Fetal heart monitoring method and device based on intelligent model

Abstract

The embodiment of the invention provides a fetal heart monitoring method and device based on an intelligent model. The method comprises the following steps: acquiring a historical fetal orientation index, a historical umbilical artery blood flow index and a historical fetal heart rate index of a specific group of users; obtaining historical bad delivery outcome types of the specific group users; associating the historical fetal orientation index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type as sample data; inputting the historical fetal orientation index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model; inputting a fetal orientation index, an umbilical artery blood flow index and a fetal heart rate index of a to-be-detected user into the trained intelligent machine learning model to obtain an output delivery outcome type; the monitoring accuracy and multifunctionality of the fetal heart monitor are improved, and the problems that the learning sample data of an existing fetal heart monitoring AI model is single, and the prediction result is inaccurate are solved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a fetal heart monitoring method based on an intelligent model, a fetal heart monitoring device based on an intelligent model, a computer device and a storage medium. Background Art

[0002] The application of AI and machine learning in fetal heart rate monitors has a significant impact and has brought revolutionary changes to fetal health monitoring. First, through deep learning and pattern recognition, fetal heart rate monitors can automatically identify and interpret fetal heart rate patterns to achieve continuous monitoring of the fetus and risk warning. This greatly reduces the workload of doctors and improves medical efficiency; secondly, AI and machine learning can help optimize ultrasound, images, and improve image quality and diagnostic accuracy. Through deep learning algorithms, fetal heart rate monitors learn from a large amount of image data and can automatically identify and mark key diagnostic information such as fetal heart rate, fetal movement, etc.; in addition, AI and machine learning can also realize the personalized settings of fetal heart rate monitors, providing personalized monitoring and warning according to the specific situation of each fetus. For example, through machine learning algorithms, fetal heart rate monitors can learn from a large amount of historical data, and then provide personalized monitoring and warning services based on the characteristics and historical data of each fetus. However, the current machine learning algorithm has a single sample data, the prediction results are not very accurate, and the subjectivity is strong, which affects the accuracy of the model output results. Summary of the invention

[0003] In view of the above problems, the present embodiment is proposed to provide a fetal heart monitoring method based on an intelligent model, a fetal heart monitoring device based on an intelligent model, a computer device and a storage medium that overcome the above problems or at least partially solve the above problems.

[0004] In order to solve the above problems, this embodiment discloses a fetal heart monitoring method based on an intelligent model, which is characterized by comprising:

[0005] Obtain historical fetal position indicators, historical umbilical artery blood flow indicators, and historical fetal heart rate indicators of specific group users;

[0006] Obtaining the historical adverse delivery outcome types of users in the specific group;

[0007] The historical fetal position index, historical umbilical artery blood flow index, historical fetal heart rate index and historical delivery outcome type were associated as sample data;

[0008] Inputting historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators, and historical delivery outcome types into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model;

[0009] The fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be tested are input into the trained intelligent machine learning model to obtain the output type of delivery outcome.

[0010] Preferably, the acquisition of historical fetal position indicators, historical umbilical artery blood flow indicators, and fetal heart rate indicators of users in a specific group includes:

[0011] Obtaining raw medical data;

[0012] The medical raw data is identified according to the preset first keyword, and the historical fetal position indicators and historical umbilical artery blood flow indicators of the specific group of users are extracted; wherein the historical fetal position indicators include left occipital anterior position, left occipital transverse position, left occipital posterior position, right occipital anterior position, right occipital transverse position, right occipital posterior position, left anterior chin position, left transverse chin position, left posterior chin position, right anterior chin position, right transverse chin position, right posterior chin position, left anterior sacral position, left transverse sacral position, left posterior sacral position, right anterior sacral position, right transverse sacral position, right posterior sacral position, left anterior shoulder position, left posterior shoulder position, right anterior shoulder position, right posterior shoulder position, posterior occipital position, occipital transverse position, breech presentation, shoulder presentation and face presentation; the historical umbilical artery blood flow indicators include umbilical artery blood flow S / D value, umbilical artery resistance index and umbilical artery pulsation index.

[0013] Preferably, the obtaining of the historical adverse delivery outcome types of the specific group of users includes:

[0014] Obtaining raw medical data;

[0015] The medical raw data is identified according to the preset second keyword, and the historical adverse delivery outcome types of the specific group of users are extracted; wherein the historical adverse delivery outcome types include cesarean section, fetal distress, umbilical cord entanglement, oligohydramnios, meconium staining, intrauterine death, congenital fetal malformation, and neonatal intrauterine asphyxia.

[0016] Preferably, the preset intelligent machine learning model includes an agent module, an environment module, and a reward system module;

[0017] The historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type are input into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model, including:

[0018] The historical fetal position index, historical umbilical artery blood flow index, historical fetal heart rate index and historical delivery outcome type are divided into a training set and a validation set; the historical fetal position index, historical umbilical artery blood flow index and historical fetal heart rate index of the training set are used as an input layer, and the historical delivery outcome type is used as an output layer, and the input layer and the output layer are set in the environment module;

[0019] Observe the state value, action value, and reward value from the environment module and store the parameters of the current cycle;

[0020] The agent module generates the next action value and inputs it into the environment module to obtain the output state value and reward value;

[0021] The environment module inputs the state value and reward value to the agent module to determine whether the first cycle is over. If so, it outputs the prediction result.

[0022] If the first cycle is not completed, the intelligent agent module is returned to the step of receiving the state value and generating the action value, and the prediction accuracy of the verification set is used as the evaluation value. If the prediction accuracy does not meet the requirements, the setting parameters of the model are optimized, and a new state value is generated. The training step is returned to obtain the trained intelligent machine learning model.

