Data management control method and system based on maternal and child one-stop learning cloud platform

Abstract

The application relates to the technical field of data management control, and particularly discloses a data management control method and system based on a maternal and infant one-stop learning cloud platform, which comprises the following steps: based on real-time collected maternal and infant physiological data, maternal and infant historical health records and user learning behavior data, a multi-modal maternal and infant user portrait containing dynamic health state labels and dynamic learning ability labels is constructed; based on multi-modal maternal and infant user portraits of all maternal and infant users, multi-dimensional health horizontal comparison factors of pregnant women and fetuses or infants in each maternal and infant user under each division attribute and each division level are determined, and maternal and infant health deviations of each maternal and infant user are analyzed; based on the multi-dimensional health horizontal comparison factors of the pregnant women and the fetuses or the infants in each maternal and infant user under each division attribute and each division level and the maternal and infant health deviations, intelligent adaptation of multiple services is driven; and a closed loop from health deviation to education content, health plan and deviation improvement is formed.

Description

Technical Field

[0001] This invention relates to the field of data management and control technology, and in particular to a data management and control method and system based on a one-stop learning cloud platform for women and children. Background Technology

[0002] In today's society, maternal and child health and education receive widespread attention. The health status of women and children not only affects their own quality of life but also has a profound impact on the growth and development of the next generation. Meanwhile, with the development of internet technology, online education has become an important way for women and children to acquire health knowledge and improve their parenting skills. Data management and control methods and systems based on one-stop learning cloud platforms for women and children have emerged, which are of great significance for optimizing maternal and child health management and education services. As people increasingly value maternal and child health and education, and with the deepening application of digital technology in the medical and educational fields, this type of data management and control technology based on one-stop learning cloud platforms is expected to be more widely applied and developed in the future, driving maternal and child health and education services towards a more intelligent and personalized direction.

[0003] However, existing maternal and child health management and online education services have many shortcomings:

[0004] First, the fragmented nature of multi-source data makes it difficult to effectively support multi-dimensional assessments of maternal and child health and learning status. Current maternal and child health management systems are often limited to collecting in-hospital prenatal checkup data and basic physiological indicators, making it difficult to access real-time data from external IoT devices and completely disconnected from user learning behavior data. Online education platforms, on the other hand, only store learning records, remaining detached from health data. This results in an inability to accurately push suitable learning content based on the health status of women and children, offering only general courses and failing to meet personalized needs.

[0005] Second, in terms of health assessment, there is a lack of cross-sectional health comparisons between individuals and groups, relying mainly on static thresholds for judgment. This approach does not fully consider the relative health level of an individual within the group, making it prone to misjudgment. Furthermore, it fails to cover the correlation between fetuses / infants and pregnant women, ignoring the mutual influence of their health status and thus failing to comprehensively and accurately assess maternal and child health.

[0006] Third, the analysis of health deviations is too simplistic, merely a qualitative assessment of whether something is abnormal, lacking in-depth, dimensional, and hierarchical quantitative analysis. It fails to pinpoint the specific stage at which the health deviation occurs, nor can it correlate it with its impact on fetal development. Furthermore, it cannot determine the specific services needed based on the deviation, resulting in a lack of targeted and often arbitrary provision of services.

[0007] Fourth, service adaptation lacks data-driven approaches, resulting in low levels of personalization and intelligence. Existing service adaptation largely relies on fixed rules or human experience, failing to dynamically adjust based on individual health deviations and learning abilities. Whether it's pushing health education content or arranging health plans and service processes, a one-size-fits-all approach is adopted, disregarding individual differences, leading to low service acceptance and difficulty in meeting the diverse needs of women and children.

[0008] Therefore, this invention proposes a data management and control method and system based on a one-stop learning cloud platform for women and children. Summary of the Invention

[0009] This invention provides a data management and control method and system based on a one-stop learning cloud platform for women and children. This method and system can integrate various types of data related to maternal and child health and learning. Through scientific data management and control, it can achieve accurate assessment of maternal and child health status and personalized delivery of educational services. It helps improve the level of maternal and child health management, reduce health risks for pregnant women and infants, and meet the learning needs of different maternal and child groups, thereby improving learning outcomes.

[0010] This invention provides a data management and control method based on a one-stop learning cloud platform for women and children, comprising:

[0011] Based on real-time collected maternal and child physiological data, maternal and child historical health records, and user learning behavior data, a multimodal maternal and child user profile is constructed, which includes dynamic health status tags and dynamic learning ability tags.

[0012] Based on the multimodal maternal and child health user profiles of all maternal and child health users, we determined the multidimensional health comparison factors of pregnant women and fetuses or infants in each maternal and child health user group at each classification level of each classification attribute.

[0013] Based on the multidimensional health comparison factors of pregnant women and fetuses or infants in each category of each category of attributes, the maternal and child health deviations of each category of maternal and child health users were analyzed.

[0014] Based on the multidimensional health comparison factors and maternal and child health deviations of pregnant women and fetuses or infants in each user group under each classification level of each classification attribute, the system drives the intelligent adaptation of multiple services.

[0015] Preferably, based on real-time collected maternal and child physiological data, historical health records, and user learning behavior data, a multimodal maternal and child user profile is constructed, including dynamic health status tags and dynamic learning ability tags, comprising:

[0016] Through a multi-protocol adaptation engine, various physiological data of maternal and child users obtained from monitoring by various IoT medical devices are collected in real time as maternal and child physiological data, and maternal and child historical health records and user learning behavior data are also collected at the same time.

[0017] The real-time collected maternal and child physiological data, maternal and child historical health records, and user learning behavior data are transmitted to the cloud computing and storage layer for preliminary cleaning and format standardization to obtain effective multi-source data.

