CN118352084A - A method and system for predicting eclampsia in pregnant women based on data analysis - Google Patents

Abstract

The invention discloses a method and a system for predicting eclampsia of a pregnant woman based on data analysis, and aims to realize early prediction and effective monitoring of eclampsia risk of the pregnant woman through advanced data processing and machine learning technologies. Firstly, a hierarchical database is constructed, and physiological data and medical history physiological data of pregnant women are collected and imported. By carrying out cluster analysis on the data, key physiological characteristics of occurrence of epilepsy in different pregnancy periods are identified. Based on the characteristics, a epilepsy prediction model is further constructed, so that the physiological index of the pregnant woman can be monitored in real time, and the risk degree of the eclampsia can be predicted. Finally, according to the predicted risk data, a risk monitoring system is constructed, and the system can divide the monitoring grade for each pregnant woman and provide a corresponding monitoring scheme, so that scientific and personalized monitoring suggestions are provided for medical institutions and pregnant women, the occurrence rate of eclampsia is effectively reduced, and the health of the pregnant women and fetuses is ensured.

Description

A method and system for predicting eclampsia in pregnant women based on data analysis

Technical Field

The present invention relates to the field of medical monitoring technology, and in particular to a method and system for predicting eclampsia in pregnant women based on data analysis.

Background technique

In the field of pregnancy management and maternal health monitoring, the prevention and early diagnosis of preeclampsia is a challenging task. Preeclampsia is a hypertensive disease specific to pregnancy. If not treated in time, it may lead to serious health problems and even endanger the life safety of pregnant women and fetuses. Traditional preventive measures rely on regular blood pressure and proteinuria testing, but these methods have limited ability to identify and personalize the risk of eclampsia in pregnant women early. In addition, the health data of pregnant women are often distributed in different medical institutions, lacking effective data integration and analysis capabilities, which limits the possibility of comprehensive assessment of the health status of pregnant women and timely intervention.

With the development of big data and machine learning technologies, data analysis-based prediction and management technologies for eclampsia in pregnant women have become a new research trend. These technologies can make comprehensive use of multi-source data such as the historical health records and physiological parameters of pregnant women, and achieve early identification and real-time monitoring of the risk of preeclampsia through accurate data analysis and model prediction. However, how to effectively integrate large-scale health data, build a high-accuracy prediction model, and apply it to actual pregnancy management and risk intervention remains a technical challenge. In addition, ensuring the generalization and interpretability of the prediction model so that medical professionals can make reasonable intervention recommendations based on the model prediction is also the key to improving the quality and efficiency of pregnancy management. Therefore, developing a method for predicting eclampsia in pregnant women that can not only efficiently integrate multi-source health data, but also provide accurate predictions and practical intervention plans is of great practical significance for improving the quality of medical services during pregnancy and the health of pregnant women and fetuses.

Summary of the invention

In order to solve at least one of the above technical problems, the present invention proposes a method and system for predicting eclampsia in pregnant women based on data analysis.

The first aspect of the present invention provides a method for predicting eclampsia in pregnant women based on data analysis, comprising:

Constructing a hierarchical database, obtaining physiological data and medical history physiological data of a preset number of pregnant women with eclampsia, and importing the physiological data and medical history physiological data into the hierarchical database;

Performing clustering operation on physiological data and medical history physiological data of each level in the hierarchical database to obtain clustering results of physiological data and medical history physiological data of each level;

Determining physiological characteristics of eclampsia in pregnant women in different pregnancy cycles according to the clustering results;

Building an eclampsia prediction model based on the physiological characteristics of eclampsia, monitoring the physiological indicators of pregnant women in real time, and predicting the risk of eclampsia in pregnant women according to the eclampsia prediction model and the physiological indicators to obtain risk prediction data;

A risk monitoring system is constructed, and the monitoring level of the risk monitoring system is determined according to the risk prediction data to obtain a monitoring plan for eclampsia in pregnant women.

In this solution, the hierarchical database is constructed to obtain physiological data and medical history physiological data of a preset number of pregnant women with eclampsia, and the physiological data and medical history physiological data are imported into the hierarchical database, specifically:

Obtain physiological data and medical history physiological data of a preset number of pregnant women with eclampsia;

Obtaining pregnancy cycle data of pregnant women with eclampsia, and dividing the physiological data and medical data into hierarchical groups according to the pregnancy cycle to obtain hierarchical data;

constructing a hierarchical database, setting a construction level of the hierarchical database according to the hierarchical data, and determining a table structure of the hierarchical database;

The hierarchical data is imported into the hierarchical database according to the constructed hierarchical and table structures.

In this solution, the physiological data and medical history physiological data of each level in the hierarchical database are clustered to obtain the clustering results of the physiological data and medical history physiological data of each level, specifically:

A hierarchical clustering algorithm is introduced, and the physiological data and medical history physiological data of each level in the hierarchical database are taken as the data to be clustered in turn;

Taking each data item in the data to be clustered as a data item to be clustered in turn, taking each data point in the data item to be clustered as a separate cluster, calculating the Manhattan distance between each cluster, and evaluating the similarity of each cluster according to the Manhattan distance to obtain a similarity matrix;

Merging two clusters with the highest similarity according to the similarity matrix to obtain a merged cluster, and updating the similarity matrix according to the merged cluster to obtain an updated matrix;

The updated matrix is cyclically merged with the two clusters with the highest similarity, and the similarity matrix is updated again until the clustering level reaches a preset number, and the clustering operation is stopped to obtain the clustering results of the physiological data and medical history physiological data of each level. The clustering results of the physiological data and medical history physiological data of each level are the results of classifying pregnant women with eclampsia whose each physiological data and each medical history physiological data of each level are within the preset data range into one category.