[0023] Preferably, the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected are input into the trained intelligent machine learning model to obtain the output type of delivery outcome, including:

[0024] Obtaining the initial fetal position index and initial umbilical artery blood flow index of the user to be detected;

[0025] Screening the initial fetal position index and the initial umbilical artery blood flow index of the user to be detected to obtain the fetal position index and the umbilical artery blood flow index;

[0026] The fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be tested are input into the trained intelligent machine learning model to obtain the output type of delivery outcome.

[0027] This embodiment discloses a fetal heart monitoring device based on an intelligent model, comprising:

[0028] A historical data acquisition module, used to obtain historical fetal position indicators, historical umbilical artery blood flow indicators, and historical fetal heart rate indicators of a specific group of users;

[0029] A type data module, used to obtain the historical adverse delivery outcome types of the specific group of users;

[0030] An association module, used to associate historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators and historical delivery outcome types as sample data;

[0031] A training module, used for inputting historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators and historical delivery outcome types into a preset intelligent machine learning model for training, thereby obtaining a trained intelligent machine learning model;

[0032] The output module is used to input the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected into the trained intelligent machine learning model to obtain the output delivery outcome type.

[0033] Preferably, the historical data acquisition module includes:

[0034] The first data acquisition submodule is used to acquire raw medical data;

[0035] a first identification submodule, for identifying the medical raw data according to a preset first keyword, and extracting historical fetal position indicators and historical umbilical artery blood flow indicators of the specific group of users; wherein the historical fetal position indicators include left occipital anterior position, left occipital transverse position, left occipital posterior position, right occipital anterior position, right occipital transverse position, right occipital posterior position, left anterior chin position, left transverse chin position, left posterior chin position, right anterior chin position, right transverse chin position, right posterior chin position, left anterior sacral position, left transverse sacral position, left posterior sacral position, right anterior sacral position, right transverse sacral position, right posterior sacral position, left anterior shoulder position, left posterior shoulder position, right anterior shoulder position, right posterior shoulder position, posterior occipital position, occipital transverse position, breech presentation, shoulder presentation and face presentation; and the historical umbilical artery blood flow indicators include umbilical artery blood flow S / D value, umbilical artery resistance index and umbilical artery pulsation index.

[0036] Preferably, the type data module includes:

[0037] The second data acquisition submodule is used to acquire raw medical data;

[0038] The second identification submodule is used to identify the medical raw data according to the preset second keyword, and extract the historical adverse delivery outcome types of the specific group of users; wherein the historical adverse delivery outcome types include cesarean section, fetal distress, umbilical cord entanglement, oligohydramnios, meconium staining, intrauterine death, congenital fetal malformation, and neonatal intrauterine asphyxia.

[0039] This embodiment also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned steps of fetal heart monitoring based on the intelligent model when executing the computer program.

[0040] This embodiment also discloses a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned steps of fetal heart monitoring based on the intelligent model are implemented.

[0041] This embodiment includes the following advantages:

[0042] In an embodiment of the present invention, the fetal heart monitoring method based on the intelligent model includes: obtaining historical fetal position indicators, historical umbilical artery blood flow indicators, and historical fetal heart rate indicators of a specific group of users; obtaining historical adverse delivery outcome types of the specific group of users; associating historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators, and historical delivery outcome types as sample data; inputting historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators, and historical delivery outcome types into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model; inputting the fetal position indicator, umbilical artery blood flow indicator, and fetal heart rate indicator of the user to be detected into the trained intelligent machine learning model to obtain an output delivery outcome type; improving the monitoring accuracy and versatility of the fetal heart monitor, and solving the problem that the existing fetal heart monitoring AI model has a single learning sample data and inaccurate prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solution in the present embodiment, the following briefly introduces the drawings required for use in the description of the embodiment. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 is a flowchart of a fetal heart rate monitoring method based on an intelligent model according to the present embodiment;

[0045] Figure 2 is a system architecture diagram of an intelligent machine learning model of this embodiment;

[0046] Figure 3 is a structural block diagram of an embodiment of a fetal heart monitoring device based on an intelligent model of the present embodiment;

[0047] Figure 4 The diagram is a diagram of the internal structure of a computer device according to an embodiment. DETAILED DESCRIPTION

[0048] In order to make the technical problems, technical solutions and beneficial effects solved by this embodiment more clearly understood, this embodiment is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0049] In a core concept of an embodiment of the present invention, the fetal position, umbilical blood flow and fetal heart rate indicators of the infant and the corresponding delivery outcome types are used as the training set of the deep learning model to train the deep learning model; different from the general judgment method of judging the condition of the undelivered infant only by the fetal heart rate, the present application adds two important indicators: the fetal position and umbilical blood flow of the infant; the indication related to the umbilical blood flow can be used to characterize the hypoxia of the undelivered infant, and the fetal position related indicators can be conditions related to the labor process and asphyxia; the above two factors are combined with the fetal heart rate and then matched with the historical adverse delivery outcome types to obtain the training data of the deep learning model, thereby improving the prediction accuracy of the deep learning model, improving the monitoring accuracy and versatility of the fetal heart monitor, and solving the problem that the existing fetal heart monitoring AI model has a single learning sample data and inaccurate prediction results.