[0018] A multimodal profile of maternal and child health users, including dynamic health status labels and dynamic learning ability labels, is constructed based on effective multi-source data.

[0019] Preferably, a multimodal maternal and child health user profile is constructed based on effective multi-source data, including dynamic health status tags and dynamic learning ability tags, comprising:

[0020] By performing threshold determination and trend analysis on valid physiological data and valid historical health data from valid multi-source data, dynamic health status labels are generated.

[0021] Cluster analysis is performed on effective learning behavior data from effective multi-source data to generate dynamic learning ability tags that include preferred course types, learning abilities, and content interaction frequency;

[0022] By aggregating the dynamic health status tags and dynamic learning ability tags of each maternal and child health user, a multimodal maternal and child health user profile is constructed.

[0023] Preferably, based on the multimodal maternal and child health user profiles of all maternal and child health users, multidimensional health comparison factors for pregnant women and fetuses or infants in each user group are determined at each level of each classification attribute, including:

[0024] Extract the health profiles of pregnant women and fetuses or infants from the multimodal maternal and child health profiles of various maternal and child health users;

[0025] Based on a variety of preset pregnant and postpartum women classification attributes and corresponding classification levels, the health profiles of pregnant and postpartum women among all maternal and child health users are classified to obtain a group of pregnant and postpartum women health profiles under each classification level of each pregnant and postpartum women classification attribute.

[0026] Based on a variety of preset fetal or infant classification attributes and corresponding classification levels, the fetal or infant health profiles of all maternal and child health users are classified to obtain a group of fetal or infant health profiles under each classification level of each fetal or infant classification attribute.

[0027] Based on the health profiles of pregnant women and fetuses or infants in each user group of maternal and child health services at each level of each attribute, we analyzed the multidimensional health comparison factors of pregnant women and fetuses or infants in each user group at each level of each attribute.

[0028] Preferably, based on the multidimensional health comparison factors of pregnant women and fetuses or infants in each category of each attribute at each classification level, the maternal and child health deviations of each user group are analyzed, including:

[0029] By mapping each level of the classification attributes of pregnant women to each level of the classification attributes of fetuses or infants, the classification levels of each unified classification requirement are obtained.

[0030] The deviation of the multidimensional health comparison factors of pregnant women and fetuses or infants in each maternal and child health user group under each unified classification level is taken as the maternal and child health deviation of each maternal and child health user group.

[0031] Preferably, based on multidimensional health comparison factors and maternal and child health biases of pregnant women and fetuses or infants among each user group at each classification level of each classification attribute, intelligent adaptation of multiple services is driven, including:

[0032] Based on the multidimensional health comparison factors of pregnant women and fetuses or infants in all maternal and child users at each level of each classification attribute, the baseline range of each dimension of the horizontal comparison factors of pregnant women at each level of each classification attribute, and the baseline range of each dimension of the horizontal comparison factors of fetuses or infants at each level of each classification attribute are calculated.

[0033] Based on the baseline range of horizontal comparison factors of pregnant women and fetuses or infants at each level of each classification attribute, multidimensional health horizontal comparison factors and maternal and child health deviations, the two-dimensional feature coordinates of health and service adaptation of pregnant women and fetuses or infants in each maternal and child user at each level of each classification attribute are determined.

[0034] Based on the pregnant women, fetuses, or infants among the users of each maternal and child health service, the two-dimensional feature coordinates of health and service adaptation under each classification level of each classification attribute drive the intelligent adaptation of multiple services.

[0035] Preferably, based on the baseline range of horizontal comparison factors for each dimension of pregnant women and fetuses or infants at each level of each classification attribute, multidimensional health horizontal comparison factors, and maternal and child health bias, the two-dimensional feature coordinates of health and service adaptation for pregnant women and fetuses or infants among each maternal and child user at each level of each classification attribute are determined, including:

[0036] The proportion of single-dimensional comparative factors that did not exceed the corresponding benchmark range among the multidimensional health comparative factors of pregnant women in each category of each category of each attribute was statistically analyzed, and this proportion was used as the health attainment of pregnant women in each category of each category of each attribute at each level.

[0037] Based on the ratio of the maternal and child health deviation of pregnant women in each category of each category at each category level to the maximum maternal and child health deviation of pregnant women in each category at all categories at all categories, the service adaptation priority of pregnant women in each category at each category level is determined.

[0038] The proportion of single-dimensional comparative factors that did not exceed the corresponding benchmark range in the multidimensional health comparative factors of fetuses or infants in each category of each category of each attribute was statistically analyzed. This proportion was used as the health attainment of fetuses or infants in each category of each category of each attribute in each category of each maternal and child health user.

[0039] Based on the ratio of the maternal and child health deviation of the fetus or infant in each user group under each classification level of each classification attribute to the maximum maternal and child health deviation of the fetus or infant in each user group under all classification levels of all classification attributes, the service adaptation priority of the fetus or infant in each user group under each classification level of each classification attribute is determined.

[0040] Based on the health attainment and service adaptation priority of pregnant women in each category of each attribute at each level among the maternal and child health users, two-dimensional feature coordinates of health and service adaptation for the corresponding pregnant women in each category of each attribute at each level are determined. Similarly, based on the health attainment and service adaptation priority of fetuses or infants in each category of each attribute at each level among the maternal and child health users, two-dimensional feature coordinates of health and service adaptation for the corresponding fetuses or infants in each category of each attribute at each level are determined.

[0041] Preferably, based on the health and service adaptation two-dimensional feature coordinates of pregnant women and fetuses or infants among each group of maternal and child health users, at each classification level of each classification attribute, the intelligent adaptation of multiple services is driven, including:

[0042] Based on the preset sorting rules, the two-dimensional feature coordinates of health and service adaptation of pregnant women in each maternal and child health user under all classification levels of all classification attributes are sorted and curve fitted to obtain the first health and service adaptation curve.