In this solution, the physiological characteristics of eclampsia in pregnant women in different pregnancy cycles are determined according to the clustering results, specifically:

Obtaining a variation range of standard physiological data of pregnant women in different pregnancy cycles, comparing the clustering results of each level with the variation range of standard physiological data according to the clustering results, marking the clustering results that are not within the variation range of standard physiological data, and obtaining a marked clustering result;

Performing a patient number counting operation on the clustering results of each item of physiological data and each item of medical history physiological data in the marked clustering results to obtain patient number data for each marked clustering result;

Performing a risk scoring operation on the data range of each clustering result in the marked clustering result according to the patient population data, to obtain risk scoring results of each physiological data at each level and each medical history physiological data in different numerical ranges;

The physiological characteristics of eclampsia occurring in different pregnancy cycles are determined according to the risk assessment results. The physiological characteristics of eclampsia occurring are physiological data items related to the occurrence of eclampsia in different pregnancy cycles of pregnant women, numerical ranges of physiological data items, and the risk of occurrence of each numerical range of physiological data items.

In this solution, the eclampsia prediction model is constructed based on the physiological characteristics of eclampsia, the physiological indicators of pregnant women are monitored in real time, and the risk of eclampsia in pregnant women is predicted according to the eclampsia prediction model and physiological indicators to obtain risk prediction data, specifically:

Based on the information entropy weight method, weight calculation is performed on the physiological characteristics of eclampsia to obtain the weight data of the influence of each physiological characteristic of eclampsia on the occurrence of eclampsia;

Constructing an eclampsia prediction model based on a logistic regression algorithm, importing the physiological characteristics and influencing weight data of the occurrence of eclampsia into the eclampsia prediction model for learning and training, and optimizing the loss function of the eclampsia prediction model by a gradient descent algorithm;

Based on the real-time monitoring of the physiological indicators of pregnant women by electronic health wearable devices, the current pregnancy cycle data of pregnant women is obtained, and the risk of eclampsia of pregnant women is predicted in real time based on the physiological indicators of pregnant women according to the eclampsia prediction model and pregnancy cycle data to obtain risk prediction data.

In this scheme, the risk monitoring system is constructed, and the monitoring level of the risk monitoring system is determined according to the risk prediction data to obtain the monitoring scheme for eclampsia in pregnant women, which is specifically:

Constructing a risk monitoring system, setting a risk monitoring threshold, and determining the monitoring level of pregnant women based on the risk monitoring threshold and risk prediction data;

If the monitoring level is a high monitoring level, a danger warning is issued in real time to remind the pregnant woman to perform emergency medical intervention, and voice data of the pregnant woman is acquired in real time, semantic analysis is performed on the voice data, the state of the pregnant woman is determined according to the semantics, and whether to perform an automatic alarm operation according to the state;

If the monitoring level is medium monitoring level, establish the recommended time and frequency of medical examinations for pregnant women;

If the monitoring level is a low monitoring level, the monitoring frequency of the physiological indicators of the pregnant woman is reduced to obtain a monitoring plan for eclampsia in the pregnant woman.

The second aspect of the present invention further provides a system for predicting eclampsia in pregnant women based on data analysis, the system comprising: a memory, a processor, the memory comprising a method program for predicting eclampsia in pregnant women based on data analysis, and when the method program for predicting eclampsia in pregnant women based on data analysis is executed by the processor, the following steps are implemented:

Constructing a hierarchical database, obtaining physiological data and medical history physiological data of a preset number of pregnant women with eclampsia, and importing the physiological data and medical history physiological data into the hierarchical database;

Performing clustering operation on physiological data and medical history physiological data of each level in the hierarchical database to obtain clustering results of physiological data and medical history physiological data of each level;

Determining physiological characteristics of eclampsia in pregnant women in different pregnancy cycles according to the clustering results;

Building an eclampsia prediction model based on the physiological characteristics of eclampsia, monitoring the physiological indicators of pregnant women in real time, and predicting the risk of eclampsia in pregnant women according to the eclampsia prediction model and the physiological indicators to obtain risk prediction data;

A risk monitoring system is constructed, and the monitoring level of the risk monitoring system is determined according to the risk prediction data to obtain a monitoring plan for eclampsia in pregnant women.

In this solution, the physiological characteristics of eclampsia in pregnant women in different pregnancy cycles are determined according to the clustering results, specifically:

Obtaining a variation range of standard physiological data of pregnant women in different pregnancy cycles, comparing the clustering results of each level with the variation range of standard physiological data according to the clustering results, marking the clustering results that are not within the variation range of standard physiological data, and obtaining a marked clustering result;

Performing a patient number counting operation on the clustering results of each item of physiological data and each item of medical history physiological data in the marked clustering results to obtain patient number data for each marked clustering result;

Performing a risk scoring operation on the data range of each clustering result in the marked clustering result according to the patient population data, to obtain risk scoring results of each physiological data at each level and each medical history physiological data in different numerical ranges;

The physiological characteristics of eclampsia occurring in different pregnancy cycles are determined according to the risk assessment results. The physiological characteristics of eclampsia occurring are physiological data items related to the occurrence of eclampsia in different pregnancy cycles of pregnant women, numerical ranges of physiological data items, and the risk of occurrence of each numerical range of physiological data items.