[0050] Reference Figure 1 , shows a flowchart of a fetal heart rate monitoring method based on an intelligent model according to the present embodiment, which may specifically include the following steps:

[0051] Step 101, obtaining historical fetal position indicators, historical umbilical artery blood flow indicators, and historical fetal heart rate indicators of users in a specific group;

[0052] In the embodiment of the present invention, the method can be applied to various different terminals, such as smart phones, laptops, desktop computers, computer equipment, etc. The embodiment of the present invention does not limit the specific type of the terminal. The operating system of the terminal may include Android, Harmony OS, IOS, Windows Phone, Windows, etc., and the present invention does not impose too many restrictions on this.

[0053] In the embodiment of the present invention, the specific group of users may refer to a group of users to be detected who have adverse delivery outcomes during pregnancy; specifically, the types of adverse delivery outcomes may include cesarean section, fetal distress, umbilical cord entanglement, oligohydramnios, meconium staining, intrauterine death, congenital fetal malformation, neonatal intrauterine asphyxia, etc. The embodiment of the present invention does not impose too many restrictions on the types of adverse delivery outcomes;

[0054] Regarding the method of obtaining the types of adverse birth outcomes, a corresponding second keyword can be preset in the database of the hospital information system (Hospital Information System), and multiple types of adverse birth outcomes can be identified from the diagnosis results of the patient medical records in the hospital database according to the second keyword; for example, the second keyword can be set to "entanglement," "deformity," "too little," "embarrassment," "contamination," "suffocation," etc., and the second keyword is identified in the electronic medical record system (EMR, Electronic Medical Record) in the hospital information system to obtain adverse birth outcomes and corresponding user identity information to be detected. The embodiment of the present invention does not impose too many restrictions on this. Further, the historical fetal position index, historical umbilical artery blood flow index, and historical fetal heart rate index of the specific group of users can be screened out according to the user identity information to be detected; specifically, after obtaining the user identity information to be detected corresponding to the adverse birth outcome, that is, determining that the user to be detected corresponding to the user identity information to be detected is a specific group user, and then filtering out the historical fetal position index, historical umbilical artery blood flow index, and historical fetal heart rate index of the specific group of users according to the user identity information to be detected; specifically, similarly The historical fetal position index, the historical umbilical artery blood flow index, and the historical fetal heart rate index can be identified in the above-mentioned electronic medical record system according to the set first keyword; for example, the first keyword can be set to "fetal position," "occipital," "chin," "sacral," "shoulder," "S / D value," "RI," "PI," "resistance index," "pulsatility index," and "heart rate," and the corresponding data can be extracted according to the above-mentioned settings of "fetal position," "occipital," "chin," "sacral," "shoulder," "S / D value," "RI," "PI," "resistance index," "pulsatility index," and "heart rate"; wherein the fetal position refers to the relationship between the indicator point of the fetal presenting part and the maternal pelvis;

[0055] In a specific embodiment, the obtaining of historical fetal position indicators, historical umbilical artery blood flow indicators, and fetal heart rate indicators of a specific group of users includes: obtaining raw medical data; identifying the raw medical data according to a preset first keyword, and extracting the historical fetal position indicators and historical umbilical artery blood flow indicators of the specific group of users;

[0056] Among them, the historical fetal position indicators include left occipital anterior position, left occipital transverse position, left occipital posterior position, right occipital anterior position, right occipital transverse position, right occipital posterior position, left chin anterior position, left chin transverse position, left chin posterior position, right chin anterior position, right chin transverse position, right chin posterior position, left sacral anterior position, left sacral transverse position, left sacral posterior position, right sacral anterior position, right sacral transverse position, right sacral posterior position, left anterior shoulder position, left posterior shoulder position, right anterior shoulder position, right posterior shoulder position, posterior occipital position, occipital transverse position, breech presentation, shoulder presentation and face presentation; of course, fetal positions divided by other methods may also be included; the embodiments of the present invention do not impose too many restrictions on the types.

[0057] In addition, the umbilical artery blood flow refers to the blood flow condition reflected by indicators such as the ratio of the highest blood flow velocity in the systolic period of the umbilical artery to the lowest blood flow velocity in the diastolic period, the yin force index and the pulsatility index. Checking the umbilical blood flow can determine the development of the fetus in the uterus; the historical umbilical artery blood flow indicators include the umbilical artery blood flow S / D value, the umbilical artery resistance index (RI), the umbilical artery pulsatility index (PI), etc., and the embodiments of the present invention do not impose too many restrictions on this.

[0058] Step 102, obtaining the historical adverse delivery outcome types of the specific group of users;

[0059] In an actual application of an embodiment of the present invention, obtaining the historical adverse birth outcome types of the specific group of users includes: obtaining medical raw data; identifying the medical raw data according to a preset second keyword, and extracting the historical adverse birth outcome types of the specific group of users; wherein the historical adverse birth outcome types include cesarean section, fetal distress, umbilical cord entanglement, oligohydramnios, meconium staining, intrauterine death, congenital fetal malformation, and neonatal intrauterine asphyxia.

[0060] The original medical data may refer to the data of the electronic medical record system in the hospital information system. The embodiment of the present invention does not impose excessive restrictions on this. Different types of adverse delivery outcomes are screened out by setting the second keyword to provide accurate sample data for model training; wherein the second keyword refers to keyword information containing type words of adverse delivery outcomes.

[0061] Step 103, associating the historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type as sample data;

[0062] In an embodiment of the present invention, the historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type are determined as sample data; it should be noted that the historical delivery outcome type may include the above-mentioned historical adverse delivery outcome type and the historical normal delivery outcome type; wherein the historical normal delivery outcome type is a delivery outcome in which there is no abnormality in the delivery process and the baby's post-delivery score is detected as normal; the historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the corresponding historical adverse delivery outcome type or historical normal delivery outcome type are determined as sample data, which ensures the comprehensiveness of the training set and improves the accuracy of model prediction.