[0043] Based on the preset sorting rules, the two-dimensional feature coordinates of health and service adaptation of fetuses or infants in each maternal and child user under all classification levels of all classification attributes are sorted and curve fitted to obtain the second health and service adaptation curve.

[0044] Generate the deviation curve between the first health and service adaptation curve and the first health and service standard adaptation curve as the first input curve, and use the deviation curve between the second health and service adaptation curve and the second health and service standard adaptation curve as the second input curve.

[0045] Intelligent adaptation of multiple services is driven by the first input curve, the second input curve, and the intelligent adaptation model.

[0046] Preferred options also include:

[0047] Based on the purpose, privacy sensitivity, and subject matter of each data block in the valid multi-source data, all data blocks are divided into a three-level classification system;

[0048] Based on the user role, domain attributes, and data ownership of the initiating user of the data access request, combined with a three-level classification system, the system dynamically responds to the data access request.

[0049] This invention provides a data management and control system based on a one-stop learning cloud platform for women and children, comprising:

[0050] The user profile building module is used to build a multimodal maternal and child user profile that includes dynamic health status tags and dynamic learning ability tags based on real-time collected maternal and child physiological data, maternal and child historical health records, and user learning behavior data.

[0051] The multidimensional horizontal comparison module is used to determine the multidimensional health horizontal comparison factors of pregnant women and fetuses or infants in each user group under each classification level of each classification attribute, based on the multimodal maternal and child user profile of all maternal and child users.

[0052] The health deviation analysis module is used to analyze the maternal and child health deviations of each user based on the multidimensional health comparison factors of pregnant women and fetuses or infants in each category at each category level of each category attribute.

[0053] The multi-directional intelligent adaptation module is used to drive the intelligent adaptation of multiple services based on the multi-dimensional health horizontal comparison factors and maternal and child health deviations of pregnant women and fetuses or infants in each category of each category attribute and at each category level.

[0054] The beneficial effects of this invention compared to existing technologies are as follows: First, by integrating maternal and child physiological data, historical health records, and user learning behavior data, a multimodal profile with dynamic health and learning ability tags is constructed, improving the accuracy of health and learning status assessment. This achieves both comprehensive assessment, covering both health and learning dimensions, and dynamic assessment, ensuring that the profile is updated in real time according to the user's actual status. Appropriate learning content is accurately pushed based on the maternal and child's health status to meet personalized needs. Second, by determining multidimensional health horizontal comparison factors and analyzing health deviations, multidimensional horizontal comparison and deviation analysis are achieved. This fully considers the relative health level of individuals within the group, covering the correlation comparison between fetuses / infants and pregnant women, taking into account the mutual influence of their health status, and comprehensively and accurately assessing the maternal and child's health status. Clearly defining demand matching provides a precise basis for service adaptation. Furthermore, driven by multidimensional horizontal comparison factors and health deviations, and combined with multimodal profiling, this method achieves data-driven intelligent service adaptation, enabling personalized adaptation by pushing differentiated services to users with different health deviations and learning abilities. It also achieves dynamic adaptation, adjusting services synchronously with changes in user status, and correlated adaptation, ensuring synergistic maternal and infant services. By considering individual differences, this approach results in high service acceptance and can meet the diverse needs of women and children. Finally, this method covers the entire life cycle of women and children, from data collection to service adaptation, improving the integration of health management and education. It continuously monitors and intervenes by linking maternal and infant health data, forming a closed loop from health deviations to educational content, health plans, and deviation improvement.

[0055] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0056] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0057] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0058] Figure 1 This is an overall architecture diagram of the data management and control method based on the one-stop learning cloud platform for women and children in this embodiment of the invention;

[0059] Figure 2 A detailed flowchart of the multimodal user profile construction in this embodiment of the invention;

[0060] Figure 3 This is a diagram of the intelligent adaptation driving mechanism in an embodiment of the present invention. Detailed Implementation

[0061] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0062] like Figure 1 As shown, this invention provides an embodiment of a data management and control method based on a one-stop learning cloud platform for women and children, comprising:

[0063] Based on real-time collected maternal and child physiological data, maternal and child historical health records, and user learning behavior data, a multimodal maternal and child user profile is constructed, which includes dynamic health status tags and dynamic learning ability tags.

[0064] Based on the multimodal maternal and child health user profiles of all maternal and child health users, we determined the multidimensional health comparison factors of pregnant women and fetuses or infants in each maternal and child health user group at each classification level of each classification attribute.

[0065] Based on the multidimensional health comparison factors of pregnant women and fetuses or infants in each category of each category of attributes, the maternal and child health deviations of each category of maternal and child health users were analyzed.

[0066] Based on the multidimensional health comparison factors and maternal and child health deviations of pregnant women and fetuses or infants in each user group under each classification level of each classification attribute, the system drives the intelligent adaptation of multiple services.

[0067] In this embodiment, the dynamic health status label is generated by threshold determination and trend analysis of effective physiological data and effective historical health data from effective multi-source data. It is used to characterize the real-time health status of women and children and will be updated as the data changes. For example, a label such as "high blood sugar risk" is generated based on the analysis results of physiological data such as blood sugar and blood pressure of pregnant women.

[0068] In this embodiment, maternal and child users refer to the group that uses the one-stop learning cloud platform for maternal and child health, including pregnant women, newborns, children and their family members, and medical professionals. The platform provides them with full-lifecycle health data management and intelligent education and training services.