In this solution, the eclampsia prediction model is constructed based on the physiological characteristics of eclampsia, the physiological indicators of pregnant women are monitored in real time, and the risk of eclampsia in pregnant women is predicted according to the eclampsia prediction model and physiological indicators to obtain risk prediction data, specifically:

Based on the information entropy weight method, weight calculation is performed on the physiological characteristics of eclampsia to obtain the weight data of the influence of each physiological characteristic of eclampsia on the occurrence of eclampsia;

Constructing an eclampsia prediction model based on a logistic regression algorithm, importing the physiological characteristics and influencing weight data of the occurrence of eclampsia into the eclampsia prediction model for learning and training, and optimizing the loss function of the eclampsia prediction model by a gradient descent algorithm;

Based on the real-time monitoring of the physiological indicators of pregnant women by electronic health wearable devices, the current pregnancy cycle data of pregnant women is obtained, and the risk of eclampsia of pregnant women is predicted in real time based on the physiological indicators of pregnant women according to the eclampsia prediction model and pregnancy cycle data to obtain risk prediction data.

In this scheme, the risk monitoring system is constructed, and the monitoring level of the risk monitoring system is determined according to the risk prediction data to obtain the monitoring scheme for eclampsia in pregnant women, which is specifically:

Constructing a risk monitoring system, setting a risk monitoring threshold, and determining a monitoring level for pregnant women based on the risk monitoring threshold and risk prediction data;

If the monitoring level is a high monitoring level, a danger warning is issued in real time to remind the pregnant woman to perform emergency medical intervention, and voice data of the pregnant woman is acquired in real time, semantic analysis is performed on the voice data, the state of the pregnant woman is determined according to the semantics, and whether to perform an automatic alarm operation according to the state;

If the monitoring level is medium monitoring level, establish the recommended time and frequency of medical examinations for pregnant women;

If the monitoring level is a low monitoring level, the monitoring frequency of the physiological indicators of the pregnant woman is reduced to obtain a monitoring plan for eclampsia in the pregnant woman.

The present invention discloses a method and system for predicting eclampsia in pregnant women based on data analysis, aiming to achieve early prediction and effective monitoring of the risk of eclampsia in pregnant women through advanced data processing and machine learning technologies. First, a hierarchical database is constructed to collect and import the physiological data and medical history physiological data of pregnant women. By clustering these data, the key physiological characteristics of eclampsia in different pregnancy cycles are identified. Based on these characteristics, an eclampsia prediction model is further constructed, which can monitor the physiological indicators of pregnant women in real time and predict the risk level of eclampsia. Finally, according to the predicted risk data, a risk monitoring system is constructed, which can divide the monitoring level for each pregnant woman and propose corresponding monitoring plans, so as to provide scientific and personalized monitoring suggestions for medical institutions and pregnant women themselves, effectively reduce the incidence of eclampsia, and protect the health of pregnant women and fetuses.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 shows a flow chart of a method for predicting eclampsia in pregnant women based on data analysis according to the present invention;

FIG2 shows a flow chart of the present invention for determining physiological characteristics of eclampsia in pregnant women in different pregnancy cycles;

FIG3 shows a flow chart of obtaining risk prediction data according to the present invention;

FIG4 shows a block diagram of a system for predicting eclampsia in pregnant women based on data analysis according to the present invention.

Detailed ways

In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

FIG1 shows a flow chart of a method for predicting eclampsia in pregnant women based on data analysis according to the present invention.

As shown in FIG1 , the first aspect of the present invention provides a method for predicting eclampsia in pregnant women based on data analysis, comprising:

S102, constructing a hierarchical database, obtaining physiological data and medical history physiological data of a preset number of pregnant women with eclampsia, and importing the physiological data and medical history physiological data into the hierarchical database;

S104, performing a clustering operation on the physiological data and medical history physiological data of each level in the hierarchical database to obtain a clustering result of the physiological data and medical history physiological data of each level;

S106, determining physiological characteristics of eclampsia in pregnant women in different pregnancy cycles according to the clustering results;

S108, constructing an eclampsia prediction model based on the physiological characteristics of eclampsia, monitoring the physiological indicators of the pregnant woman in real time, and predicting the risk of eclampsia in the pregnant woman according to the eclampsia prediction model and the physiological indicators to obtain risk prediction data;

S110, constructing a risk monitoring system, determining a monitoring level of the risk monitoring system according to the risk prediction data, and obtaining a monitoring plan for eclampsia in pregnant women.

It should be noted that, since a large amount of physiological data and physical therapy history physiological data of pregnant women with eclampsia need to be collected, the data processing and management capabilities can be greatly improved by constructing a hierarchical database to store the data. The physiological data include blood routine indicators, proteinuria, oxygen saturation, heart rate, and weight. The medical history physiological data refers to the physiological data recorded in the historical medical records during pregnancy. Because the probability of occurrence and physiological characteristics of eclampsia in pregnant women are different in different pregnancy cycles, the hierarchical data formed by the data of pregnant women in different pregnancy cycles are clustered, and the occurrence of eclampsia in different pregnancy cycles is determined by the clustering results. The physiological characteristics of pregnant women are identified and an eclampsia prediction model is constructed. The risk of eclampsia in pregnant women is predicted based on the prediction model. Finally, a risk monitoring system is built to conduct real-time monitoring operations on pregnant women. Through early identification and accurate prediction of eclampsia risk, as well as the implementation of effective monitoring and intervention measures, the health problems caused by eclampsia in pregnant women and fetuses can be significantly reduced, the safety of pregnancy can be improved, and the waste of medical resources can be reduced. By analyzing the risk prediction data, the risk monitoring system can determine different monitoring levels and formulate personalized monitoring plans for pregnant women with eclampsia accordingly. This personalized monitoring plan can provide more accurate and effective monitoring and intervention measures for pregnant women with different risk levels.

