Perinatal maternal and infant health monitoring method and system

Abstract

The invention provides a perinatal maternal and infant health monitoring method and system. A perinatal maternal and fetal systolic pressure-heart rate correlation model is established according to historical systolic pressure data and historical heart rate data; predicting heart rate predicted values corresponding to the systolic pressure of the target puerpera at different monitoring moments through a systolic pressure-heart rate association model, and further determining the physiological index association deviation degree of the target puerpera and the fetus at different monitoring moments; for each monitoring moment, if the physiological index correlation deviation degree is greater than a preset correlation deviation threshold value, correcting the actual heart rate value of the target fetus at the corresponding monitoring moment according to the body position data of the target puerpera at the corresponding monitoring moment and the fetal position data of the target fetus; and extracting a maternal and infant physiological collaborative health index of the target puerpera and the fetus, and pushing an abnormality monitoring report to a medical care terminal based on the maternal and infant physiological collaborative health index. By adopting the scheme of the invention, the maternal and infant physiological association can be quantified so as to improve the perinatal maternal and infant potential health risk identification capability.

Description

Technical Field

[0001] This application relates to the field of maternal and infant health monitoring technology, and in particular to a perinatal maternal and infant health monitoring method and system. Background Technology

[0002] Perinatal maternal and infant health monitoring is crucial, as its results directly affect the life, health, and subsequent development of both mother and child. Traditional methods of perinatal maternal and infant health monitoring often focus on monitoring the physiological indicators of the mother and fetus separately. This approach can only obtain physiological data of a single individual, completely ignoring the close physiological relationship between the mother and fetus. In fact, changes in the mother's physical condition, such as abnormal fluctuations in blood pressure, are very likely to affect the fetal heart rate.

[0003] With the advancement of technology, existing monitoring methods have achieved technological upgrades to some extent, with some adopting electronic devices for data collection, such as ultrasonic Doppler fetal heart rate monitors for monitoring fetal heart rate and electronic blood pressure monitors for measuring maternal blood pressure. These methods can also digitally record and store the data. However, these methods still have significant shortcomings. They fail to deeply explore and quantify the correlation and deviation between maternal and fetal physiological indicators. In actual monitoring, faced with a large amount of complex and fluctuating data, relying solely on fixed thresholds for judgment can lead to difficulties in accurately identifying abnormal data, resulting in a lack of precision in assessing maternal and infant health. Therefore, how to quantify the physiological correlation between mother and child to improve the ability to identify potential health risks to mothers and children during the perinatal period has become a challenge for the industry. Summary of the Invention

[0004] Based on this, this application provides a perinatal maternal and infant health monitoring method and system that quantifies the physiological correlation between mother and child to improve the ability to identify potential health risks to mothers and children during the perinatal period.

[0005] Firstly, this application provides a method for perinatal maternal and infant health monitoring, comprising the following steps: Collect historical systolic blood pressure data of different healthy mothers during the perinatal period and historical heart rate data of corresponding healthy fetuses; A systolic blood pressure-heart rate correlation model between perinatal mothers and fetuses was established based on all historical systolic blood pressure data and all historical heart rate data. The systolic blood pressure and body position characteristics of the target pregnant woman at different monitoring times during the perinatal period, as well as the actual heart rate and fetal position characteristics of the target fetus, are obtained. The heart rate prediction value corresponding to the systolic blood pressure of the target pregnant woman at different monitoring times is predicted by the systolic blood pressure-heart rate correlation model. Based on each heart rate prediction value and the actual heart rate value of the target fetus at the corresponding monitoring time, the correlation deviation degree of the physiological indicators between the target pregnant woman and the fetus at different monitoring times is determined. For each monitoring time, if the correlation deviation of physiological indicators is greater than the preset correlation deviation threshold, the actual heart rate value of the target fetus at the corresponding monitoring time is corrected based on the position data of the target mother and the fetal position data of the target fetus at the corresponding monitoring time. The maternal and infant physiological health index of the target mother and fetus is extracted from all corrected actual heart rate values, all systolic blood pressure, all body position characteristics and all fetal position characteristics, and then abnormal monitoring reports are pushed to the medical terminal based on the maternal and infant physiological health index.

[0006] In some embodiments, establishing a perinatal maternal-fetal systolic blood pressure-heart rate correlation model based on all historical systolic blood pressure data and all historical heart rate data specifically includes: Data cleaning was performed on all historical systolic blood pressure data and all historical heart rate data to obtain cleaned historical systolic blood pressure dataset and historical heart rate dataset. The cleaned historical systolic blood pressure dataset and historical heart rate dataset are divided into training set, validation set and test set; The association prediction framework based on multiple linear regression and gradient boosting tree is trained using the training set, the validation set, and the test set. Historical systolic blood pressure is used as the input feature, and historical heart rate is used as the output feature to obtain the systolic blood pressure-heart rate association model.

[0007] In some embodiments, predicting the predicted heart rate value corresponding to the systolic blood pressure at different monitoring times for the target parturient using the systolic blood pressure-heart rate correlation model specifically includes: Obtain gestational week information corresponding to systolic blood pressure at different monitoring times for the target pregnant woman; Based on the information of each gestational week, the corresponding time-segmented prediction parameters in the systolic blood pressure-heart rate correlation model are matched; The systolic blood pressure at different monitoring times is input into the matched systolic blood pressure-heart rate correlation model, and the predicted heart rate value corresponding to different monitoring times is output.

[0008] In some embodiments, determining the correlation deviation between the physiological indicators of the target mother and the fetus at different monitoring times based on each predicted heart rate value and the actual heart rate value of the target fetus at the corresponding monitoring time specifically includes: For each monitoring time, determine the allowable error of heart rate at that time. Determine the heart rate difference between the predicted heart rate at the monitoring time and the actual heart rate corresponding to the target fetus; Based on the allowable error of heart rate and the heart rate difference, the correlation deviation of physiological indicators between the target mother and the fetus at the monitoring time is determined, and then the correlation deviation of physiological indicators between the target mother and the fetus at different monitoring times is obtained.

[0009] In some embodiments, correcting the actual heart rate value of the target fetus at the corresponding monitoring time based on the maternal position data and fetal position data of the target fetus at the corresponding monitoring time specifically includes: Based on the target mother's body position data, the target fetus's fetal position data, the target mother's systolic blood pressure fluctuation range, and the target mother's gestational age at the corresponding monitoring time, determine the type of dominant deviation factor at the corresponding monitoring time. The comprehensive correction coefficient for the actual heart rate value is determined based on the gestational age of the target mother, the duration of the target mother's position, the fluctuation range of the target mother's systolic blood pressure, and the fetal position type of the target fetus. According to the priority order of the dominant deviation factors and the comprehensive correction coefficient, the actual heart rate value of the target fetus at the monitoring time is subjected to hierarchical progressive correction to obtain the preliminary corrected heart rate value. The initial corrected heart rate value is validated in two dimensions to obtain the actual heart rate value of the target fetus after correction at the corresponding monitoring time.