[0063] Specifically, the historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical adverse delivery outcome types or the historical normal delivery outcome types are associated with the user information to be detected, and a model training set including the historical adverse delivery outcome types or the historical normal delivery outcome types and the corresponding heart rate and blood data is obtained.

[0064] For example, training set 1 is: "Pregnant woman A's fetal position before delivery: right occipital transverse position, umbilical artery blood flow S / D value: 2.67±0.44, RI value: 0.54±0.02, PI value: 1.72±0.06, the first continuously measured heart rate value (bpm):, the second continuously measured heart rate value (bpm): 122, 130, 125, 156, 155········147, 170, 192, 136, the second continuously measured heart rate value (bpm): 122, 130, 125, 156, 155········147, 133, 152, 136, historical delivery outcome type: neonatal intrauterine asphyxia;"

[0065] Training set 2 is: "Pregnant woman B's fetal position before delivery: occipital left posterior position, umbilical artery blood flow S / D value: 3.39±0.62, RI value: 0.52±0.03, PI value: 1.64±0.07, the first continuously measured heart rate value: , the second continuously measured heart rate value: 132, 130, 142, 150, 158·······132, 165, 122, 125, the second continuously measured heart rate value: 142, 160, 121, 133, 145·······134, 141, 158, 127, historical delivery outcome type: umbilical cord entanglement;"

[0066] Training set 3 is: "Pregnant woman C's fetal position before delivery: right posterior shoulder, umbilical artery blood flow S / D value: 3.08±0.35, RI value: 0.59±0.03, PI value: 1.56±0.07, the first continuously measured heart rate value: , the second continuously measured heart rate value: 155, 163, 143, 156, 165·······127, 145, 132, 147, the second continuously measured heart rate value: 133, 146, 143, 136, 153·······157, 123, 142, 144, historical delivery outcome type: fetal distress;"

[0067] Training set 4 is: "Pregnant woman D's fetal position before delivery: left anterior shoulder, umbilical artery blood flow S / D value: 2.89±0.5, RI value: 0.56±0.04, PI value: 1.8±0.07, the first continuous measurement of heart rate value: , the second continuous measurement of heart rate value: 160, 145, 143, 123, 134········133, 126, 128, 148, the second continuous measurement of heart rate value: 157, 138, 149, 122, 141········142, 153, 148, 156, historical delivery outcome type: oligohydramnios, meconium stain;"

[0068] Step 104, inputting the historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type into a preset intelligent machine learning model for training, to obtain a trained intelligent machine learning model;

[0069] After obtaining the sample data of the historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type, the historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type are input into the preset intelligent machine learning model for training.

[0070] In actual application to the embodiment of the present invention, the historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type are input into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model, including:

[0071] The preset intelligent machine learning model includes an agent module, an environment module, and a reward system module;

[0072] The historical fetal position index, historical umbilical artery blood flow index, and historical fetal heart rate index of the training set are used as the input layer, and the historical delivery outcome type is used as the output layer, and the input layer and the output layer are set in the environment module;

[0073] Observe the state value, action value, and reward value from the environment module and store the above parameters of the current cycle;

[0074] The agent module generates the next action value and inputs it into the environment module to obtain the output state value and reward value;

[0075] The environment module inputs the state value and reward value to the agent module to determine whether the first cycle is over. If so, it inputs the prediction result;

[0076] If the first cycle is not completed, the intelligent agent module is returned to the step of receiving the state value and generating the action value, and the prediction accuracy of the verification set is used as the evaluation value. If the prediction accuracy does not meet the requirements, the setting parameters of the model are optimized, and a new state value is generated, and the training step is returned.

[0077] like Figure 2 As shown, in the embodiment of the present invention, a deep reinforcement learning model is used as the intelligent machine learning model, and the intelligent machine learning model is trained using sample data to obtain a trained intelligent machine learning model; specifically, by setting the historical fetal position index, the historical umbilical artery blood flow index, and the historical fetal heart rate index as the input layer of the environment module, and setting the historical delivery outcome type as the input layer of the environment module, and then transmitting the state value and the reward value to the intelligent agent module, the intelligent agent module generates an action value to the environment module, and iterates back and forth to obtain a trained deep reinforcement learning model. The deep reinforcement learning model in the embodiment of the present invention uses a deep neural network to approximate the strategy or value function in reinforcement learning, thereby solving complex multi-factor decision-making probability problems.

[0078] A preferred embodiment of the embodiments of the present invention, the above-mentioned intelligent machine learning model is only an example of the embodiments of the present invention, and other deep learning neural network models can also be used to train the sample data such as the historical fetal position index, historical umbilical artery blood flow index, historical fetal heart rate index and historical delivery outcome type of the embodiments of the present invention, and the embodiments of the present invention do not impose too many restrictions on this; specifically, a supervised neural network model or an unsupervised neural network model can be used, for example, a supervised neural network model or an unsupervised neural network model can include a feedforward neural network, a convolutional neural network, a recurrent neural network, a self-attention network, a generative adversarial network, and an autoencoder, and the embodiments of the present invention do not impose too many restrictions on this.