According to an embodiment of the present invention, the hierarchical database is constructed to obtain physiological data and medical history physiological data of a preset number of pregnant women with eclampsia, and the physiological data and medical history physiological data are imported into the hierarchical database, specifically:

Obtain physiological data and medical history physiological data of a preset number of pregnant women with eclampsia;

Obtaining pregnancy cycle data of pregnant women with eclampsia, and dividing the physiological data and medical data into hierarchical groups according to the pregnancy cycle to obtain hierarchical data;

constructing a hierarchical database, setting a construction level of the hierarchical database according to the hierarchical data, and determining a table structure of the hierarchical database;

The hierarchical data is imported into the hierarchical database according to the constructed hierarchical and table structures.

It should be noted that the hierarchical data is to divide the physiological data and medical history physiological data of pregnant women with eclampsia in the same pregnancy cycle into the same area.

According to an embodiment of the present invention, the clustering operation is performed on the physiological data and medical history physiological data of each level in the hierarchical database to obtain the clustering results of the physiological data and medical history physiological data of each level, specifically:

A hierarchical clustering algorithm is introduced, and the physiological data and medical history physiological data of each level in the hierarchical database are taken as the data to be clustered in turn;

Taking each data item in the data to be clustered as a data item to be clustered in turn, taking each data point in the data item to be clustered as a separate cluster, calculating the Manhattan distance between each cluster, and evaluating the similarity of each cluster according to the Manhattan distance to obtain a similarity matrix;

Merging two clusters with the highest similarity according to the similarity matrix to obtain a merged cluster, and updating the similarity matrix according to the merged cluster to obtain an updated matrix;

The updated matrix is cyclically merged with the two clusters with the highest similarity, and the similarity matrix is updated again until the clustering level reaches a preset number, and the clustering operation is stopped to obtain the clustering results of the physiological data and medical history physiological data of each level. The clustering results of the physiological data and medical history physiological data of each level are the results of classifying pregnant women with eclampsia whose each physiological data and each medical history physiological data of each level are within the preset data range into one category.

It should be noted that by introducing a hierarchical clustering algorithm to conduct a detailed clustering analysis of the physiological data and medical history of pregnant women, the hierarchical clustering algorithm can classify data with similar characteristics into one category, accurately identify groups of different risk levels, and gradually merge data points according to the similarity of physiological data and medical history to form a risk stratification from fine to coarse. This method can reveal subtle differences in the risk of eclampsia in pregnant women; the clustering operation is to cluster data of the same category, and data of non-same categories cannot be clustered. For example, blood pressure data within a certain level is clustered, that is, pregnant women with eclampsia with similar blood pressure within the level are classified into one category; the data items are all physiological data items and historical medical data items at each level. For example, if the physiological data contains blood pressure, blood pressure is a physiological data item.

FIG. 2 shows a flow chart of the present invention for determining physiological characteristics of eclampsia in pregnant women in different pregnancy cycles.

According to an embodiment of the present invention, the physiological characteristics of eclampsia occurring in pregnant women in different pregnancy cycles are determined according to the clustering results, specifically:

S202, obtaining a variation range of standard physiological data of pregnant women in different pregnancy cycles, comparing the clustering results of each level with the variation range of standard physiological data according to the clustering results, marking the clustering results that are not in the variation range of standard physiological data, and obtaining a marked clustering result;

S204, performing a patient number counting operation on the clustering results of each item of physiological data and each item of medical history physiological data in the marked clustering results to obtain patient number data for each marked clustering result;

S206, performing a risk scoring operation on the data range of each clustering result in the marked clustering result according to the patient number data, to obtain risk scoring results of each physiological data of each level and each medical history physiological data in different numerical ranges;

S208, determining the physiological characteristics of eclampsia in different pregnancy cycles based on the risk assessment result, wherein the physiological characteristics of eclampsia are physiological data items related to the occurrence of eclampsia in different pregnancy cycles of pregnant women, numerical ranges of physiological data items, and the risk of occurrence of each numerical range of physiological data items.

It should be noted that the marked clustering results are clustering results that are not within the standard physiological data variation range. Marking the clustering results that are not within the standard physiological data variation range can preliminarily identify the abnormal physiological data of pregnant women. Abnormal physiological data is likely to cause eclampsia in pregnant women. First, excluding normal data can reduce the amount of data analysis and reduce the difficulty of analysis; the risk scoring operation refers to an assessment method for the risk of eclampsia in pregnant women based on the number of people in each clustering result in each marked clustering result. For example, in the clustering results of blood pressure items, the clustering results with blood pressure values between 130-140 have the most pregnant women with eclampsia. It can be considered that the risk probability of eclampsia in pregnant women with blood pressure values between 130-140 is the highest, and the risk score is also higher.

FIG3 shows a flow chart of obtaining risk prediction data according to the present invention.

According to an embodiment of the present invention, the eclampsia prediction model is constructed based on the physiological characteristics of eclampsia, the physiological indicators of pregnant women are monitored in real time, and the risk of eclampsia in pregnant women is predicted according to the eclampsia prediction model and the physiological indicators to obtain risk prediction data, specifically:

S302, weighting the physiological characteristics of eclampsia based on an information entropy weighting method to obtain weight data of the influence of each physiological characteristic of eclampsia on the occurrence of eclampsia;

S304, constructing an eclampsia prediction model based on a logistic regression algorithm, importing the physiological characteristics and influencing weight data of the occurrence of eclampsia into the eclampsia prediction model for learning and training, and optimizing the loss function of the eclampsia prediction model by a gradient descent algorithm;

S306, based on real-time monitoring of the physiological indicators of pregnant women using electronic health wearable devices, obtaining the current pregnancy cycle data of the pregnant woman, and making a real-time prediction of the risk of eclampsia of the pregnant woman based on the physiological indicators of the pregnant woman according to the eclampsia prediction model and the pregnancy cycle data to obtain risk prediction data.