[0010] In some embodiments, extracting the maternal-infant physiological synergistic health index from all corrected actual heart rate values, all systolic blood pressure values, all postural characteristics, and all fetal position characteristics specifically includes: Extract the systolic blood pressure-fetal heart rate matching degree between the target mother and the fetus from all corrected actual heart rate values ​​and all systolic blood pressure values; The position-fetal position compatibility coefficient between the target mother and the fetus is extracted from all positional characteristics and all fetal position characteristics. The systolic blood pressure-fetal heart rate matching degree and the body position-fetal position adaptation coefficient are synergistically fused to obtain the maternal and infant physiological synergistic health index of the target mother and fetus.

[0011] In some embodiments, if the correlation deviation of physiological indicators is less than or equal to a preset correlation deviation threshold, no processing is performed.

[0012] Secondly, this application provides a perinatal maternal and infant health monitoring system, comprising: The data acquisition module is used to collect historical systolic blood pressure data of different healthy mothers during the perinatal period and historical heart rate data of the corresponding healthy fetuses. The processing module is used to build a systolic blood pressure-heart rate correlation model between the perinatal mother and the fetus based on all historical systolic blood pressure data and all historical heart rate data. The processing module is also used to acquire the systolic blood pressure and body position characteristics of the target mother at different monitoring times during the perinatal period, the actual heart rate value and fetal position characteristics of the target fetus, and predict the heart rate prediction value corresponding to the systolic blood pressure of the target mother at different monitoring times through the systolic blood pressure-heart rate correlation model. Based on each heart rate prediction value and the actual heart rate value of the target fetus at the corresponding monitoring time, the correlation deviation degree of the physiological indicators between the target mother and the fetus at different monitoring times is determined. The processing module is also used to, for each monitoring time, if the correlation deviation of physiological indicators is greater than a preset correlation deviation threshold, correct the actual heart rate value of the target fetus at the corresponding monitoring time based on the position data of the target mother and the fetal position data of the target fetus at the corresponding monitoring time. The execution module is used to extract the maternal and infant physiological health index of the target mother and fetus from all corrected actual heart rate values, all systolic blood pressure, all body position characteristics and all fetal position characteristics, and then push abnormal monitoring reports to the medical terminal based on the maternal and infant physiological health index.

[0013] Thirdly, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the perinatal maternal and infant health monitoring method described above.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the perinatal maternal and infant health monitoring method described above.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The perinatal maternal and infant health monitoring method and system provided in this application first collects historical systolic blood pressure data of different healthy mothers and corresponding historical heart rate data of healthy fetuses during the perinatal period; based on all historical systolic blood pressure data and all historical heart rate data, a systolic blood pressure-heart rate correlation model between the perinatal mother and fetus is established; the systolic blood pressure and body position characteristics of the target mother at different monitoring times, and the actual heart rate value and fetal position characteristics of the target fetus are obtained; the predicted heart rate value corresponding to the systolic blood pressure of the target mother at different monitoring times is predicted through the systolic blood pressure-heart rate correlation model; and based on each predicted heart rate value and the corresponding monitoring time, the predicted heart rate value is calculated. The actual heart rate value of the target fetus is used to determine the correlation deviation of physiological indicators between the target mother and the fetus at different monitoring times. For each monitoring time, if the correlation deviation of physiological indicators is greater than the preset correlation deviation threshold, the actual heart rate value of the target fetus at the corresponding monitoring time is corrected based on the body position data of the target mother and the fetal position data of the target fetus at the corresponding monitoring time. From all corrected actual heart rate values, all systolic blood pressure, all body position characteristics and all fetal position characteristics, the maternal and infant physiological synergy health index of the target mother and fetus is extracted, and then an abnormal monitoring report is pushed to the medical terminal based on the maternal and infant physiological synergy health index.

[0016] Therefore, this application first establishes a systolic blood pressure-heart rate correlation model between the perinatal mother and fetus based on all historical systolic blood pressure and heart rate data. This step enables accurate prediction of fetal heart rate at different time periods, conforming to the physiological patterns of different gestational weeks, thereby improving the accuracy of the correlation prediction of maternal and fetal physiological indicators. Secondly, the systolic blood pressure-heart rate correlation model predicts the predicted heart rate values ​​corresponding to the systolic blood pressure of the target mother at different monitoring times. Based on each predicted heart rate value and the actual heart rate value of the target fetus at the corresponding monitoring time, the correlation deviation of physiological indicators between the target mother and fetus at different monitoring times is determined. This correlation deviation is an indicator that quantifies the difference between the predicted fetal heart rate value corresponding to the perinatal mother's systolic blood pressure and the actual fetal heart rate value at the same monitoring time. Its core function is to serve as a preliminary basis for judging whether the correlation of "mother-fetus" physiological indicators is abnormal. Through this indicator, deviations between the actual fetal heart rate and the expected correlation pattern can be quickly identified. This triggers subsequent "heart rate correction based on body position / fetal position" operations, avoiding health risks caused by untimely handling of abnormal indicator correlations. Then, the maternal-infant physiological synergy health index is extracted from all corrected actual heart rate values, all systolic blood pressure, all body position characteristics, and all fetal position characteristics. This maternal-infant physiological synergy health index comprehensively reflects the quantitative indicators of the synergy and stability of the physiological state of the target mother and fetus during the perinatal period (28 weeks of gestation to delivery). Its core function is to serve as the core basis for judging whether the overall health status of the mother and baby is normal. By integrating two key dimensions, "matching of maternal systolic blood pressure and fetal heart rate" and "adaptability of maternal body position and fetal position," it intuitively reflects the synergistic pattern between maternal and infant physiological indicators. Finally, based on the maternal-infant physiological synergy health index, an abnormal monitoring report is pushed to the medical terminal. In summary, the solution of this application can quantify the physiological correlation between mother and baby to improve the ability to identify potential health risks of mother and baby during the perinatal period. Attached Figure Description

[0017] Figure 1 This is an exemplary flowchart of a perinatal maternal and infant health monitoring method according to some embodiments of this application; Figure 2 This is a schematic diagram illustrating an application scenario of a perinatal maternal and infant health monitoring system according to some embodiments of this application; Figure 3 This is a schematic flowchart illustrating the process of correcting the actual heart rate value according to some embodiments of this application; Figure 4 This is a schematic diagram of the perinatal maternal and infant health monitoring system according to some embodiments of this application; Figure 5 This is a schematic diagram of the structure of a computer device for implementing a perinatal maternal and infant health monitoring method according to some embodiments of this application. Detailed Implementation

[0018] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0019] refer to Figure 1 The figure is an exemplary flowchart of a perinatal maternal and infant health monitoring method according to some embodiments of this application. The perinatal maternal and infant health monitoring method mainly includes the following steps: In step 101, historical systolic blood pressure data of different healthy mothers during the perinatal period and historical heart rate data of the corresponding healthy fetuses are collected.