[0069] In this embodiment, each classification level of the classification attribute is a classification method set for the refined management of maternal and child health. Taking pregnant women as an example, the classification attributes include pregnancy stage, health risk level, etc., and the classification level is a further subdivision of the attributes. For example, the pregnancy stage is divided into early pregnancy, mid pregnancy, and late pregnancy, and the health risk level is divided into low risk, medium risk, and high risk, so as to achieve accurate analysis and services.

[0070] In this embodiment, the multidimensional health comparison factor is used to compare and analyze multiple dimensions of maternal and child health status at each level of each attribute. For example, at the mid-pregnancy level for pregnant women, it includes BMI deviation, blood glucose similarity, and prenatal checkup trend consistency, in order to comprehensively understand the health differences with the population benchmark.

[0071] In this embodiment, the maternal and child health deviation is derived from the multidimensional health cross-comparison factor analysis of pregnant women and fetuses or infants at each division level of each division attribute. It is the deviation calculated by comparing factors at the corresponding levels of the two, such as the deviation of blood glucose similarity between pregnant women and their corresponding fetuses or infants.

[0072] In this embodiment, multiple services are provided: various services offered by the one-stop learning cloud platform for women and children, including health management services such as electronic health records, remote monitoring and early warning, and personalized health plan development; and cloud academy services such as diversified curriculum systems, adaptive learning systems, and skills training modules (such as VR newborn care simulation).

[0073] like Figure 2 As shown, to construct a multimodal maternal and child user profile with dynamic labels using multi-source data, providing a foundation for subsequent maternal and child health analysis and service adaptation, this paper proposes a method based on real-time collected maternal and child physiological data, historical maternal and child health records, and user learning behavior data. This method constructs a multimodal maternal and child user profile including dynamic health status labels and dynamic learning ability labels, including:

[0074] Through a multi-protocol adaptation engine, various physiological data of maternal and child users obtained from monitoring by various IoT medical devices are collected in real time as maternal and child physiological data, and maternal and child historical health records and user learning behavior data are also collected at the same time.

[0075] The real-time collected maternal and child physiological data, maternal and child historical health records, and user learning behavior data are transmitted to the cloud computing and storage layer for preliminary cleaning and format standardization to obtain effective multi-source data.

[0076] A multimodal profile of maternal and child health users, including dynamic health status labels and dynamic learning ability labels, is constructed based on effective multi-source data.

[0077] In this embodiment, the multi-protocol adaptation engine is a component in the system that supports access to various IoT medical devices, user terminal apps, and third-party medical information systems through open interfaces. It is compatible with multiple protocols such as MQTT, HTTP / HTTPS, etc., and realizes real-time collection and unified access of multi-source heterogeneous data, ensuring that data from different sources can smoothly enter the system for subsequent processing.

[0078] In this embodiment, the various physiological data of each female and child user obtained by each IoT medical device monitoring refers to the various physiological information obtained by real-time monitoring of female and child users through various IoT medical devices, such as blood glucose meters, such as blood glucose and blood pressure values ​​of pregnant women, fetal heart rate of fetuses, and body temperature of infants.

[0079] In this embodiment, the maternal and child health history record includes past health-related information of maternal and child users, such as the pregnant woman's past prenatal check-up reports and medical history, and the child's vaccination records and past illnesses, which can reflect the long-term health changes of maternal and child users.

[0080] In this embodiment, user learning behavior data refers to data generated by maternal and child health users during their use of the cloud academy service, such as the user's learning duration, learning progress, course completion status, frequency of interaction with learning content, and preferences for different types of courses.

[0081] In this embodiment, the cloud computing and storage layer, employing a hybrid cloud architecture, is a crucial component for system data processing and storage. On one hand, it utilizes distributed data storage, employing time-series databases to monitor data, relational databases to store business data, and object storage to store multimedia course resources. On the other hand, a data platform cleans, transforms, and standardizes the collected data to form a unified data model, providing data services for upper-layer applications.

[0082] In this embodiment, real-time collected maternal and child physiological data, maternal and child historical health records, and user learning behavior data are transmitted to the cloud computing and storage layer for preliminary cleaning and format standardization to obtain effective multi-source data. The three types of data collected in real-time from different sources are transmitted to the cloud computing and storage layer. At this layer, preliminary cleaning operations such as noise removal and error correction are performed, and the data is standardized according to the system's required format. This transforms the diverse data from different sources into effective multi-source data with a unified format that meets the requirements of subsequent processing, laying the foundation for building user profiles and other subsequent analyses and applications.

[0083] like Figure 2 As shown, to generate dynamic health status and learning ability labels from effective multi-source data, and then construct a complete multimodal maternal and child user profile to achieve a comprehensive depiction of maternal and child health and learning status, this paper proposes to construct a multimodal maternal and child user profile based on effective multi-source data, including dynamic health status labels and dynamic learning ability labels, as follows:

[0084] By performing threshold determination and trend analysis on valid physiological data and valid historical health data from valid multi-source data, dynamic health status labels are generated.

[0085] Cluster analysis is performed on effective learning behavior data from effective multi-source data to generate dynamic learning ability tags that include preferred course types, learning abilities, and content interaction frequency;

[0086] By aggregating the dynamic health status tags and dynamic learning ability tags of each maternal and child health user, a multimodal maternal and child health user profile is constructed.

[0087] In this embodiment, effective physiological data refers to data that accurately reflects the physiological condition of women and children after initial cleaning and format standardization by cloud computing and storage layers. Examples include blood glucose and blood pressure data of pregnant women collected from IoT medical devices after noise reduction and format standardization, as well as heart rate and body temperature data of fetuses or infants. This data can be used for subsequent analysis of the health status of women and children.