It should be noted that the introduction of the information entropy weight method can objectively evaluate the impact of each physiological characteristic on the risk of eclampsia, reduce the interference of subjective judgment, and improve the accuracy and scientificity of feature weight allocation. The application of the information entropy weight method ensures the objectivity and rationality of the weights of each physiological characteristic in the eclampsia prediction model, so that the model can more accurately reflect the probability of eclampsia in pregnant women; the logistic regression algorithm can accurately predict the risk level of eclampsia in pregnant women, providing doctors with a powerful tool to achieve personalized medical decisions and timely intervention measures, which helps to reduce the incidence of eclampsia and improve the health and safety of pregnant women and fetuses; electronic health wearable devices can continuously and in real time monitor the physiological indicators of pregnant women, such as heart rate, blood pressure, etc., which provides technical support for the construction of a real-time dynamic maternal health monitoring system. Through real-time monitoring, abnormal changes in the physiological indicators of pregnant women can be discovered in time, so as to give timely warnings before eclampsia occurs. This early warning mechanism can greatly improve the safety of pregnant women and fetuses.

According to an embodiment of the present invention, the risk monitoring system is constructed, and the monitoring level of the risk monitoring system is determined according to the risk prediction data to obtain a monitoring plan for eclampsia in pregnant women, specifically:

Constructing a risk monitoring system, setting a risk monitoring threshold, and determining a monitoring level for pregnant women based on the risk monitoring threshold and risk prediction data;

If the monitoring level is a high monitoring level, a danger warning is issued in real time to remind the pregnant woman to perform emergency medical intervention, and voice data of the pregnant woman is acquired in real time, semantic analysis is performed on the voice data, the state of the pregnant woman is determined according to the semantics, and whether to perform an automatic alarm operation according to the state;

If the monitoring level is medium monitoring level, establish the recommended time and frequency of medical examinations for pregnant women;

If the monitoring level is a low monitoring level, the monitoring frequency of the physiological indicators of the pregnant woman is reduced to obtain a monitoring plan for eclampsia in the pregnant woman.

It should be noted that by setting a reasonable risk monitoring threshold, the system can ensure that it can respond in time when a risk signal is detected. This method is based on a large amount of data analysis and can accurately determine when the health status of pregnant women requires closer observation or intervention, ensuring that pregnant women receive immediate attention and intervention when the risk of eclampsia increases, reducing the risk of eclampsia and improving the safety of pregnant women and fetuses; when the monitoring level reaches a high-risk level, the system can issue a real-time danger warning and obtain and analyze voice data in real time. This process uses advanced data processing and natural language processing technology. This mechanism can respond quickly to emergencies in pregnant women and provide medical assistance in a timely manner, reducing potential threats to the health of pregnant women and improving the efficiency of emergency medical response; at the medium monitoring level, the system will formulate personalized medical examination recommendations for pregnant women based on risk prediction data and historical health records. Time and frequency of examinations; at the low monitoring level, the system will reduce the frequency of monitoring of physiological indicators of pregnant women. This adjustment is based on the analysis of a large amount of data to ensure that the monitoring effectiveness is maintained while reducing interference with pregnant women.

According to an embodiment of the present invention, it also includes:

If the monitoring level is high, the movement trajectory data of the electronic health wearable device is obtained in real time;

According to the motion trajectory data, a preset distance trajectory before and after the mutation point of the motion trajectory of the pregnant woman is framed to obtain a sensitive trajectory map;

Obtain a historical abnormal trajectory map data set of pregnant women in various abnormal motion states, compare the sensitive trajectory map with the abnormal trajectory map, and calculate the similarity between the sensitive trajectory map and each abnormal trajectory map;

Selecting the top three abnormal trajectory graphs with the highest similarity as comparison results, and predicting the abnormal motion state of the pregnant woman's current motion trajectory according to the comparison results and the abnormal trajectory graphs;

Determining the danger level of the pregnant woman's current state according to the abnormal movement state;

According to the degree of danger, the current location of the pregnant woman is obtained in real time, and the current location is sent to the communication equipment of relevant personnel for emergency processing to obtain emergency warning information;

The emergency warning information is used to supplement the eclampsia monitoring program for pregnant women.

It should be noted that under the monitoring of electronic health wearable devices, the real-time movement trajectory of pregnant women will be recorded, and the abnormal state of pregnant women will be judged based on the movement trajectory. For example, in the case of a sudden fall, the movement trajectory will show a sudden change, and the mutation point will be analyzed before and after to further determine whether the trajectory near the mutation point is abnormal, thereby analyzing the possible abnormal movement state of the pregnant woman, analyzing the degree of danger based on the abnormal movement state, and finally forming early warning information; the early warning information is used to supplement the eclampsia monitoring program for pregnant women, further optimize the application scenarios of the program, and enable the eclampsia monitoring program for pregnant women to adapt to and identify more dangerous scenarios for pregnant women.

FIG4 shows a block diagram of a system for predicting eclampsia in pregnant women based on data analysis according to the present invention.

The second aspect of the present invention further provides a system 4 for predicting eclampsia in pregnant women based on data analysis, the system comprising: a memory 41, a processor 42, the memory comprising a method program for predicting eclampsia in pregnant women based on data analysis, and when the method program for predicting eclampsia in pregnant women based on data analysis is executed by the processor, the following steps are implemented:

Constructing a hierarchical database, obtaining physiological data and medical history physiological data of a preset number of pregnant women with eclampsia, and importing the physiological data and medical history physiological data into the hierarchical database;

Performing clustering operation on physiological data and medical history physiological data of each level in the hierarchical database to obtain clustering results of physiological data and medical history physiological data of each level;

Determining physiological characteristics of eclampsia in pregnant women in different pregnancy cycles according to the clustering results;

Building an eclampsia prediction model based on the physiological characteristics of eclampsia, monitoring the physiological indicators of pregnant women in real time, and predicting the risk of eclampsia in pregnant women according to the eclampsia prediction model and the physiological indicators to obtain risk prediction data;

A risk monitoring system is constructed, and the monitoring level of the risk monitoring system is determined according to the risk prediction data to obtain a monitoring plan for eclampsia in pregnant women.