[0020] In specific implementation, the historical systolic blood pressure data of different healthy mothers and the historical heart rate data of corresponding healthy fetuses during the perinatal period can be collected in the following way: the historical systolic blood pressure data of different healthy mothers and the historical heart rate data of corresponding healthy fetuses during the perinatal period can be collected from the hospital's electronic medical record database. The electronic medical record database records various test data of different healthy mothers and corresponding fetuses during the physiological index testing in history, including: the systolic blood pressure of healthy mothers and the heart rate test values ​​of corresponding fetuses. In other embodiments, the historical systolic blood pressure data of different healthy mothers and the historical heart rate data of corresponding healthy fetuses during the perinatal period can also be collected by other methods, which are not limited here.

[0021] It should be noted that the perinatal period refers to a specific stage surrounding delivery, including the critical periods of pregnancy, delivery, and postpartum. In this application, maternal and infant health monitoring is conducted during the perinatal period of the pregnant woman's pregnancy (from 28 weeks of gestation to delivery).

[0022] In addition, it should be noted that the healthy pregnant women mentioned in this application must exclude individuals with gestational hypertension (systolic blood pressure ≥140 mmHg), gestational diabetes (fasting blood glucose ≥5.1 mmol / L), or congenital heart disease; the healthy fetuses must exclude those with intrauterine distress (fetal heart rate baseline >160 beats / min or <110 beats / min for more than 10 minutes) or fetal growth restriction (estimated weight below the 10th percentile for the same gestational age). The screening criteria must be independently reviewed and confirmed by two senior obstetricians (consistency ≥95%).

[0023] In some embodiments, reference Figure 2As shown in the figure, this diagram illustrates an application scenario of the perinatal maternal and infant health monitoring system according to some embodiments of this application. The figure includes three main components: a data acquisition device, a server, and a data storage device. The data acquisition device is responsible for collecting historical systolic blood pressure data of different healthy mothers during the perinatal period and historical heart rate data of corresponding healthy fetuses, systolic blood pressure and body position characteristics of the target mother at different monitoring times, and actual heart rate and fetal position characteristics of the target fetus. The collected historical systolic blood pressure data, historical heart rate data, all systolic blood pressure, all body position characteristics, all actual heart rate values, and all fetal position characteristics are sent to the server through a communication network. The perinatal maternal and infant health monitoring system runs on the server, which stores the processing results in the data storage device and visualizes them.

[0024] In step 102, a systolic blood pressure-heart rate correlation model between the perinatal mother and the fetus is established based on all historical systolic blood pressure data and all historical heart rate data.

[0025] In some embodiments, establishing a perinatal maternal-fetal systolic blood pressure-heart rate correlation model based on all historical systolic blood pressure data and all historical heart rate data can be achieved by the following steps: Data cleaning was performed on all historical systolic blood pressure data and all historical heart rate data to obtain cleaned historical systolic blood pressure dataset and historical heart rate dataset. The cleaned historical systolic blood pressure dataset and historical heart rate dataset are divided into training set, validation set and test set; The association prediction framework based on multiple linear regression and gradient boosting tree is trained using the training set, the validation set, and the test set. Historical systolic blood pressure is used as the input feature, and historical heart rate is used as the output feature to obtain the systolic blood pressure-heart rate association model.

[0026] In specific implementation, all historical systolic blood pressure data and all historical heart rate data are cleaned to obtain the cleaned historical systolic blood pressure dataset and historical heart rate dataset. This can be achieved by removing samples with systolic blood pressure deviations of 80-140 mmHg, fetal heart rate deviations of 110-160 beats / min, and data missing of more than 3% in a single time period. Other methods can also be used in other embodiments, which are not limited here.

[0027] In specific implementation, the cleaned historical systolic blood pressure dataset and historical heart rate dataset can be divided into training set, validation set and test set in the following way: the training set, validation set and test set can be divided in a ratio of 7:2:1. In other embodiments, other methods can also be used to divide them, which are not limited here.

[0028] In specific implementation, the correlation prediction framework constructed based on multiple linear regression and gradient boosting tree is trained according to the training set, the validation set, and the test set. Historical systolic blood pressure is used as the input feature, and historical heart rate is used as the output feature. The systolic blood pressure-heart rate correlation model can be implemented as follows: Based on the training set, the validation set, and the test set, the correlation prediction framework is constructed using the "multiple linear regression + gradient boosting tree" algorithm, with historical systolic blood pressure as the input feature and historical heart rate as the output feature. Multiple linear regression initially fits the linear correlation between systolic blood pressure and heart rate. The training set is input into the correlation prediction framework for iterative training. The validation set is used to adjust hyperparameters such as the tree depth (3-8 layers) and the number of iterations (100-200 times) of the gradient boosting tree. The model performance is verified using the test set (i.e., ensuring the model meets the requirements of root mean square error ≤ 5 Hz and coefficient of determination ≥ 0.85). The optimal parameters are then integrated to form the systolic blood pressure-heart rate correlation model.

[0029] It should be noted that, as a preferred embodiment, to improve the prediction accuracy of the model at different gestational weeks, this application can further subdivide and clean the historical systolic blood pressure data of healthy mothers and the corresponding historical fetal heart rate data into three sub-periods according to gestational week: 28-32 weeks of gestation, 33-36 weeks of gestation, and 37 weeks of gestation to delivery. The specific gestational week (accurate to the day) and daily monitoring time (8-12 am, 12-4 pm, 4-8 pm, 8 pm to 8 am the next day) of each data collection are supplemented. Samples with systolic blood pressure deviations of 80-140 mmHg, fetal heart rate deviations of 110-160 beats / min, and data missing in a single time period exceeding 3% are removed. The training set, validation set, and test set are then divided in a 7:2:1 ratio for each sub-period. Based on the datasets after the sub-period division, the "multiple linear regression + gradient boosting tree" algorithm is used to construct a time-segmented correlation prediction framework, which will correlate the mother's systolic blood pressure and specific fetal heart rate data within each sub-period. Gestational age and monitoring time period are used as input features, and the corresponding fetal heart rate baseline value (5-minute mean heart rate ± 2 beats / min) for each sub-time period is used as output features. Multiple linear regression initially fits the linear relationship between systolic blood pressure and heart rate, while gradient boosting tree optimizes the non-linear impact of different gestational ages and monitoring time periods on heart rate. The training sets of each sub-time period are input into the corresponding prediction framework for iterative training. The hyperparameters such as the tree depth (3-8 layers) and number of iterations (100-200 times) of the gradient boosting tree are adjusted using the validation sets of each sub-time period. The model performance is verified using the test sets of each sub-time period (ensuring that the model for each sub-time period meets the requirements of root mean square error ≤ 5 beats / min and coefficient of determination ≥ 0.85). The optimal parameters of each sub-time period are integrated to form a systolic blood pressure-heart rate correlation model that can be called in different time periods, ensuring that the predicted heart rate value can be matched to the sub-time period of the target pregnant woman's current gestational age during subsequent real-time monitoring. Other methods can also be used in other embodiments, which will not be elaborated here.