[0079] The improvement on the intelligent machine learning model in the embodiment of the present invention is that the sample data: "historical fetal position index, historical umbilical artery blood flow index, historical fetal heart rate index and historical delivery outcome type" sets a conversion module, and the above data is dimensionally converted and directly input into the environment module, thereby improving conversion efficiency, eliminating the need for conversion within the module, saving model time, and improving training efficiency.

[0080] Regarding the optimization of the setting parameters of the above-mentioned model, the setting parameters may include the model learning rate setting, the number of hidden layers, the number of nodes in each hidden layer, etc. The embodiment of the present invention does not impose too many restrictions on this, and the optimization process of the setting parameters may be a process of iterative adjustment of the parameters.

[0081] Step 105, inputting the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected into the trained intelligent machine learning model to obtain the output delivery outcome type.

[0082] In an embodiment of the present invention, after obtaining the trained intelligent machine learning model, the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected are input into the trained intelligent machine learning model to obtain a predicted delivery outcome type;

[0083] For example, after inputting the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be tested: "fetal position: xxx, umbilical artery blood flow S / D value: xxx, RI value: xxx, PI value: xxx, first continuously measured heart rate value: xxx, second continuously measured heart rate value: xxx, second continuously measured heart rate value: xxx", the preset delivery outcome type can be output as an 80% probability of neonatal intrauterine asphyxia; or the preset delivery outcome type can be output as an 86% probability of umbilical cord entanglement.

[0084] In a specific embodiment, the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected are input into the trained intelligent machine learning model to obtain the output delivery outcome type, including:

[0085] Obtaining the initial fetal position index and initial umbilical artery blood flow index of the user to be detected;

[0086] Screening the initial fetal position index and the initial umbilical artery blood flow index of the user to be detected to obtain the fetal position index and the umbilical artery blood flow index;

[0087] The fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be tested are input into the trained intelligent machine learning model to obtain the output type of delivery outcome.

[0088] In an embodiment of the present invention, after obtaining the initial fetal position index and the initial umbilical artery blood flow index, it is necessary to pre-process the data to improve the model data training efficiency. Specifically, the data cleaning and data screening can be performed by setting a numerical range to obtain the fetal position index and the umbilical artery blood flow index; then the processed fetal position index and umbilical artery blood flow index, the fetal position index, umbilical artery blood flow index and the fetal heart rate index of the user to be detected are input into the trained intelligent machine learning model to obtain the output type of delivery outcome.

[0089] In an embodiment of the present invention, the fetal heart monitoring method based on the intelligent model includes: obtaining historical fetal position indicators, historical umbilical artery blood flow indicators, and historical fetal heart rate indicators of a specific group of users; obtaining historical adverse delivery outcome types of the specific group of users; associating historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators, and historical delivery outcome types as sample data; inputting historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators, and historical delivery outcome types into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model; inputting the fetal position indicator, umbilical artery blood flow indicator, and fetal heart rate indicator of the user to be detected into the trained intelligent machine learning model to obtain an output delivery outcome type; improving the monitoring accuracy and versatility of the fetal heart monitor, and solving the problem that the existing fetal heart monitoring AI model has a single learning sample data and inaccurate prediction results.

[0090] It should be noted that, for the method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that this embodiment is not limited by the order of the actions described, because according to this embodiment, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by this embodiment.

[0091] Reference Figure 3 , shows a structural block diagram of an embodiment of a fetal heart monitoring device based on an intelligent model of this embodiment, which may specifically include the following modules:

[0092] The historical data acquisition module 301 is used to acquire the historical fetal position index, historical umbilical artery blood flow index, and historical fetal heart rate index of a specific group of users;

[0093] Type data module 302, used to obtain the historical adverse delivery outcome types of the specific group of users;

[0094] An association module 303 is used to associate historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators and historical delivery outcome types as sample data;

[0095] A training module 304 is used to input historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators and historical delivery outcome types into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model;

[0096] The output module 305 is used to input the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected into the trained intelligent machine learning model to obtain the output delivery outcome type.

[0097] Preferably, the historical data acquisition module includes:

[0098] The first data acquisition submodule is used to acquire raw medical data;

[0099] a first identification submodule, for identifying the medical raw data according to a preset first keyword, and extracting historical fetal position indicators and historical umbilical artery blood flow indicators of the specific group of users; wherein the historical fetal position indicators include left occipital anterior position, left occipital transverse position, left occipital posterior position, right occipital anterior position, right occipital transverse position, right occipital posterior position, left anterior chin position, left transverse chin position, left posterior chin position, right anterior chin position, right transverse chin position, right posterior chin position, left anterior sacral position, left transverse sacral position, left posterior sacral position, right anterior sacral position, right transverse sacral position, right posterior sacral position, left anterior shoulder position, left posterior shoulder position, right anterior shoulder position, right posterior shoulder position, posterior occipital position, occipital transverse position, breech presentation, shoulder presentation and face presentation; and the historical umbilical artery blood flow indicators include umbilical artery blood flow S / D value, umbilical artery resistance index and umbilical artery pulsation index.

[0100] Preferably, the type data module includes:

[0101] The second data acquisition submodule is used to acquire raw medical data;

[0102] The second identification submodule is used to identify the medical raw data according to the preset second keyword, and extract the historical adverse delivery outcome types of the specific group of users; wherein the historical adverse delivery outcome types include cesarean section, fetal distress, umbilical cord entanglement, oligohydramnios, meconium staining, intrauterine death, congenital fetal malformation, and neonatal intrauterine asphyxia.