According to an embodiment of the present invention, the hierarchical database is constructed to obtain physiological data and medical history physiological data of a preset number of pregnant women with eclampsia, and the physiological data and medical history physiological data are imported into the hierarchical database, specifically:

Obtain physiological data and medical history physiological data of a preset number of pregnant women with eclampsia;

Obtaining pregnancy cycle data of pregnant women with eclampsia, and dividing the physiological data and medical data into hierarchical groups according to the pregnancy cycle to obtain hierarchical data;

constructing a hierarchical database, setting a construction level of the hierarchical database according to the hierarchical data, and determining a table structure of the hierarchical database;

The hierarchical data is imported into the hierarchical database according to the constructed hierarchical and table structures.

It should be noted that the hierarchical data is to divide the physiological data and medical history physiological data of pregnant women with eclampsia in the same pregnancy cycle into the same area.

According to an embodiment of the present invention, the clustering operation is performed on the physiological data and medical history physiological data of each level in the hierarchical database to obtain the clustering results of the physiological data and medical history physiological data of each level, specifically:

A hierarchical clustering algorithm is introduced, and the physiological data and medical history physiological data of each level in the hierarchical database are taken as the data to be clustered in turn;

Taking each data item in the data to be clustered as a data item to be clustered in turn, taking each data point in the data item to be clustered as a separate cluster, calculating the Manhattan distance between each cluster, and evaluating the similarity of each cluster according to the Manhattan distance to obtain a similarity matrix;

Merging two clusters with the highest similarity according to the similarity matrix to obtain a merged cluster, and updating the similarity matrix according to the merged cluster to obtain an updated matrix;

The updated matrix is cyclically merged with the two clusters with the highest similarity, and the similarity matrix is updated again until the clustering level reaches a preset number, and the clustering operation is stopped to obtain the clustering results of the physiological data and medical history physiological data of each level. The clustering results of the physiological data and medical history physiological data of each level are the results of classifying pregnant women with eclampsia whose each physiological data and each medical history physiological data of each level are within the preset data range into one category.

According to an embodiment of the present invention, the physiological characteristics of eclampsia occurring in pregnant women in different pregnancy cycles are determined according to the clustering results, specifically:

Obtaining a variation range of standard physiological data of pregnant women in different pregnancy cycles, comparing the clustering results of each level with the variation range of standard physiological data according to the clustering results, marking the clustering results that are not within the variation range of standard physiological data, and obtaining a marked clustering result;

Performing a patient number counting operation on the clustering results of each item of physiological data and each item of medical history physiological data in the marked clustering results to obtain patient number data for each marked clustering result;

Performing a risk scoring operation on the data range of each clustering result in the marked clustering result according to the patient population data, to obtain risk scoring results of each physiological data at each level and each medical history physiological data in different numerical ranges;

The physiological characteristics of eclampsia occurring in different pregnancy cycles are determined according to the risk assessment results. The physiological characteristics of eclampsia occurring are physiological data items related to the occurrence of eclampsia in different pregnancy cycles of pregnant women, numerical ranges of physiological data items, and the risk of occurrence of each numerical range of physiological data items.

According to an embodiment of the present invention, the eclampsia prediction model is constructed based on the physiological characteristics of eclampsia, the physiological indicators of pregnant women are monitored in real time, and the risk of eclampsia in pregnant women is predicted according to the eclampsia prediction model and the physiological indicators to obtain risk prediction data, specifically:

Based on the information entropy weight method, weight calculation is performed on the physiological characteristics of eclampsia to obtain the weight data of the influence of each physiological characteristic of eclampsia on the occurrence of eclampsia;

Constructing an eclampsia prediction model based on a logistic regression algorithm, importing the physiological characteristics and influencing weight data of the occurrence of eclampsia into the eclampsia prediction model for learning and training, and optimizing the loss function of the eclampsia prediction model by a gradient descent algorithm;

Based on the real-time monitoring of the physiological indicators of pregnant women by electronic health wearable devices, the current pregnancy cycle data of pregnant women is obtained, and the risk of eclampsia of pregnant women is predicted in real time based on the physiological indicators of pregnant women according to the eclampsia prediction model and pregnancy cycle data to obtain risk prediction data.

According to an embodiment of the present invention, the risk monitoring system is constructed, and the monitoring level of the risk monitoring system is determined according to the risk prediction data to obtain a monitoring plan for eclampsia in pregnant women, specifically:

Constructing a risk monitoring system, setting a risk monitoring threshold, and determining a monitoring level for pregnant women based on the risk monitoring threshold and risk prediction data;

If the monitoring level is a high monitoring level, a danger warning is issued in real time to remind the pregnant woman to perform emergency medical intervention, and voice data of the pregnant woman is acquired in real time, semantic analysis is performed on the voice data, the state of the pregnant woman is determined according to the semantics, and whether to perform an automatic alarm operation according to the state;

If the monitoring level is medium monitoring level, establish the recommended time and frequency of medical examinations for pregnant women;

If the monitoring level is a low monitoring level, the monitoring frequency of the physiological indicators of the pregnant woman is reduced to obtain a monitoring plan for eclampsia in the pregnant woman.