[0030] In step 103, the systolic blood pressure and body position characteristics of the target pregnant woman at different monitoring times during the perinatal period, the actual heart rate value and fetal position characteristics of the target fetus, and the predicted heart rate value corresponding to the systolic blood pressure of the target pregnant woman at different monitoring times are obtained through the systolic blood pressure-heart rate correlation model. Based on each predicted heart rate value and the actual heart rate value of the target fetus at the corresponding monitoring time, the correlation deviation degree of the physiological indicators between the target pregnant woman and the fetus at different monitoring times is determined.

[0031] In practice, obtaining systolic blood pressure and body position characteristics of the target pregnant woman at different monitoring times during the perinatal period, as well as the actual heart rate and fetal position characteristics of the target fetus, can be achieved in the following way: First, a combination of medical-grade monitoring equipment can be provided to the target pregnant woman, including an upper arm cuff blood pressure monitor (accuracy ±2mmHg), a wearable body position sensor (sampling frequency 10Hz), and an ultrasonic Doppler fetal heart rate monitor (sampling rate ≥100Hz). All devices are synchronized with timestamps (error ≤1 second) to ensure the correlation of data at the same monitoring time. Second, the monitoring frequency is set to collect data once every 10 minutes. During data collection, the mother must be at rest. Systolic blood pressure is collected using an upper arm cuff blood pressure monitor (two consecutive measurements are taken and the average is recorded; the cuff size matches the upper arm circumference). Body position is identified by a wearable position sensor into four categories: left lateral decubitus, right lateral decubitus, supine, and sitting. (The sensor's X-axis tilt angle is 30°-60° for left lateral decubitus, -30° to -60° for right lateral decubitus, 0°±10° for supine, and 80°-100° for sitting; the position is automatically categorized based on preset angle thresholds). The fetal heart rate is measured using an ultrasound Doppler fetal heart rate monitor probe. Fetal heart sounds were collected after locating them on the abdominal wall (each session lasted 30 seconds, and the average value was taken as the heart rate value at that moment; if the signal was interrupted, medical staff were alerted in real time to relocate the heart sound). Next, fetal position characteristics were obtained using a combination of real-time palpation and daily ultrasound confirmation. At the monitoring time, obstetric medical staff made a preliminary assessment of fetal position (cephalic / breech / transverse) through abdominal palpation. At fixed times each day, ultrasound examinations were used to precisely record fetal position details (such as whether the cephalic position was occiput anterior). The palpation results and ultrasound results were then linked and stored. Finally, all collected data were uploaded to a central database in real time, with each data entry accompanied by the "monitoring time". The unique identifier is "interval stamp + maternal ID + device number". The systolic blood pressure and heart rate data are automatically marked with numerical ranges (systolic blood pressure 80-140 mmHg, heart rate 110-160 beats / min). When the data exceeds the range, a data anomaly alert is triggered to ensure that the basic data used for subsequent model prediction and deviation calculation is accurate and complete, and is consistent with historical data collection standards (such as systolic blood pressure measurement accuracy and heart rate collection duration) to avoid affecting the applicability of the associated model due to differences in data specifications. It should be noted that the collection frequency in this application can also be adjusted according to the actual situation, and is not limited here.

[0032] In some embodiments, predicting the predicted heart rate value corresponding to the systolic blood pressure at different monitoring times for the target parturient using the systolic blood pressure-heart rate correlation model can be achieved through the following steps: Obtain gestational week information corresponding to systolic blood pressure at different monitoring times for the target pregnant woman; Based on the information of each gestational week, the corresponding time-segmented prediction parameters in the systolic blood pressure-heart rate correlation model are matched; The systolic blood pressure at different monitoring times is input into the matched systolic blood pressure-heart rate correlation model, and the predicted heart rate value corresponding to different monitoring times is output.

[0033] In specific implementation, obtaining the gestational age information corresponding to the systolic blood pressure at different monitoring times for the target pregnant woman can be achieved in the following way: the gestational age information corresponding to the systolic blood pressure at different monitoring times for the target pregnant woman can be retrieved from the hospital's obstetrics information database, and the gestational age information and the systolic blood pressure data at the corresponding monitoring time can be associated and stored with a unique identifier (such as "pregnant woman ID-monitoring timestamp") to ensure that the systolic blood pressure at each monitoring time corresponds one-to-one with the gestational age information; in other embodiments, the gestational age information can also be manually entered by medical staff for supplementary verification, and this application does not limit this.

[0034] In specific implementation, matching the corresponding time-segmented prediction parameters in the systolic blood pressure-heart rate correlation model based on gestational week information can be achieved in the following way: For gestational week information at each monitoring time, the corresponding parameter subset is located through interval matching logic (e.g., 30 weeks of gestation corresponds to the subset of 28-32 weeks of gestation, and 35 weeks of gestation corresponds to the subset of 33-36 weeks of gestation), and this subset is used as the time-segmented prediction parameter for the current monitoring time; if the gestational week information is at the critical value of a certain subset (e.g., 32 weeks + 6 days of gestation), the parameters of two adjacent subsets are automatically matched and the average value is taken as the transition parameter, thereby ensuring the continuity of prediction; in other embodiments, a finer-grained division of gestational week can also be used (e.g., dividing the parameter subset by week), which is not limited in this application.

[0035] In specific implementation, the systolic blood pressure at different monitoring times is input into the matched systolic blood pressure-heart rate correlation model, and the predicted heart rate value corresponding to different monitoring times is output in the following way: For each monitoring time, the systolic blood pressure value at the monitoring time is converted into a feature vector that the model can recognize (consistent with the feature dimension of historical systolic blood pressure data), and input into the systolic blood pressure-heart rate correlation model with matched time-segment prediction parameters; after the model is processed by a preset algorithm (e.g., multiple linear regression to calculate the basic predicted value, gradient boosting tree for nonlinear correction), the predicted heart rate value at the monitoring time is output, and the confidence interval of the predicted heart rate value (e.g., ±3 beats / min) is marked simultaneously; the output predicted heart rate value is associated and stored with the timestamp, systolic blood pressure, and gestational age information of the corresponding monitoring time to form an association record of "monitoring time-systolic blood pressure-gestational age-predicted heart rate value", which provides a data basis for subsequent calculation of the correlation deviation of physiological indicators. Other methods can also be used in other embodiments, which are not limited here.

[0036] In some embodiments, determining the correlation deviation between the physiological indicators of the target mother and fetus at different monitoring times based on each predicted heart rate value and the actual heart rate value of the target fetus at the corresponding monitoring time can be achieved by sampling the following steps: For each monitoring time, determine the allowable error of heart rate at that time. Determine the heart rate difference between the predicted heart rate at the monitoring time and the actual heart rate corresponding to the target fetus; Based on the allowable error of heart rate and the heart rate difference, the correlation deviation of physiological indicators between the target mother and the fetus at the monitoring time is determined, and then the correlation deviation of physiological indicators between the target mother and the fetus at different monitoring times is obtained.

