse scanning techniques, it significantly reduces the recognition deviation caused by inconsistent operations. Coupled with the deep - learning model that can learn and classify different fetal position characteristics, this application can accurately judge the fetal position in real - time during the delivery process, providing an objective basis for clinical decision - making, with a higher degree of automation, reducing human intervention, and ensuring a one - to - one correspondence between image features and the fetal head direction.
[0072] Further referring to Figure 3 as an implementation of the method shown above, this application provides an embodiment of a fetal position recognition device. This device embodiment corresponds to the method embodiment shown in Figure 2 and can be specifically applied to various electronic devices. Figure 2
[0073] Figure 3 As shown in
[0074]
[0075] The transverse scanning module 301 is used to select the upper edge of the pubic symphysis or the perineum for transverse scanning according to the fetal head presentation situation, and obtain a real - time ultrasound image including anatomical structure parameters or fetal head direction parameters.
[0076] The pre - processing module 302 is used to pre - process the ultrasound image, and the pre - processing includes denoising, enhancement, and normalization processing.
[0077] The fetal position judgment module 303 is used to input the pre - processed ultrasound image into a pre - trained fetal position deep - learning model, and the fetal position deep - learning model outputs the corresponding fetal position information according to the anatomical structure parameters or fetal head direction parameters.
[0077] This application effectively reduces the dependence on the operator's experience. At the same time, by using fixed measurement sites and cross-sectional scanning techniques, it significantly reduces the recognition deviation caused by inconsistent operations. Combined with the deep learning model for learning and classifying different fetal position characteristics, this application can accurately determine the fetal position in real time during childbirth, providing an objective basis for clinical decision-making, with a higher degree of automation, reducing human intervention, and ensuring a one-to-one correspondence between image features and the fetal head direction.
[0078] To solve the above technical problems, an embodiment of this application also provides a computer device. Specifically, please refer to Figure 3 , Figure 3 which is the basic structural block diagram of the computer device in this embodiment.
[0079] The computer device 40 includes a memory 41, a processor 42, and a network interface 43 that are communicatively connected to each other through a system bus. It should be noted that only the computer device 40 with components 41 - 43 is shown in the figure, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented. Among them, those skilled in the art of this technology can understand that a computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.
[0080] The computer device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The computer device can interact with the user through means such as a keyboard, a mouse, a remote control, a touchpad, or a voice control device.
[0081] The memory 41 at least includes one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disc, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 40, such as the hard disk or memory of the computer device 40. In other embodiments, the memory 41 may also be an external storage device of the computer device 40, such as a plug-in hard disk equipped on the computer device 40, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Of course, the memory 41 may also include both the internal storage unit of the computer device 40 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system installed on the computer device 40 and various application software, such as computer-readable instructions of the method for fetal position identification. In addition, the memory 41 may also be used to temporarily store various data that have been output or will be output.
[0082] In some embodiments, the processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 42 is generally used to control the overall operation of the computer device 40. In this embodiment, the processor 42 is used to run the computer-readable instructions stored in the memory 41 or process data, such as running the computer-readable instructions of the method for fetal position identification.
[0083] The network interface 43 may include a wireless network interface or a wired network interface, and the network interface 43 is generally used to establish a communication connection between the computer device 40 and other electronic devices.
[0084] This application effectively reduces the dependence on the operator's experience. At the same time, by using fixed measurement parts and transverse scanning techniques, it greatly reduces the recognition deviation caused by inconsistent operations. Combined with the deep learning model that can learn and classify different fetal position characteristics, this application can accurately judge the fetal position in real time during childbirth, providing an objective basis for clinical decision-making, with a higher degree of automation, reducing manual intervention, and ensuring a one-to-one correspondence between image features and the fetal head direction.
[0085] The present application also provides another implementation manner, that is, to provide a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor, so that the at least one processor executes the steps of the method for fetal position recognition as described above.
[0086] The present application effectively reduces the dependence on the operator's experience. At the same time, by using fixed measurement parts and transverse scanning techniques, the recognition deviation caused by inconsistent operations is greatly reduced. Combined with the deep learning model for learning and classifying different fetal position characteristics, the present application can accurately judge the fetal position in real time during childbirth, provide an objective basis for clinical decision-making, have a higher degree of automation, reduce human intervention, and ensure the one-to-one correspondence between image features and the fetal head direction.
[0087] Through the description of the above implementation manners, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation manner. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc) and includes several instructions to enable a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of the present application.