The present invention discloses a method and system for predicting eclampsia in pregnant women based on data analysis, aiming to achieve early prediction and effective monitoring of the risk of eclampsia in pregnant women through advanced data processing and machine learning technologies. First, a hierarchical database is constructed to collect and import the physiological data and medical history physiological data of pregnant women. By clustering these data, the key physiological characteristics of eclampsia in different pregnancy cycles are identified. Based on these characteristics, an eclampsia prediction model is further constructed, which can monitor the physiological indicators of pregnant women in real time and predict the risk level of eclampsia. Finally, according to the predicted risk data, a risk monitoring system is constructed, which can divide the monitoring level for each pregnant woman and propose corresponding monitoring plans, so as to provide scientific and personalized monitoring suggestions for medical institutions and pregnant women themselves, effectively reduce the incidence of eclampsia, and protect the health of pregnant women and fetuses.

In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

A person skilled in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, etc. Various media that can store program codes.

Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims (10)

1. A method for predicting eclampsia in pregnant women based on data analysis, characterized in that it comprises the following steps: Constructing a hierarchical database, obtaining physiological data and medical history physiological data of a preset number of pregnant women with eclampsia, and importing the physiological data and medical history physiological data into the hierarchical database; Performing clustering operation on physiological data and medical history physiological data of each level in the hierarchical database to obtain clustering results of physiological data and medical history physiological data of each level; Determining physiological characteristics of eclampsia in pregnant women in different pregnancy cycles according to the clustering results; Building an eclampsia prediction model based on the physiological characteristics of eclampsia, monitoring the physiological indicators of pregnant women in real time, and predicting the risk of eclampsia in pregnant women according to the eclampsia prediction model and the physiological indicators to obtain risk prediction data; A risk monitoring system is constructed, and the monitoring level of the risk monitoring system is determined according to the risk prediction data to obtain a monitoring plan for eclampsia in pregnant women. 2. A method for predicting eclampsia in pregnant women based on data analysis according to claim 1, characterized in that the hierarchical database is constructed, physiological data and medical history physiological data of a preset number of pregnant women with eclampsia are obtained, and the physiological data and medical history physiological data are imported into the hierarchical database, specifically: Obtain physiological data and medical history physiological data of a preset number of pregnant women with eclampsia; Obtaining pregnancy cycle data of pregnant women with eclampsia, and dividing the physiological data and medical data into hierarchical groups according to the pregnancy cycle to obtain hierarchical data; constructing a hierarchical database, setting a construction level of the hierarchical database according to the hierarchical data, and determining a table structure of the hierarchical database; The hierarchical data is imported into the hierarchical database according to the constructed hierarchical and table structures. 3. A method for predicting eclampsia in pregnant women based on data analysis according to claim 1, characterized in that the physiological data and medical history physiological data of each level in the hierarchical database are clustered to obtain the clustering results of the physiological data and medical history physiological data of each level, specifically: A hierarchical clustering algorithm is introduced, and the physiological data and medical history physiological data of each level in the hierarchical database are taken as the data to be clustered in turn; Taking each data item in the data to be clustered as a data item to be clustered in turn, taking each data point in the data item to be clustered as a separate cluster, calculating the Manhattan distance between each cluster, and evaluating the similarity of each cluster according to the Manhattan distance to obtain a similarity matrix; Merging two clusters with the highest similarity according to the similarity matrix to obtain a merged cluster, and updating the similarity matrix according to the merged cluster to obtain an updated matrix; The updated matrix is cyclically merged with the two clusters with the highest similarity, and the similarity matrix is updated again until the clustering level reaches a preset number, and the clustering operation is stopped to obtain the clustering results of the physiological data and medical history physiological data of each level. The clustering results of the physiological data and medical history physiological data of each level are the results of classifying pregnant women with eclampsia whose each physiological data and each medical history physiological data of each level are within the preset data range into one category. 4. A method for predicting eclampsia in pregnant women based on data analysis according to claim 1, characterized in that the physiological characteristics of eclampsia in pregnant women in different pregnancy cycles are determined according to the clustering results, specifically: Obtaining a variation range of standard physiological data of pregnant women in different pregnancy cycles, comparing the clustering results of each level with the variation range of standard physiological data according to the clustering results, marking the clustering results that are not within the variation range of standard physiological data, and obtaining a marked clustering result; Performing a patient number counting operation on the clustering results of each item of physiological data and each item of medical history physiological data in the marked clustering results to obtain patient number data for each marked clustering result; Performing a risk scoring operation on the data range of each clustering result in the marked clustering result according to the patient population data, to obtain risk scoring results of each physiological data at each level and each medical history physiological data in different numerical ranges; The physiological characteristics of eclampsia occurring in different pregnancy cycles are determined according to the risk assessment results. The physiological characteristics of eclampsia occurring are physiological data items related to the occurrence of eclampsia in different pregnancy cycles of pregnant women, numerical ranges of physiological data items, and the risk of occurrence of each numerical range of physiological data items. 