[0037] It should be noted that the physiological indicator correlation deviation in this application is an indicator that quantifies the difference between the predicted fetal heart rate corresponding to the target maternal systolic blood pressure during the perinatal period and the actual fetal heart rate at the same monitoring time. Its core function is to serve as a preliminary basis for judging whether the correlation between "maternal-fetal" physiological indicators is abnormal. Through this indicator, deviations between the actual fetal heart rate and the expected coordination pattern can be quickly identified, thereby triggering the subsequent operation of "correcting heart rate based on body position / fetal position" to avoid health risks caused by failure to deal with abnormal indicator correlation in a timely manner.

[0038] Specifically, the allowable error for heart rate at monitoring times can be determined as follows: First, based on historical datasets of healthy mothers and infants during the perinatal period, stratified statistical analysis is performed according to gestational week (28-32 weeks, 33-36 weeks, 37 weeks to delivery) and monitoring time period (8-12 am, 12-4 pm, 4-8 pm, 8 pm to 8 am the next day). The normal fluctuation range of fetal heart rate within each stratum is calculated (mean ± 2 standard deviation). Then, the upper limit of the normal fluctuation range for each stratum is determined. The value is used as the allowable error of heart rate for the corresponding layer (e.g., the allowable error from 37 weeks of pregnancy to 20 minutes before delivery to 8 am the next day is set to 8 beats / min, and from 28 to 32 weeks of pregnancy to 8 am to 12 pm is set to 12 beats / min). At the same time, the dynamic adjustment interface of the error of each layer is retained (which can be updated regularly according to the new healthy samples). Finally, the allowable error of heart rate corresponding to the monitoring time is matched. In other embodiments, the fetal heart rate fluctuation threshold recommended by clinical guidelines (e.g., 10-15 beats / min) can also be used for correction, which is not limited in this application.

[0039] In specific implementation, the heart rate difference between the predicted heart rate value at the monitoring time and the actual heart rate value corresponding to the target fetus can be determined as follows: calculate the difference between the predicted heart rate value at the monitoring time and the actual heart rate value corresponding to the target fetus, and use the absolute value of the difference as the heart rate difference between the predicted heart rate value at the monitoring time and the actual heart rate value corresponding to the target fetus. In specific implementation, the correlation deviation between the physiological indicators of the target mother and the fetus at the monitoring time can be determined based on the allowable heart rate error and the heart rate difference as follows: use the ratio of the heart rate difference to the allowable heart rate error as the correlation deviation between the physiological indicators of the target mother and the fetus at the monitoring time.

[0040] It should be noted that in other embodiments, a combined calculation method of "trend consistency + numerical difference" can be used to determine the heart rate variability. For example, firstly, the trend of the predicted heart rate value and the actual heart rate value can be analyzed using a sliding window (e.g., 3 consecutive monitoring times) (synchronous increase / synchronous decrease / reverse change). If the trends are inconsistent, a penalty coefficient of 20% is added to the numerical difference. Then, the "composite variability = (|actual value - predicted value / predicted value) × 100% × (1 + trend penalty coefficient)" is calculated, which reflects both the numerical difference and the trend deviation of the physiological correlation between the two (e.g., a decrease in the predicted value while an increase in the actual value may indicate an abnormal correlation). For cases where the trends are consistent, the numerical variability is used directly to avoid over-correction. "Probability of deviation risk" can be introduced to replace direct numerical calculation to determine the deviation of the physiological indicator correlation. For example, firstly, a correlation model between the deviation and abnormal pregnancy outcomes (e.g., fetal distress) can be constructed based on historical data, and the "probability of abnormal outcome under the current variability" can be calculated using logistic regression. Then, the risk level is determined by combining the allowable error of heart rate (e.g., risk probability < 5% is "low deviation risk", 5%-20% is "low deviation risk"). (The risk level is defined as "medium risk" and >20% as "high risk"). Finally, the risk level and the corresponding probability value are used together as the output of the physiological indicator correlation deviation (e.g., "medium risk, probability 12%)", which more intuitively supports subsequent clinical decision-making. When connected with the "extracting maternal and infant health characteristics" step, the risk level can be directly used as the feature dimension. Other methods can also be used in other embodiments, which are not limited here.

[0041] It should be noted that existing technologies often use fixed thresholds for deviation calculations, without considering the physiological differences between gestational age and monitoring period, resulting in a false negative rate as high as 20% in late pregnancy. This application reduces the false negative rate and improves the accuracy of abnormal early warning in clinical testing by using time-segmented dynamic thresholds and maternal-infant synergy coefficient correction.

[0042] In step 104, for each monitoring time, if the correlation deviation of the physiological indicators is greater than the preset correlation deviation threshold, the actual heart rate value of the target fetus at the corresponding monitoring time is corrected based on the position data of the target mother and the fetal position data of the target fetus at the corresponding monitoring time.

[0043] It should be noted that the correlation deviation threshold in this application can be determined using the following method: In some embodiments, reference Figure 3 As shown in the figure, this is a flowchart illustrating the correction of actual heart rate values ​​in some embodiments of this application. In this embodiment, the correction of the actual heart rate value of the target fetus at the corresponding monitoring time based on the target mother's position data and the target fetus's fetal position data at the corresponding monitoring time can be achieved by the following steps: In step 1031, the type of dominant deviation factor at the corresponding monitoring time is determined based on the target mother's position data, the target fetus's position data, the target mother's systolic blood pressure fluctuation range, and the target mother's gestational age. In step 1032, a comprehensive correction coefficient for the actual heart rate value is determined based on the gestational age of the target mother, the duration of the target mother's position, the fluctuation range of the target mother's systolic blood pressure, and the fetal position type of the target fetus. In step 1033, the actual heart rate value of the target fetus at the monitoring time is subjected to hierarchical progressive correction according to the priority order of the dominant deviation factor type and the comprehensive correction coefficient to obtain the preliminary corrected heart rate value; In step 1034, the preliminary corrected heart rate value is validated in two dimensions to obtain the actual heart rate value of the target fetus after correction at the corresponding monitoring time.

[0044] It should be noted that the types of deviation-dominant factors include: position-dominant, fetal position-dominant, and position-fetal position mixed-dominant; the position duration is: the continuous duration for which the target mother maintains the current position before the monitoring time; the systolic blood pressure fluctuation amplitude is: the absolute value of the difference between the target mother's systolic blood pressure at the monitoring time and the previous monitoring time; the fetal position type includes: cephalic presentation (occiput anterior), cephalic presentation (occiput posterior), breech presentation, and transverse presentation; the priority order is: position-dominant > position-fetal position mixed-dominant > fetal position-dominant.

[0045] In specific implementation, determining the dominant deviation factor type for the corresponding monitoring time based on the target mother's position data, the target fetus's position data, the target mother's systolic blood pressure fluctuation amplitude, and the target mother's gestational age can be achieved in the following ways: A model can be trained based on a historical dataset of healthy mothers and infants, and then the target mother's position data, target fetus's position data, target mother's systolic blood pressure fluctuation amplitude, and target mother's gestational age at the corresponding monitoring time can be input into the model to determine the dominant deviation factor type for the corresponding monitoring time. Alternatively, professional medical personnel can determine the dominant deviation factor type for the corresponding monitoring time based on the target mother's position data, target fetus's position data, target mother's systolic blood pressure fluctuation amplitude, and target mother's gestational age. Other methods can also be used in other embodiments, which will not be elaborated here.