[0088] In this embodiment, valid historical health data refers to usable past health-related records of women and children after processing by cloud computing and storage layers. Data such as standardized and organized prenatal checkup reports and medical records for pregnant women, as well as children's growth and health records, can provide a basis for understanding the development of maternal and child health and are important references for generating dynamic health status labels.

[0089] In this embodiment, dynamic health status labels are generated by performing threshold determination and trend analysis on valid physiological data and valid historical health data from valid multi-source data. For the processed valid physiological and historical health data, a normal range threshold is set, and the data is compared to this threshold while observing the trend of data changes over time. For example, if a pregnant woman's blood glucose level consistently exceeds the normal threshold and shows an upward trend, a dynamic health status label such as "hyperglycemia risk" may be generated, thereby reflecting the dynamic changes in maternal and child health status in real time.

[0090] In this embodiment, effective learning behavior data refers to data processed at the cloud computing and storage layer that can be used to analyze the learning behavior characteristics of female and child users. This data includes information such as the user's course learning duration, learning order, completion progress, and interaction with the learning content on the cloud academy platform. After processing, this data supports the generation of dynamic learning ability tags.

[0091] In this embodiment, cluster analysis is performed on effective learning behavior data from valid multi-source data to generate dynamic learning ability tags that include preferred course types, learning abilities, and content interaction frequency. Cluster analysis is used to process the effective learning behavior data, grouping data with similar learning behavior patterns into one category. For example, analysis reveals that some users spend a lot of time and interact frequently with prenatal nutrition courses, indicating that their preferred course type is prenatal nutrition. Learning abilities are then assessed based on course difficulty and completion speed. Content interaction frequency is statistically analyzed based on actions such as clicking on course content and asking questions, thereby generating dynamic learning ability tags.

[0092] In this embodiment, the dynamic health status tags and dynamic learning ability tags of each maternal and child health user are aggregated to construct a multimodal maternal and child health user profile. This involves integrating the dynamic health status tags and dynamic learning ability tags generated separately for each user to comprehensively characterize user features from two dimensions: health and learning behavior, thus forming a multimodal maternal and child health user profile. For example, a pregnant woman's profile might be a combination of a health status tag indicating high blood sugar risk and learning ability tags indicating a preference for prenatal exercise courses, strong learning ability, and high frequency of content interaction, laying the foundation for providing personalized services.

[0093] To provide a basis for analyzing maternal and child health biases, this paper proposes a multimodal maternal and child user profile based on all maternal and child users. This profile is created by categorizing the health profiles of pregnant women and fetuses or infants according to preset attributes and levels, and determining multidimensional health comparison factors. The comparison factors include:

[0094] Extract the health profiles of pregnant women and fetuses or infants from the multimodal maternal and child health profiles of various maternal and child health users;

[0095] Based on a variety of preset pregnant and postpartum women classification attributes and corresponding classification levels, the health profiles of pregnant and postpartum women among all maternal and child health users are classified to obtain a group of pregnant and postpartum women health profiles under each classification level of each pregnant and postpartum women classification attribute.

[0096] Based on a variety of preset fetal or infant classification attributes and corresponding classification levels, the fetal or infant health profiles of all maternal and child health users are classified to obtain a group of fetal or infant health profiles under each classification level of each fetal or infant classification attribute.

[0097] Based on the health profiles of pregnant women and fetuses or infants in each user group of maternal and child health services at each level of each attribute, we analyzed the multidimensional health comparison factors of pregnant women and fetuses or infants in each user group at each level of each attribute.

[0098] In this embodiment, a maternal health profile and a fetal or infant health profile are extracted from the multimodal maternal and child health profiles of each user. From the multimodal maternal and child health profiles that comprehensively reflect the user's health and learning behavior, health characteristic information related to the pregnant woman is extracted to form a maternal health profile, and health characteristic information related to the fetus or infant is extracted to form a fetal or infant health profile. For example, from a user's multimodal profile, information such as the pregnant woman's prenatal checkup indicators and physical condition is compiled into a maternal health profile, and information such as fetal development indicators is grouped into a fetal health profile.

[0099] In this embodiment, the maternal health profile is a collection of information related to the health of pregnant women, including physiological indicators at different stages of pregnancy (such as changes in blood pressure, blood sugar, and weight), health risk factors (such as family medical history and age risk), and health management needs (such as the need for nutritional guidance and psychological counseling), which comprehensively shows the health status of pregnant women during pregnancy.

[0100] In this embodiment, the fetal or infant health profile is a collection of information that mainly records the health status of the fetus or infant. For the fetus, it includes growth and development indicators (such as biparietal diameter, femur length, fetal heart rate, etc.); for the infant, it includes physical development data (such as height, weight, head circumference), health status (such as whether the child is sick, vaccination status, etc.) to reflect the health status of the fetus or infant.

[0101] In this embodiment, there are multiple preset attributes for classifying pregnant and postpartum women and corresponding classification levels: these are pre-defined attributes used to classify the health profiles of pregnant and postpartum women and the sub-levels under each attribute. For example, the classification attributes may include pregnancy stage (including early pregnancy, mid pregnancy, and late pregnancy), health risk level (divided into low risk, medium risk, and high risk), and health management needs (such as nutritional needs, exercise needs, and psychological needs).

[0102] In this embodiment, based on multiple preset pregnancy and childbirth classification attributes and corresponding classification levels, the health profiles of pregnant and childbirth women among all maternal and child health users are classified to obtain groups of health profiles of pregnant and child health women under each classification level of each pregnancy and childbirth classification attribute. According to the preset classification attributes and levels, the health profiles of pregnant and child health women among all maternal and child health users are categorized. For example, using pregnancy stage as a classification attribute and early pregnancy as a classification level, all health profiles of pregnant and child health women in early pregnancy are grouped together to form a group of health profiles of pregnant and child health women under the early pregnancy classification level. Similarly, this classification operation is performed on other classification attributes and levels.