5. A method for predicting eclampsia in pregnant women based on data analysis according to claim 1, characterized in that an eclampsia prediction model is constructed based on the physiological characteristics of eclampsia, the physiological indicators of pregnant women are monitored in real time, and the risk of eclampsia in pregnant women is predicted according to the eclampsia prediction model and physiological indicators to obtain risk prediction data, specifically: Based on the information entropy weight method, weight calculation is performed on the physiological characteristics of eclampsia to obtain the weight data of the influence of each physiological characteristic of eclampsia on the occurrence of eclampsia; Constructing an eclampsia prediction model based on a logistic regression algorithm, importing the physiological characteristics and influencing weight data of the occurrence of eclampsia into the eclampsia prediction model for learning and training, and optimizing the loss function of the eclampsia prediction model by a gradient descent algorithm; Based on the real-time monitoring of the physiological indicators of pregnant women by electronic health wearable devices, the current pregnancy cycle data of pregnant women is obtained, and the risk of eclampsia of pregnant women is predicted in real time based on the physiological indicators of pregnant women according to the eclampsia prediction model and pregnancy cycle data to obtain risk prediction data. 6. A method for predicting eclampsia in pregnant women based on data analysis according to claim 1, characterized in that the risk monitoring system is constructed, the monitoring level of the risk monitoring system is determined according to the risk prediction data, and the monitoring plan for eclampsia in pregnant women is obtained, specifically: Constructing a risk monitoring system, setting a risk monitoring threshold, and determining a monitoring level for pregnant women based on the risk monitoring threshold and risk prediction data; If the monitoring level is a high monitoring level, a danger warning is issued in real time to remind the pregnant woman to perform emergency medical intervention, and voice data of the pregnant woman is acquired in real time, semantic analysis is performed on the voice data, the state of the pregnant woman is determined according to the semantics, and whether to perform an automatic alarm operation according to the state; If the monitoring level is medium monitoring level, establish the recommended time and frequency of medical examinations for pregnant women; If the monitoring level is a low monitoring level, the monitoring frequency of the physiological indicators of the pregnant woman is reduced to obtain a monitoring plan for eclampsia in the pregnant woman. 7. A system for predicting eclampsia in pregnant women based on data analysis, characterized in that the system for predicting eclampsia in pregnant women based on data analysis comprises a storage device and a processor, the storage device comprises a method program for predicting eclampsia in pregnant women based on data analysis, and when the method program for predicting eclampsia in pregnant women based on data analysis is executed by the processor, the following steps are implemented: Constructing a hierarchical database, obtaining physiological data and medical history physiological data of a preset number of pregnant women with eclampsia, and importing the physiological data and medical history physiological data into the hierarchical database; Performing clustering operation on physiological data and medical history physiological data of each level in the hierarchical database to obtain clustering results of physiological data and medical history physiological data of each level; Determining physiological characteristics of eclampsia in pregnant women in different pregnancy cycles according to the clustering results; Building an eclampsia prediction model based on the physiological characteristics of eclampsia, monitoring the physiological indicators of pregnant women in real time, and predicting the risk of eclampsia in pregnant women according to the eclampsia prediction model and the physiological indicators to obtain risk prediction data; A risk monitoring system is constructed, and the monitoring level of the risk monitoring system is determined according to the risk prediction data to obtain a monitoring plan for eclampsia in pregnant women. 8. A system for predicting eclampsia in pregnant women based on data analysis according to claim 7, characterized in that the physiological characteristics of eclampsia in pregnant women in different pregnancy cycles are determined according to the clustering results, specifically: Obtaining a variation range of standard physiological data of pregnant women in different pregnancy cycles, comparing the clustering results of each level with the variation range of standard physiological data according to the clustering results, marking the clustering results that are not within the variation range of standard physiological data, and obtaining a marked clustering result; Performing a patient number counting operation on the clustering results of each item of physiological data and each item of medical history physiological data in the marked clustering results to obtain patient number data for each marked clustering result; Performing a risk scoring operation on the data range of each clustering result in the marked clustering result according to the patient population data, to obtain risk scoring results of each physiological data at each level and each medical history physiological data in different numerical ranges; The physiological characteristics of eclampsia occurring in different pregnancy cycles are determined according to the risk assessment results. The physiological characteristics of eclampsia occurring are physiological data items related to the occurrence of eclampsia in different pregnancy cycles of pregnant women, numerical ranges of physiological data items, and the risk of occurrence of each numerical range of physiological data items. 9. A system for predicting eclampsia in pregnant women based on data analysis according to claim 7, characterized in that the eclampsia prediction model is constructed based on the physiological characteristics of the occurrence of eclampsia, the physiological indicators of pregnant women are monitored in real time, and the risk of eclampsia in pregnant women is predicted according to the eclampsia prediction model and physiological indicators to obtain risk prediction data, specifically: Based on the information entropy weight method, weight calculation is performed on the physiological characteristics of eclampsia to obtain the weight data of the influence of each physiological characteristic of eclampsia on the occurrence of eclampsia; Constructing an eclampsia prediction model based on a logistic regression algorithm, importing the physiological characteristics and influencing weight data of the occurrence of eclampsia into the eclampsia prediction model for learning and training, and optimizing the loss function of the eclampsia prediction model by a gradient descent algorithm; Based on the real-time monitoring of the physiological indicators of pregnant women by electronic health wearable devices, the current pregnancy cycle data of pregnant women is obtained, and the risk of eclampsia of pregnant women is predicted in real time based on the physiological indicators of pregnant women according to the eclampsia prediction model and pregnancy cycle data to obtain risk prediction data. 10. A system for predicting eclampsia in pregnant women based on data analysis according to claim 7, characterized in that the risk monitoring system is constructed, and the monitoring level of the risk monitoring system is determined according to the risk prediction data to obtain the monitoring plan for eclampsia in pregnant women, specifically: Constructing a risk monitoring system, setting a risk monitoring threshold, and determining a monitoring level for pregnant women based on the risk monitoring threshold and risk prediction data; If the monitoring level is a high monitoring level, a danger warning is issued in real time to remind the pregnant woman to perform emergency medical intervention, and voice data of the pregnant woman is acquired in real time, semantic analysis is performed on the voice data, the state of the pregnant woman is determined according to the semantics, and whether to perform an automatic alarm operation according to the state; If the monitoring level is medium monitoring level, establish the recommended time and frequency of medical examinations for pregnant women; If the monitoring level is a low monitoring level, the monitoring frequency of the physiological indicators of the pregnant woman is reduced to obtain a monitoring plan for eclampsia in the pregnant woman.