[0046] In practice, the comprehensive correction coefficient for determining the actual heart rate value based on the gestational age of the target mother, the duration of the target mother's position, the fluctuation range of the target mother's systolic blood pressure, and the fetal position type of the target fetus can be achieved in the following way: the comprehensive correction coefficient can be determined through "weighted fusion of multi-dimensional sub-matrixes". For example, for the gestational age sub-matrix: sub-matrices are divided according to 28-32 weeks of gestation, 33-36 weeks of gestation, and 37 weeks to delivery. Based on historical data, the position and fetal position of each sub-period are statistically analyzed. The coefficients for the influence on heart rate (e.g., 0.95-1.05 for 37 weeks of gestation to pre-delivery, 0.9-1.1 for 28-32 weeks of gestation); for the submatrix of body position duration: divided into submatrices of <10 minutes, 10-20 minutes, and >20 minutes, with the coefficients for body position duration increasing with duration (e.g., <10 minutes = 1.0, 10-20 minutes = 1.03, >20 minutes = 1.07); the submatrix of systolic blood pressure fluctuations: divided into <5 mmHg, 5-10 mmHg... The system is divided into sub-matrices based on blood pressure readings >10 mmHg, with the systolic blood pressure fluctuation coefficient increasing with the fluctuation amplitude (e.g., <5 mmHg = 1.0, 5-10 mmHg = 1.02, >10 mmHg = 1.05); a sub-matrix is ​​based on fetal position type: cephalic presentation (occiput anterior), cephalic presentation (occiput posterior), breech, and transverse presentation, with the fetal position coefficient increasing sequentially (1.0, 1.02, 1.05, 1.07); finally, for a given monitoring time, the coefficients for the corresponding dimensions are matched from each sub-matrix (e.g., 35 weeks gestation + supine position 25...). (Minutes + systolic blood pressure fluctuation of 12 mmHg + breech presentation, matching coefficients are 1.02, 1.07, 1.05, 1.05). A comprehensive correction coefficient is calculated by weighted product, where the comprehensive correction coefficient = gestational age coefficient × (position duration coefficient × 0.4 + systolic blood pressure fluctuation coefficient × 0.3 + fetal position coefficient × 0.3). The weights are determined based on the influence of each dimension on the deviation, with position duration having the greatest impact. Other methods can be used in other embodiments, which are not limited here.

[0047] In practice, the actual heart rate value of the target fetus at the monitoring time is progressively corrected according to the priority order of the dominant deviation factors and the comprehensive correction coefficient. The preliminary corrected heart rate value can be obtained in the following way: First, the actual heart rate value is corrected based on the correction coefficient of the corresponding dominant factor. Then, the correction result is fine-tuned based on the correction coefficient of the secondary factor to obtain the preliminary corrected heart rate value. For example, for position-dominant correction, the first step is to correct the actual heart rate value using the position duration coefficient × systolic blood pressure fluctuation coefficient, with the formula: Preliminary corrected value = actual heart rate value × (1 / (position duration coefficient × systolic blood pressure fluctuation coefficient)). The second step is to use the fetal position coefficient... Fine-tuning is performed using the formula: Preliminary corrected heart rate value = Preliminary corrected value × (1 / fetal position coefficient). For mixed-dominant correction, the comprehensive correction coefficient is used directly, with the formula: Preliminary corrected heart rate value = Actual heart rate value × (1 / comprehensive correction coefficient). For fetal position-dominant correction, the first step is to correct the actual heart rate value using the fetal position coefficient, with the formula: Preliminary corrected value = Actual heart rate value × (1 / fetal position coefficient). The second step is to fine-tune using the body position duration coefficient × systolic blood pressure fluctuation coefficient, with the formula: Preliminary corrected heart rate value = Preliminary corrected value × (1 / (body position duration coefficient × systolic blood pressure fluctuation coefficient)). Other methods can be used in other embodiments, which are not limited here.

[0048] In specific implementation, the initial corrected heart rate value is validated using a two-dimensional method to obtain the corrected actual heart rate value of the target fetus at the corresponding monitoring time. This can be achieved as follows: the two dimensions include a first dimension and a second dimension. The first dimension is the correlation deviation between the initial corrected heart rate value and the predicted heart rate value at the corresponding monitoring time. The second dimension is the heart rate variability (HRV) of the target fetus within 5 minutes before and after the monitoring time. If the correlation deviation of the first dimension is less than or equal to a preset correlation deviation threshold and the heart rate variability of the second dimension is within the normal range of 5-15 beats / min, then the correction is deemed valid, and the initial corrected heart rate value is used as the final corrected heart rate value. The corrected heart rate value is then evaluated. If the correction is deemed ineffective, the specific implementation steps of "if the correlation deviation of physiological indicators is greater than the preset correlation deviation threshold, then the actual heart rate value of the target fetus at the corresponding monitoring time is corrected based on the position data of the target mother and the fetal position data of the target fetus at the corresponding monitoring time" are re-executed to confirm whether the type of deviation dominant factor is misjudged. If the type of deviation dominant factor is correct, the potential physiological abnormality is determined based on the systolic blood pressure trend of the target mother and the heart rate trend of the target fetus in the past hour. If the potential physiological abnormality exists, an abnormality warning is pushed to the medical terminal. Other methods can be used in other embodiments, which are not limited here.

[0049] In addition, in some embodiments, if the correlation deviation of physiological indicators is less than or equal to a preset correlation deviation threshold, no processing is performed.

[0050] In step 105, the maternal and infant physiological health index of the target mother and fetus is extracted from all corrected actual heart rate values, all systolic blood pressure, all body position characteristics and all fetal position characteristics, and then an abnormal monitoring report is pushed to the medical terminal based on the maternal and infant physiological health index.

[0051] In some embodiments, extracting the maternal-fetal physiological synergistic health index from all corrected actual heart rate values, all systolic blood pressure values, all body position characteristics, and all fetal position characteristics can be achieved by the following steps: Extract the systolic blood pressure-fetal heart rate matching degree between the target mother and the fetus from all corrected actual heart rate values ​​and all systolic blood pressure values; The position-fetal position compatibility coefficient between the target mother and the fetus is extracted from all positional characteristics and all fetal position characteristics. The systolic blood pressure-fetal heart rate matching degree and the body position-fetal position adaptation coefficient are synergistically fused to obtain the maternal and infant physiological synergistic health index of the target mother and fetus.

[0052] It should be noted that the maternal and infant physiological synergy health index in this application comprehensively reflects the quantitative indicators of the synergy and stability of the physiological state of the target mother and fetus during the perinatal period (from 28 weeks of gestation to delivery). Its core function is to serve as the core basis for judging whether the overall health status of the mother and baby is normal. By integrating two key dimensions, namely "the matching between maternal systolic blood pressure and fetal heart rate" and "the compatibility between maternal position and fetal position", it intuitively reflects the synergistic pattern between maternal and infant physiological indicators.