[0103] In this embodiment, multiple pre-defined fetal or infant classification attributes and corresponding classification levels are used: similar to the classification of pregnant women, attributes and sub-levels for classifying fetal or infant health profiles are pre-defined. Classification attributes can be growth and development stages (such as early pregnancy, mid-pregnancy, and late pregnancy for fetuses, and neonatal, infancy, and early childhood for infants), health status categories (such as healthy, with minor illnesses, and with serious illnesses), etc., with each attribute further subdivided into different levels.

[0104] In this embodiment, based on preset multiple fetal or infant classification attributes and corresponding classification levels, the fetal or infant health profiles of all maternal and child users are classified to obtain fetal or infant health profile groups under each classification level of each fetal or infant classification attribute. According to the aforementioned classification attributes and levels, the fetal or infant health profiles of all maternal and child users are grouped. For example, using growth and development stage as an attribute, and the neonatal period as one level, all infant health profiles in the neonatal period are grouped together, forming a fetal or infant health profile group under the neonatal period classification level. Other attributes and levels are operated on in the same way.

[0105] In this embodiment, based on the health profiles of pregnant women and fetuses or infants among each user group under each classification level of each attribute, multidimensional health comparison factors are analyzed for each user group. By comparing the health profiles of pregnant women (or fetuses or infants) under each corresponding classification level, factors reflecting differences or correlations in their health status are identified from multiple dimensions. For example, under the corresponding level of mid-pregnancy for pregnant women, the analyzed multidimensional health comparison factors might include the proportion of abnormal weight gain in pregnant women and abnormal blood sugar levels in pregnant women. Under the corresponding level of pregnancy in fetuses, the multidimensional health comparison factors might include the proportion of abnormal biparietal diameter growth and abnormal fetal development indicators.

[0106] To accurately identify maternal and child health issues by using the multidimensional health comparison factors of pregnant women and fetuses or infants at each classification level as maternal and child health deviations, this paper proposes a method based on multidimensional health comparison factors of pregnant women and fetuses or infants at each classification level of each classification attribute among maternal and child health users. This method analyzes the maternal and child health deviations of each user, including:

[0107] By mapping each level of the classification attributes of pregnant women to each level of the classification attributes of fetuses or infants, the classification levels of each unified classification requirement are obtained.

[0108] The deviation of the multidimensional health comparison factors of pregnant women and fetuses or infants in each maternal and child health user group under each unified classification level is taken as the maternal and child health deviation of each maternal and child health user group.

[0109] In this embodiment, a one-to-one correspondence is established between the classification levels of each pregnant woman's attributes and the classification levels of each fetus or infant's attributes to obtain the classification levels for each unified classification requirement. This involves matching the various classification attributes and their sub-levels set for pregnant women with the corresponding classification attributes and sub-levels set for fetuses or infants according to a certain logical relationship. For example, the early stage of pregnancy for pregnant women is correlated with the early stage of pregnancy for fetuses, and the health risk level of pregnant women and high risk is correlated with the health status category of fetuses and serious diseases. Through this correspondence, a series of classification levels based on unified classification requirements are obtained, facilitating the comparison and analysis of the health status of pregnant women and fetuses or infants from a consistent perspective.

[0110] In this embodiment, the deviation of multidimensional health comparison factors for pregnant women and fetuses or infants among each user group is analyzed under each unified classification level of the needs. Based on the previously obtained unified classification levels of the needs, for each user group, the degree of difference in multidimensional health comparison factors for pregnant women and fetuses or infants under each level is compared. For example, under the unified classification level of mid-pregnancy for pregnant women and mid-pregnancy for fetuses, multidimensional health comparison factors include the relationship between abnormal weight gain ratio of pregnant women and abnormal weight gain ratio of fetuses, and the correlation between abnormal blood glucose levels of pregnant women and abnormal blood glucose levels of fetuses. The differences or deviations of these factors in terms of values ​​and trends are calculated. This difference or deviation is the deviation of the multidimensional health comparison factors, which reflects the inconsistency of women and children in the corresponding health dimensions and provides a basis for subsequent health management.

[0111] like Figure 3 As shown, to determine the two-dimensional feature coordinates of health and service adaptation by calculating the benchmark range, and to provide a quantitative basis for intelligent adaptation of multiple services, a multi-dimensional health horizontal comparison factor and maternal and child health deviation are proposed based on the pregnant women and fetuses or infants in each category of each classification attribute at each classification level. This drives the intelligent adaptation of multiple services, including:

[0112] Based on the multidimensional health comparison factors of pregnant women and fetuses or infants in all maternal and child users at each level of each classification attribute, the baseline range of each dimension of the horizontal comparison factors of pregnant women at each level of each classification attribute, and the baseline range of each dimension of the horizontal comparison factors of fetuses or infants at each level of each classification attribute are calculated.

[0113] Based on the baseline range of horizontal comparison factors of pregnant women and fetuses or infants at each level of each classification attribute, multidimensional health horizontal comparison factors and maternal and child health deviations, the two-dimensional feature coordinates of health and service adaptation of pregnant women and fetuses or infants in each maternal and child user at each level of each classification attribute are determined.

[0114] Based on the pregnant women, fetuses, or infants among the users of each maternal and child health service, the two-dimensional feature coordinates of health and service adaptation under each classification level of each classification attribute drive the intelligent adaptation of multiple services.