[0103] Preferably, the preset intelligent machine learning model includes an agent module, an environment module, and a reward system module;

[0104] The training module includes:

[0105] Setting a submodule, for dividing the historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type into a training set and a validation set; using the historical fetal position index, the historical umbilical artery blood flow index and the historical fetal heart rate index of the training set as an input layer, and the historical delivery outcome type as an output layer, and setting the input layer and the output layer in the environment module;

[0106] The storage submodule is used to observe the state value, action value, and reward value from the environment module and store the parameters of the current cycle;

[0107] The generation submodule is used for the agent module to generate the next action value and input it into the environment module to obtain the output state value and reward value;

[0108] The judgment submodule is used for the environment module to input the state value and reward value into the agent module to determine whether the first cycle is completed. If so, the prediction result is output;

[0109] The optimization submodule is used to return to the step of receiving the state value and generating the action value of the intelligent agent module if the first cycle has not ended, and use the prediction accuracy of the verification set as the evaluation value. If the prediction accuracy does not meet the requirements, the setting parameters of the model are optimized, and a new state value is generated, and the training step is returned to obtain the trained intelligent machine learning model.

[0110] Preferably, the output module comprises:

[0111] The third data acquisition submodule is used to obtain the initial fetal position index and the initial umbilical artery blood flow index of the user to be detected;

[0112] A screening submodule, used for screening the initial fetal position index and the initial umbilical artery blood flow index of the user to be detected to obtain the fetal position index and the umbilical artery blood flow index;

[0113] The output submodule is used to input the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected into the trained intelligent machine learning model to obtain the output delivery outcome type.

[0114] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0115] For the specific definition of the fetal heart monitoring device based on the intelligent model, please refer to the definition of the fetal heart monitoring method based on the intelligent model above, which will not be repeated here. Each module in the above-mentioned fetal heart monitoring device based on the intelligent model can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0116] The fetal heart monitoring device based on the intelligent model provided above can be used to execute the fetal heart monitoring method based on the intelligent model provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0117] In one embodiment, a computer device is provided, whose internal structure diagram can be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for simulating daylighting is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0118] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0119] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0120] Obtain historical fetal position indicators, historical umbilical artery blood flow indicators, and historical fetal heart rate indicators of specific group users;

[0121] Obtaining the historical adverse delivery outcome types of users in the specific group;

[0122] The historical fetal position index, historical umbilical artery blood flow index, historical fetal heart rate index and historical delivery outcome type were associated as sample data;

[0123] Inputting historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators, and historical delivery outcome types into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model;

[0124] The fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be tested are input into the trained intelligent machine learning model to obtain the output type of delivery outcome.

[0125] Preferably, the acquisition of historical fetal position indicators, historical umbilical artery blood flow indicators, and fetal heart rate indicators of users in a specific group includes:

[0126] Obtaining raw medical data;

[0127] The medical raw data is identified according to the preset first keyword, and the historical fetal position indicators and historical umbilical artery blood flow indicators of the specific group of users are extracted; wherein the historical fetal position indicators include left occipital anterior position, left occipital transverse position, left occipital posterior position, right occipital anterior position, right occipital transverse position, right occipital posterior position, left anterior chin position, left transverse chin position, left posterior chin position, right anterior chin position, right transverse chin position, right posterior chin position, left anterior sacral position, left transverse sacral position, left posterior sacral position, right anterior sacral position, right transverse sacral position, right posterior sacral position, left anterior shoulder position, left posterior shoulder position, right anterior shoulder position, right posterior shoulder position, posterior occipital position, occipital transverse position, breech presentation, shoulder presentation and face presentation; the historical umbilical artery blood flow indicators include umbilical artery blood flow S / D value, umbilical artery resistance index and umbilical artery pulsation index.

[0128] Preferably, the obtaining of the historical adverse delivery outcome types of the specific group of users includes:

[0129] Obtaining raw medical data;

[0130] The medical raw data is identified according to the preset second keyword, and the historical adverse delivery outcome types of the specific group of users are extracted; wherein the historical adverse delivery outcome types include cesarean section, fetal distress, umbilical cord entanglement, oligohydramnios, meconium staining, intrauterine death, congenital fetal malformation, and neonatal intrauterine asphyxia.

[0131] Preferably, the preset intelligent machine learning model includes an agent module, an environment module, and a reward system module;

[0132] The historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type are input into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model, including:

[0133] The historical fetal position index, historical umbilical artery blood flow index, historical fetal heart rate index and historical delivery outcome type are divided into a training set and a validation set; the historical fetal position index, historical umbilical artery blood flow index and historical fetal heart rate index of the training set are used as an input layer, and the historical delivery outcome type is used as an output layer, and the input layer and the output layer are set in the environment module;

[0134] Observe the state value, action value, and reward value from the environment module and store the parameters of the current cycle;

[0135] The agent module generates the next action value and inputs it into the environment module to obtain the output state value and reward value;

[0136] The environment module inputs the state value and reward value to the agent module to determine whether the first cycle is over. If so, it outputs the prediction result.

[0137] If the first cycle is not completed, the intelligent agent module is returned to the step of receiving the state value and generating the action value, and the prediction accuracy of the verification set is used as the evaluation value. If the prediction accuracy does not meet the requirements, the setting parameters of the model are optimized, and a new state value is generated. The training step is returned to obtain the trained intelligent machine learning model.

[0138] Preferably, the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected are input into the trained intelligent machine learning model to obtain the output type of delivery outcome, including:

[0139] Obtaining the initial fetal position index and initial umbilical artery blood flow index of the user to be detected;

[0140] Screening the initial fetal position index and the initial umbilical artery blood flow index of the user to be detected to obtain the fetal position index and the umbilical artery blood flow index;

[0141] The fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be tested are input into the trained intelligent machine learning model to obtain the output type of delivery outcome.