[0053] Specifically, in implementation, the systolic blood pressure-fetal heart rate matching degree between the target mother and fetus can be extracted from all corrected actual heart rate values ​​and all systolic blood pressure values ​​in the following manner: First, it can be determined whether all corrected actual heart rate values ​​fall within the range of 110-160 beats / min. If so, it is marked as "fetal heart rate meets the standard". Then, it can be determined whether all systolic blood pressures fall within the range of 110-140 mmHg. If so, it is marked as "systolic blood pressure meets the standard". Next, the difference between all actual heart rate values ​​and the predicted heart rate values ​​at the corresponding monitoring time is calculated. If the difference is ≤ the standard allowable error (±5 beats / min), it is marked as "correlation meets the standard". Finally, the total number of times corresponding to "fetal heart rate meets the standard", "systolic blood pressure meets the standard", and "correlation meets the standard" is counted. The total number is divided by the total number of monitoring times to obtain the systolic blood pressure-fetal heart rate matching degree. Other methods can also be used in other embodiments, which are not limited here.

[0054] In practice, extracting the maternal-fetal position compatibility coefficient from all maternal and fetal position characteristics can be achieved as follows: First, referring to clinical perinatal monitoring guidelines, assign a baseline compatibility level (range 0.5-1.0) to the maternal-fetal position combinations at different monitoring times. For example, for a high compatibility combination (baseline compatibility level 1.0): maternal lateral decubitus position + fetal cephalic presentation (clinically recommended combination, with minimal ultrasound signal interference, requiring little or no compensation correction); for a medium compatibility combination (baseline compatibility level 0... 0.8): Mother sitting position + fetal breech position (moderate signal interference, small compensation correction). For low fit combination (basic fit level is 0.5): Mother supine position + fetal transverse position (maximum signal interference, large compensation correction). Then, the average of the basic fit levels corresponding to all monitoring times is used as the position-fetal position fit coefficient of the target mother and fetus. In other embodiments, the mode of the basic fit levels corresponding to all monitoring times can also be used as the position-fetal position fit coefficient. In other embodiments, other methods can also be used, which are not limited here.

[0055] In specific implementation, the systolic blood pressure-fetal heart rate matching degree and the body position-fetal position adaptation coefficient are synergistically fused to obtain the maternal and infant physiological health index of the target mother and fetus. This can be achieved by weighting the systolic blood pressure-fetal heart rate matching degree and the body position-fetal position adaptation coefficient according to a weight ratio of 6:4 to obtain the maternal and infant physiological health index of the target mother and fetus. Other methods can also be used in other embodiments, which are not limited here.

[0056] In some embodiments, pushing abnormal monitoring reports to medical terminals based on the maternal and infant physiological co-health index can be achieved through the following steps: Obtain preset thresholds for maternal and infant physiological health coordination; The maternal and infant physiological health index is compared with the maternal and infant physiological health threshold. If the maternal and infant physiological health index is greater than or equal to the maternal and infant physiological health threshold, no action is taken. When the maternal and infant physiological health index is less than the maternal and infant physiological health threshold, an abnormal monitoring report is pushed to the medical terminal.

[0057] It should be noted that the maternal-infant physiological coordination health threshold refers to a critical health value characterizing whether the physiological state of the mother and infant is within the normal range over a certain period of time. As a preferred embodiment, the maternal-infant physiological coordination health threshold can be determined by the following method: Historical maternal-infant physiological indicator data of multiple healthy mothers can be obtained from a maternal-infant health database, including: systolic blood pressure data of the healthy mothers, body position data of the mothers, and corresponding fetal heart rate data and fetal position data. Based on the maternal-infant physiological indicator data, the systolic blood pressure-fetal heart rate matching degree and body position-fetal position matching degree between each healthy mother and fetus can be extracted. The fetal position compatibility coefficient is then used to synthesize the mean values ​​of all systolic blood pressure-fetal heart rate compatibility and all body position-fetal position compatibility coefficients to obtain an initial maternal and infant physiological health threshold. This initial maternal and infant physiological health threshold is then compared with the experience of medical experts and clinical diagnostic and treatment standards. If necessary, the initial maternal and infant physiological health threshold can be modified in stages according to the maternal recovery stage (different stages during pregnancy and postpartum) and the infant's age characteristics to obtain the final maternal and infant physiological health threshold. Other methods can also be used in other embodiments, which are not limited here.

[0058] It should be noted that although this application provides examples of various gestational weeks and the corresponding parameters, the specific monitoring time period can be adjusted according to needs. As a preferred embodiment, this application outputs a monitoring report once a day and only monitors the data of the target mother and fetus from the previous day.

[0059] In another aspect, in some embodiments, this application provides a perinatal maternal and infant health monitoring system, with reference to... Figure 4 The figure is a schematic diagram of the perinatal maternal and infant health monitoring system according to some embodiments of this application. The perinatal maternal and infant health monitoring system 400 includes: a data acquisition module 401, a processing module 402, and an execution module 403, which are described below: The data acquisition module 401 in this application is mainly used to collect historical systolic blood pressure data of different healthy mothers during the perinatal period and historical heart rate data of corresponding healthy fetuses. Processing module 402, in this application, is mainly used to establish a systolic blood pressure-heart rate correlation model between perinatal mothers and fetuses based on all historical systolic blood pressure data and all historical heart rate data; The processing module 402 described in this application is also used to acquire the systolic blood pressure and body position characteristics of the target pregnant woman at different monitoring times during the perinatal period, the actual heart rate value and fetal position characteristics of the target fetus, and predict the heart rate prediction value corresponding to the systolic blood pressure of the target pregnant woman at different monitoring times through the systolic blood pressure-heart rate correlation model. Based on each heart rate prediction value and the actual heart rate value of the target fetus at the corresponding monitoring time, the correlation deviation degree of the physiological indicators between the target pregnant woman and the fetus at different monitoring times is determined. The processing module 402 described in this application is also used to, for each monitoring time, if the correlation deviation of physiological indicators is greater than a preset correlation deviation threshold, correct the actual heart rate value of the target fetus at the corresponding monitoring time based on the body position data of the target mother and the fetal position data of the target fetus at the corresponding monitoring time. The execution module 403 in this application is mainly used to extract the maternal and infant physiological health index of the target mother and fetus from all corrected actual heart rate values, all systolic blood pressure, all body position characteristics and all fetal position characteristics, and then push abnormal monitoring reports to the medical terminal based on the maternal and infant physiological health index.

[0060] Each module in the aforementioned perinatal maternal and infant health monitoring system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0061] In another embodiment, this application provides a computer device, which may be a server, and its internal structure diagram may be as follows. Figure 5 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores perinatal maternal and infant health monitoring data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a perinatal maternal and infant health monitoring method.

[0062] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0063] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described embodiments of the perinatal maternal and infant health monitoring method.