[0115] In this embodiment, for all maternal and child health users, the situation of pregnant women and fetuses or infants in each segmentation attribute and corresponding segmentation level needs to be considered separately. Taking pregnant women as an example, the segmentation attributes include pregnancy stage and health risk level, and the segmentation levels are like the early pregnancy, mid pregnancy, and late pregnancy stages. For each such combination of segmentation attributes and levels, the multidimensional health comparison factors of pregnant women in all maternal and child health users under that combination are collected.

[0116] For example, under the combination of mid-pregnancy and low health risk level, multidimensional health comparison factors include BMI deviation and blood glucose similarity. By statistically analyzing these factor data from a large number of pregnant women and children in this combination—for example, calculating the mean and standard deviation—a reasonable range is determined. This range serves as the baseline range for this dimension of comparison factors within this classification level for pregnant women. Assuming BMI deviation data for 100 pregnant women in the mid-pregnancy and low health risk level combination are collected, these data are first summed to a total of 120. Dividing this by 100 yields a mean of 1.2. Next, the difference between each data point and the mean is calculated, squared, and summed to obtain 10. Dividing this by 100 yields 0.1. The square root of 0.1 gives a standard deviation of approximately 0.32. Centered on the mean, the baseline range is determined by adding or subtracting twice the standard deviation from the mean. The lower limit is 1.2 and 2×0.32=0.56, and the upper limit is 1.2+2×0.32=1.84. That is, the baseline range of BMI deviation under this combination is approximately 0.56 to 1.84. The same method can be used to determine the baseline range of other multidimensional health cross-comparison factors such as blood glucose similarity.

[0117] The same method is applied to fetuses or infants. Under each combination of categorization attributes and levels, statistical analysis is performed on the corresponding multidimensional health comparison factor data to determine the baseline ranges for each dimension of the comparison factors for fetuses or infants under each categorization level of each attribute. For example, under the combination of gestational age and normal development in the fetus, the baseline ranges for factors such as biparietal diameter deviation and fetal heart rate similarity are determined. These baseline ranges can serve as reference standards for measuring maternal and child health status.

[0118] To determine health attainment levels based on multidimensional health comparison factors, benchmark ranges, and maternal and child health deviations, and to prioritize service adaptation based on deviation ratios, thereby establishing two-dimensional feature coordinates for health and service adaptation among maternal and child health users, this paper proposes a method to accurately assess the degree of health status and service demand. This method is based on the benchmark ranges of multidimensional health comparison factors for pregnant women and fetuses or infants at each level of each attribute, multidimensional health comparison factors, and maternal and child health deviations. The method determines the two-dimensional feature coordinates for health and service adaptation among pregnant women and fetuses or infants at each level of each attribute for each user group. These coordinates include:

[0119] The proportion of single-dimensional comparative factors that did not exceed the corresponding benchmark range among the multidimensional health comparative factors of pregnant women in each category of each category of each attribute was statistically analyzed, and this proportion was used as the health attainment of pregnant women in each category of each category of each attribute at each level.

[0120] Based on the ratio of the maternal and child health deviation of pregnant women in each category of each category at each category level to the maximum maternal and child health deviation of pregnant women in each category at all categories at all categories, the service adaptation priority of pregnant women in each category at each category level is determined.

[0121] The proportion of single-dimensional comparative factors that did not exceed the corresponding benchmark range in the multidimensional health comparative factors of fetuses or infants in each category of each category of each attribute was statistically analyzed. This proportion was used as the health attainment of fetuses or infants in each category of each category of each attribute in each category of each maternal and child health user.

[0122] Based on the ratio of the maternal and child health deviation of the fetus or infant in each user group under each classification level of each classification attribute to the maximum maternal and child health deviation of the fetus or infant in each user group under all classification levels of all classification attributes, the service adaptation priority of the fetus or infant in each user group under each classification level of each classification attribute is determined.

[0123] Based on the health attainment and service adaptation priority of pregnant women in each category of each attribute at each level among the maternal and child health users, two-dimensional feature coordinates of health and service adaptation for the corresponding pregnant women in each category of each attribute at each level are determined. Similarly, based on the health attainment and service adaptation priority of fetuses or infants in each category of each attribute at each level among the maternal and child health users, two-dimensional feature coordinates of health and service adaptation for the corresponding fetuses or infants in each category of each attribute at each level are determined.

[0124] In this embodiment, the proportion of single-dimensional comparative factors that do not exceed the corresponding benchmark range among the multi-dimensional health comparative factors for pregnant women in each user group under each classification level of each classification attribute is statistically analyzed. For each pregnant woman in the user group, under each classification attribute (such as pregnancy stage, health risk level, etc.) and its corresponding classification level (such as early pregnancy, mid pregnancy, late pregnancy, low risk, medium risk, high risk), there are multiple dimensions of health comparative factors (such as BMI deviation value, blood glucose similarity, etc.). Each single-dimensional comparative factor is compared with the previously calculated corresponding benchmark range, and the number of single-dimensional comparative factors that do not exceed the benchmark range is counted. This number is then divided by the total number of single-dimensional comparative factors to obtain the proportion of single-dimensional comparative factors that do not exceed the corresponding benchmark range for that pregnant woman under this classification attribute and level. For example, under the classification of mid pregnancy and low health risk level, there are 5 single-dimensional comparative factors, of which 3 do not exceed the corresponding benchmark range, so the proportion of dimensions is 3 ÷ 5 = 0.6.