[0088] Obviously, the embodiments described above are only a part of the embodiments of the present application, rather than all the embodiments. The drawings show the preferred embodiments of the present application, but do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosed content of the present application more thorough and comprehensive. Although the present application has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions described in the foregoing specific embodiments, or perform equivalent replacements for some of the technical features. Any equivalent structure directly or indirectly using the content of the specification and drawings of the present application in other related technical fields is equally within the scope of the patent protection of the present application.
Claims
1. A method for identifying fetal position, characterized in that: The steps include: S1. Select the upper edge of the pubic symphysis or the perineum for transverse scanning according to the fetal head presentation to obtain real-time ultrasound images including anatomical structure parameters or fetal head direction parameters; S2, preprocessing the ultrasound image, wherein the preprocessing includes denoising, enhancement and normalization processing; S3. Input the preprocessed ultrasound image into a pre-trained fetal position deep learning model, and the fetal position deep learning model outputs corresponding fetal position information according to anatomical structure parameters or fetal head direction parameters.
2. The method for identifying fetal position according to claim 1, characterized in that: The step S3 further includes: S4. Displaying the fetal position information.
3. The method for identifying fetal position according to claim 2, characterized in that: The step S4 further comprises: S5. Determine whether the labor process is continuing. If so, repeat steps S1 to S4.
4. The method for identifying fetal position according to claim 1, characterized in that: The step S1 specifically includes: S11. Determine the fetal head presentation; S12, if the fetal head presentation is greater than +1, a transverse scan is performed on the upper edge of the pubic symphysis to obtain an ultrasound image containing anatomical structure parameters, wherein the anatomical structure parameters include the orbit, the midline of the brain, the cerebellum, and the cerebral horn; S13. If the fetal head presentation is less than +1, select the perineum for transverse scanning to obtain an ultrasound image containing at least fetal head direction parameters, wherein the fetal head direction parameters include the brain midline and the cerebral angle.
5. The method for identifying fetal position according to claim 1, characterized in that: The denoising in step S2 is performed by median filtering or Gaussian filtering, and the enhancement is performed by histogram equalization.
6. The method for identifying fetal position according to claim 1, characterized in that: The training steps of the fetal position deep learning model in step S3 include: S31, according to the method of step S1, obtain a set number of ultrasound images of the labor process, extract and mark the fetal position information corresponding to each ultrasound image, wherein the fetal position information includes posterior occipital, posterior right occipital, posterior left occipital, transverse right occipital, transverse left occipital, anterior occipital, anterior left occipital, and anterior right occipital; S32, using a convolutional neural network or a deep learning model integrating a self-attention mechanism to train a model on a set number of ultrasound image datasets obtained during the labor process using the S1 method; S33, using random gradient descent or Adam and other optimization algorithms to train the deep learning model to obtain a fetal position deep learning model corresponding to the key features of the eye sockets, brain midline, cerebellum and cerebral horns in the ultrasound image and the fetal position information; S34. Iteratively optimize the model accuracy through the cross entropy loss function and the validation set to obtain the final fetal position deep learning model.
7. The method for identifying fetal position according to claim 6, characterized in that: The step S32 specifically includes: When the fetal orbital structure appears in the image, the judgment is made based on the symmetry of the orbit: if the left and right orbits are symmetrical, it is judged as right occipital posterior; if the orbit is deflected to the left, it is judged as right occipital posterior; if the orbit is deflected to the right, it is judged as left occipital posterior; When the brain midline and cerebellum structure are clearly visible in the image, the clock position pointed by the cerebellum is used for judgment. When the cerebellum points to the 8-10 o'clock position, it is the right occipital transverse position; when the cerebellum points to the 2-4 o'clock position, it is the left occipital transverse position. When there is an obvious brain angle structure in the image, judge it according to the clock position it points to. When it points to the 6 o'clock position, it is the front occipital position; when it points to the 7-8 o'clock position, it is the left occipital position; when it points to the 4-5 o'clock position, it is the right occipital position.
8. A device for identifying fetal position, characterized in that: include: A transverse scanning module is used to select the upper edge of the pubic symphysis or the perineum for transverse scanning according to the fetal head presentation to obtain a real-time ultrasound image including anatomical structure parameters or fetal head direction parameters; A preprocessing module, used for preprocessing the ultrasound image, wherein the preprocessing includes denoising, enhancement and normalization processing; The fetal position determination module is used to input the preprocessed ultrasound image into a pre-trained fetal position deep learning model, and the fetal position deep learning model outputs corresponding fetal position information according to anatomical structure parameters or fetal head direction parameters.
9. A computer device, comprising a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the method for fetal position identification according to any one of claims 1 to 7 when executing the computer-readable instructions.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the method for fetal position identification according to any one of claims 1 to 7 are implemented.