[0064] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps described in the perinatal maternal and infant health monitoring method embodiment.

[0065] In one embodiment, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps described in the perinatal maternal and infant health monitoring method embodiment.

[0066] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0067] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0068] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for monitoring maternal and infant health during the perinatal period, characterized in that, Includes the following steps: Collect historical systolic blood pressure data of different healthy mothers during the perinatal period and historical heart rate data of corresponding healthy fetuses; A systolic blood pressure-heart rate correlation model between perinatal mothers and fetuses was established based on all historical systolic blood pressure data and all historical heart rate data. The systolic blood pressure and body position characteristics of the target pregnant woman at different monitoring times during the perinatal period, as well as the actual heart rate and fetal position characteristics of the target fetus, are obtained. The heart rate prediction value corresponding to the systolic blood pressure of the target pregnant woman at different monitoring times is predicted by the systolic blood pressure-heart rate correlation model. Based on each heart rate prediction value and the actual heart rate value of the target fetus at the corresponding monitoring time, the correlation deviation degree of the physiological indicators between the target pregnant woman and the fetus at different monitoring times is determined. For each monitoring time, if the correlation deviation of physiological indicators is greater than the preset correlation deviation threshold, the actual heart rate value of the target fetus at the corresponding monitoring time is corrected based on the position data of the target mother and the fetal position data of the target fetus at the corresponding monitoring time. The maternal and infant physiological health index of the target mother and fetus is extracted from all corrected actual heart rate values, all systolic blood pressure, all body position characteristics and all fetal position characteristics, and then abnormal monitoring reports are pushed to the medical terminal based on the maternal and infant physiological health index.

2. The method as described in claim 1, characterized in that, A perinatal maternal-fetal systolic blood pressure-heart rate correlation model was established based on all historical systolic blood pressure and heart rate data, specifically including: Data cleaning was performed on all historical systolic blood pressure data and all historical heart rate data to obtain cleaned historical systolic blood pressure dataset and historical heart rate dataset. The cleaned historical systolic blood pressure dataset and historical heart rate dataset are divided into training set, validation set and test set; The association prediction framework based on multiple linear regression and gradient boosting tree is trained using the training set, the validation set, and the test set. Historical systolic blood pressure is used as the input feature, and historical heart rate is used as the output feature to obtain the systolic blood pressure-heart rate association model.

3. The method as described in claim 1, characterized in that, The heart rate prediction values ​​corresponding to systolic blood pressure at different monitoring times for the target parturient, obtained through the systolic blood pressure-heart rate correlation model, specifically include: Obtain gestational week information corresponding to systolic blood pressure at different monitoring times for the target pregnant woman; Based on the information of each gestational week, the corresponding time-segmented prediction parameters in the systolic blood pressure-heart rate correlation model are matched; The systolic blood pressure at different monitoring times is input into the matched systolic blood pressure-heart rate correlation model, and the predicted heart rate value corresponding to different monitoring times is output.

4. The method as described in claim 1, characterized in that, Based on the predicted heart rate values ​​and the actual heart rate values ​​of the target fetus at the corresponding monitoring time, the correlation deviation between the physiological indicators of the target mother and the fetus at different monitoring times is determined, specifically including: For each monitoring time, determine the allowable error of heart rate at that time. Determine the heart rate difference between the predicted heart rate at the monitoring time and the actual heart rate corresponding to the target fetus; Based on the allowable error of heart rate and the heart rate difference, the correlation deviation of physiological indicators between the target mother and the fetus at the monitoring time is determined, and then the correlation deviation of physiological indicators between the target mother and the fetus at different monitoring times is obtained.

5. The method as described in claim 1, characterized in that, The correction of the actual heart rate value of the target fetus at the corresponding monitoring time, based on the maternal position data and fetal position data of the target mother at the corresponding monitoring time, specifically includes: Based on the target mother's body position data, the target fetus's fetal position data, the target mother's systolic blood pressure fluctuation range, and the target mother's gestational age at the corresponding monitoring time, determine the type of dominant deviation factor at the corresponding monitoring time. The comprehensive correction coefficient for the actual heart rate value is determined based on the gestational age of the target mother, the duration of the target mother's position, the fluctuation range of the target mother's systolic blood pressure, and the fetal position type of the target fetus. According to the priority order of the dominant deviation factors and the comprehensive correction coefficient, the actual heart rate value of the target fetus at the monitoring time is subjected to hierarchical progressive correction to obtain the preliminary corrected heart rate value. The initial corrected heart rate value is validated in two dimensions to obtain the actual heart rate value of the target fetus after correction at the corresponding monitoring time.

6. The method as described in claim 1, characterized in that, The maternal-fetal physiological synergy health index for the target mother and fetus was extracted from all corrected actual heart rate values, all systolic blood pressure values, all body position characteristics, and all fetal position characteristics. Specifically, it includes: Extract the systolic blood pressure-fetal heart rate matching degree between the target mother and the fetus from all corrected actual heart rate values ​​and all systolic blood pressure values; The position-fetal position compatibility coefficient between the target mother and the fetus is extracted from all positional characteristics and all fetal position characteristics. The systolic blood pressure-fetal heart rate matching degree and the body position-fetal position adaptation coefficient are synergistically fused to obtain the maternal and infant physiological synergistic health index of the target mother and fetus.

7. The method as described in claim 1, characterized in that, If the correlation deviation of physiological indicators is less than or equal to the preset correlation deviation threshold, no action will be taken.

8. A perinatal maternal and infant health monitoring system, characterized in that, include: The data acquisition module is used to collect historical systolic blood pressure data of different healthy mothers during the perinatal period and historical heart rate data of the corresponding healthy fetuses. The processing module is used to build a systolic blood pressure-heart rate correlation model between the perinatal mother and the fetus based on all historical systolic blood pressure data and all historical heart rate data. The processing module is also used to acquire the systolic blood pressure and body position characteristics of the target mother at different monitoring times during the perinatal period, the actual heart rate value and fetal position characteristics of the target fetus, and predict the heart rate prediction value corresponding to the systolic blood pressure of the target mother at different monitoring times through the systolic blood pressure-heart rate correlation model. Based on each heart rate prediction value and the actual heart rate value of the target fetus at the corresponding monitoring time, the correlation deviation degree of the physiological indicators between the target mother and the fetus at different monitoring times is determined. The processing module is also used to, for each monitoring time, if the correlation deviation of physiological indicators is greater than a preset correlation deviation threshold, correct the actual heart rate value of the target fetus at the corresponding monitoring time based on the position data of the target mother and the fetal position data of the target fetus at the corresponding monitoring time. The execution module is used to extract the maternal and infant physiological health index of the target mother and fetus from all corrected actual heart rate values, all systolic blood pressure, all body position characteristics and all fetal position characteristics, and then push abnormal monitoring reports to the medical terminal based on the maternal and infant physiological health index.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the perinatal maternal and infant health monitoring method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the perinatal maternal and infant health monitoring method as described in any one of claims 1 to 7.

Images