[0142] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented;

[0143] Obtain historical fetal position indicators, historical umbilical artery blood flow indicators, and historical fetal heart rate indicators of specific group users;

[0144] Obtaining the historical adverse delivery outcome types of users in the specific group;

[0145] The historical fetal position index, historical umbilical artery blood flow index, historical fetal heart rate index and historical delivery outcome type were associated as sample data;

[0146] Inputting historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators, and historical delivery outcome types into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model;

[0147] The fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be tested are input into the trained intelligent machine learning model to obtain the output type of delivery outcome.

[0148] Preferably, the acquisition of historical fetal position indicators, historical umbilical artery blood flow indicators, and fetal heart rate indicators of users in a specific group includes:

[0149] Obtaining raw medical data;

[0150] The medical raw data is identified according to the preset first keyword, and the historical fetal position indicators and historical umbilical artery blood flow indicators of the specific group of users are extracted; wherein the historical fetal position indicators include left occipital anterior position, left occipital transverse position, left occipital posterior position, right occipital anterior position, right occipital transverse position, right occipital posterior position, left anterior chin position, left transverse chin position, left posterior chin position, right anterior chin position, right transverse chin position, right posterior chin position, left anterior sacral position, left transverse sacral position, left posterior sacral position, right anterior sacral position, right transverse sacral position, right posterior sacral position, left anterior shoulder position, left posterior shoulder position, right anterior shoulder position, right posterior shoulder position, posterior occipital position, occipital transverse position, breech presentation, shoulder presentation and face presentation; the historical umbilical artery blood flow indicators include umbilical artery blood flow S / D value, umbilical artery resistance index and umbilical artery pulsation index.

[0151] Preferably, the obtaining of the historical adverse delivery outcome types of the specific group of users includes:

[0152] Obtaining raw medical data;

[0153] The medical raw data is identified according to the preset second keyword, and the historical adverse delivery outcome types of the specific group of users are extracted; wherein the historical adverse delivery outcome types include cesarean section, fetal distress, umbilical cord entanglement, oligohydramnios, meconium staining, intrauterine death, congenital fetal malformation, and neonatal intrauterine asphyxia.

[0154] Preferably, the preset intelligent machine learning model includes an agent module, an environment module, and a reward system module;

[0155] The historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type are input into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model, including:

[0156] The historical fetal position index, historical umbilical artery blood flow index, historical fetal heart rate index and historical delivery outcome type are divided into a training set and a validation set; the historical fetal position index, historical umbilical artery blood flow index and historical fetal heart rate index of the training set are used as an input layer, and the historical delivery outcome type is used as an output layer, and the input layer and the output layer are set in the environment module;

[0157] Observe the state value, action value, and reward value from the environment module and store the parameters of the current cycle;

[0158] The agent module generates the next action value and inputs it into the environment module to obtain the output state value and reward value;

[0159] The environment module inputs the state value and reward value to the agent module to determine whether the first cycle is over. If so, it outputs the prediction result.

[0160] If the first cycle is not completed, the intelligent agent module is returned to the step of receiving the state value and generating the action value, and the prediction accuracy of the verification set is used as the evaluation value. If the prediction accuracy does not meet the requirements, the setting parameters of the model are optimized, and a new state value is generated. The training step is returned to obtain the trained intelligent machine learning model.

[0161] Preferably, the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected are input into the trained intelligent machine learning model to obtain the output type of delivery outcome, including:

[0162] Obtaining the initial fetal position index and initial umbilical artery blood flow index of the user to be detected;

[0163] Screening the initial fetal position index and the initial umbilical artery blood flow index of the user to be detected to obtain the fetal position index and the umbilical artery blood flow index;

[0164] The fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be tested are input into the trained intelligent machine learning model to obtain the output type of delivery outcome.

[0165] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0166] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0167] Those skilled in the art will appreciate that the embodiments of the present embodiment may be provided as methods, devices, or computer program products. Therefore, the present embodiment may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present embodiment may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0168] The present embodiment is described with reference to the flowchart and / or block diagram of the apparatus, terminal device (system), and computer program product according to the present embodiment. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0169] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction method, which is implemented in the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0170] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0171] Although the preferred embodiment of the present embodiment has been described, other changes and modifications may be made to these embodiments once those skilled in the art have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiment and all changes and modifications falling within the scope of the present embodiment.

[0172] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, device, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a..." do not exclude the existence of other identical elements in the process, device, article or terminal device including the elements.

[0173] The above is a detailed introduction to a fetal heart monitoring method based on an intelligent model and a fetal heart monitoring device based on an intelligent model, a computer device and a storage medium provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A fetal heart rate monitoring method based on an intelligent model, characterized in that: include: Obtain historical fetal position indicators, historical umbilical artery blood flow indicators, and historical fetal heart rate indicators of specific group users; Obtaining the historical adverse delivery outcome types of users in the specific group; The historical fetal position index, historical umbilical artery blood flow index, historical fetal heart rate index and historical delivery outcome type were associated as sample data; Inputting historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators, and historical delivery outcome types into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model; The fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be tested are input into the trained intelligent machine learning model to obtain the output type of delivery outcome.

2. The method according to claim 1, characterized in that The acquisition of historical fetal position indicators, historical umbilical artery blood flow indicators, and fetal heart rate indicators of users in a specific group includes: Obtaining raw medical data; The medical raw data is identified according to the preset first keyword, and the historical fetal position indicators and historical umbilical artery blood flow indicators of the specific group of users are extracted; wherein the historical fetal position indicators include left occipital anterior position, left occipital transverse position, left occipital posterior position, right occipital anterior position, right occipital transverse position, right occipital posterior position, left anterior chin position, left transverse chin position, left posterior chin position, right anterior chin position, right transverse chin position, right posterior chin position, left anterior sacral position, left transverse sacral position, left posterior sacral position, right anterior sacral position, right transverse sacral position, right posterior sacral position, left anterior shoulder position, left posterior shoulder position, right anterior shoulder position, right posterior shoulder position, posterior occipital position, occipital transverse position, breech presentation, shoulder presentation and face presentation; the historical umbilical artery blood flow indicators include umbilical artery blood flow S / D value, umbilical artery resistance index and umbilical artery pulsation index.

3. The method according to claim 2, characterized in that The acquired historical adverse delivery outcome types of the specific group of users include: Obtaining raw medical data; The medical raw data is identified according to the preset second keyword, and the historical adverse delivery outcome types of the specific group of users are extracted; wherein the historical adverse delivery outcome types include cesarean section, fetal distress, umbilical cord entanglement, oligohydramnios, meconium staining, intrauterine death, congenital fetal malformation, and neonatal intrauterine asphyxia.

4. The method according to claim 2, characterized in that: The preset intelligent machine learning model includes an agent module, an environment module, and a reward system module; The historical fetal position index, the historical umbilical artery blood flow index, the historical fetal heart rate index and the historical delivery outcome type are input into a preset intelligent machine learning model for training to obtain a trained intelligent machine learning model, including: The historical fetal position index, historical umbilical artery blood flow index, historical fetal heart rate index and historical delivery outcome type are divided into a training set and a validation set; the historical fetal position index, historical umbilical artery blood flow index and historical fetal heart rate index of the training set are used as an input layer, and the historical delivery outcome type is used as an output layer, and the input layer and the output layer are set in the environment module; Observe the state value, action value, and reward value from the environment module and store the parameters of the current cycle; The agent module generates the next action value and inputs it into the environment module to obtain the output state value and reward value; The environment module inputs the state value and reward value to the agent module to determine whether the first cycle is over. If so, it outputs the prediction result. If the first cycle is not completed, the intelligent agent module is returned to the step of receiving the state value and generating the action value, and the prediction accuracy of the verification set is used as the evaluation value. If the prediction accuracy does not meet the requirements, the setting parameters of the model are optimized, and a new state value is generated. The training step is returned to obtain the trained intelligent machine learning model.

5. The method according to claim 2, characterized in that: The fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected are input into the trained intelligent machine learning model to obtain the output delivery outcome type, including: Obtaining the initial fetal position index and initial umbilical artery blood flow index of the user to be detected; Screening the initial fetal position index and the initial umbilical artery blood flow index of the user to be detected to obtain the fetal position index and the umbilical artery blood flow index; The fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be tested are input into the trained intelligent machine learning model to obtain the output type of delivery outcome.

6. A fetal heart rate monitoring device based on an intelligent model, characterized in that: include: A historical data acquisition module, used to obtain historical fetal position indicators, historical umbilical artery blood flow indicators, and historical fetal heart rate indicators of a specific group of users; A type data module, used to obtain the historical adverse delivery outcome types of the specific group of users; A correlation module, used to correlate historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators and historical delivery outcome types as sample data; A training module, used for inputting historical fetal position indicators, historical umbilical artery blood flow indicators, historical fetal heart rate indicators and historical delivery outcome types into a preset intelligent machine learning model for training, thereby obtaining a trained intelligent machine learning model; The output module is used to input the fetal position index, umbilical artery blood flow index, and fetal heart rate index of the user to be detected into the trained intelligent machine learning model to obtain the output delivery outcome type.

7. The device according to claim 6, characterized in that The historical data acquisition module includes: The first data acquisition submodule is used to acquire raw medical data; a first identification submodule, for identifying the medical raw data according to a preset first keyword, and extracting historical fetal position indicators and historical umbilical artery blood flow indicators of the specific group of users; wherein the historical fetal position indicators include left occipital anterior position, left occipital transverse position, left occipital posterior position, right occipital anterior position, right occipital transverse position, right occipital posterior position, left anterior chin position, left transverse chin position, left posterior chin position, right anterior chin position, right transverse chin position, right posterior chin position, left anterior sacral position, left transverse sacral position, left posterior sacral position, right anterior sacral position, right transverse sacral position, right posterior sacral position, left anterior shoulder position, left posterior shoulder position, right anterior shoulder position, right posterior shoulder position, posterior occipital position, occipital transverse position, breech presentation, shoulder presentation and face presentation; and the historical umbilical artery blood flow indicators include umbilical artery blood flow S / D value, umbilical artery resistance index and umbilical artery pulsation index.

8. The device according to claim 7, characterized in that The type data module includes: The second data acquisition submodule is used to acquire raw medical data; The second identification submodule is used to identify the medical raw data according to the preset second keyword, and extract the historical adverse delivery outcome types of the specific group of users; wherein the historical adverse delivery outcome types include cesarean section, fetal distress, umbilical cord entanglement, oligohydramnios, meconium staining, intrauterine death, congenital fetal malformation, and neonatal intrauterine asphyxia.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the fetal heart monitoring method based on the intelligent model described in any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the fetal heart monitoring method based on an intelligent model as described in any one of claims 1 to 6 are implemented.

Images