[0125] In this embodiment, the dimensional proportion of single-dimensional horizontal comparison factors that do not exceed the corresponding benchmark range among the multidimensional health horizontal comparison factors of fetuses or infants in each user group under each classification level of each classification attribute is statistically analyzed. Similar to the statistical method for pregnant women, for each fetus or infant in each user group, under various classification attributes and classification levels, the multidimensional health horizontal comparison factors (such as biparietal diameter deviation value, fetal heart rate similarity, etc.) are compared one by one with the corresponding benchmark range. The number of single-dimensional horizontal comparison factors that do not exceed the benchmark range is counted, and the dimensional proportion is obtained by dividing by the total number. For example, under the classification of gestational age and normal development in fetuses, there are 4 single-dimensional horizontal comparison factors, 2 of which do not exceed the benchmark range, and the dimensional proportion is 2÷4=0.5.

[0126] In this embodiment, based on the health attainment and service adaptation priority of pregnant women among the maternal and child health users at each level of each attribute, the two-dimensional feature coordinates of health and service adaptation for the corresponding pregnant women at each level of each attribute are determined. Similarly, based on the health attainment and service adaptation priority of fetuses or infants among the maternal and child health users at each level of each attribute, the two-dimensional feature coordinates of health and service adaptation for the corresponding fetuses or infants at each level of each attribute are determined. Here, the health attainment is the dimensional proportion of the single-dimensional horizontal comparison factor that does not exceed the corresponding benchmark range, as previously statistically calculated. The service adaptation priority is determined based on factors such as maternal and child health deviations (e.g., the greater the deviation, the higher the priority). Taking pregnant women as an example, using the health attainment under each attribute and level as the horizontal axis and the service adaptation priority as the vertical axis, a point is formed on a two-dimensional plane. This point represents the two-dimensional feature coordinates of health and service adaptation for the corresponding pregnant woman at that attribute and level. Similarly, for fetuses or infants, the corresponding two-dimensional feature coordinates are determined in the same way using their health attainment and service adaptation priority under each classification attribute and level. For example, if a pregnant woman has a health attainment of 0.4 and a service adaptation priority of 0.8 under the late pregnancy and high-risk classification, then her two-dimensional feature coordinates for health and service adaptation are (0.4, 0.8); if a fetus has a health attainment of 0.3 and a service adaptation priority of 0.9 under the late pregnancy and developmental delay classification, then its two-dimensional feature coordinates are (0.3, 0.9).

[0127] like Figure 3 As shown, to generate a deviation curve from the standard adaptation curve by sorting and fitting the two-dimensional feature coordinates of health and service adaptation for pregnant women and fetuses or infants, and providing input for the intelligent adaptation model to drive intelligent adaptation such as health education, this paper proposes to drive intelligent adaptation of multiple services based on the two-dimensional feature coordinates of health and service adaptation for pregnant women and fetuses or infants in each user group, at each division level of each attribute, including:

[0128] Based on the preset sorting rules, the two-dimensional feature coordinates of health and service adaptation of pregnant women in each maternal and child health user under all classification levels of all classification attributes are sorted and curve fitted to obtain the first health and service adaptation curve.

[0129] Based on the preset sorting rules, the two-dimensional feature coordinates of health and service adaptation of fetuses or infants in each maternal and child user under all classification levels of all classification attributes are sorted and curve fitted to obtain the second health and service adaptation curve.

[0130] Generate the deviation curve between the first health and service adaptation curve and the first health and service standard adaptation curve as the first input curve, and use the deviation curve between the second health and service adaptation curve and the second health and service standard adaptation curve as the second input curve.

[0131] Intelligent adaptation of multiple services is driven by the first input curve, the second input curve, and the intelligent adaptation model.

[0132] In this embodiment, the preset sorting rule is a pre-defined criterion used to determine the order of the health and service adaptation two-dimensional feature coordinates of pregnant women or fetuses / infants among each maternal and child health user under different classification attributes and levels. For example, the health and service adaptation two-dimensional feature coordinates under all classification levels are sorted in the order of early pregnancy, mid-pregnancy, late pregnancy, low risk, medium risk, and high risk.

[0133] In this embodiment, based on a preset sorting rule, the health and service adaptation two-dimensional feature coordinates of pregnant women among all users are sorted and curve-fitted under all classification levels of all classification attributes to obtain the first health and service adaptation curve: According to the preset sorting rule, the health and service adaptation two-dimensional feature coordinates of all pregnant women among all users under each classification attribute (such as pregnancy stage, health risk level, etc.) and its classification level (such as early pregnancy, mid pregnancy, late pregnancy, low risk, medium risk, high risk) are arranged in order. Then, the mathematical method of curve fitting is used to process these ordered coordinate points to find a curve that can best approximate the distribution trend of these points. This curve is the first health and service adaptation curve.

[0134] In this embodiment, based on preset sorting rules, the two-dimensional feature coordinates of health and service adaptation for fetuses or infants in each maternal and child user under all classification attributes and all classification levels are sorted and curve-fitted to obtain a second health and service adaptation curve. Similar to the operation for pregnant women, the two-dimensional feature coordinates of health and service adaptation for fetuses or infants in each maternal and child user under various classification attributes and classification levels are sorted according to preset sorting rules. Then, through curve fitting, a curve that reflects the overall adaptation characteristics of fetuses or infants under different health and service conditions is obtained, namely the second health and service adaptation curve.

[0135] In this embodiment, the first health and service standard fit curve represents the ideal fit between maternal health and services, serving as a standard reference for measuring the actual situation. It may be determined based on numerous successful maternal health management cases, medical standards, or industry consensus. By comparing the actual first health and service fit curve with this standard curve, the difference between the current maternal health and service fit and the ideal state can be clearly seen.

[0136] In this embodiment, a deviation curve is generated between the first health and service fit curve and the first health and service standard fit curve: corresponding points of the first health and service fit curve and the first health and service standard fit curve are compared, and the difference between them on the ordinate corresponding to the same horizontal axis is calculated. Using these differences as data, a new curve is plotted. This curve reflects the degree and trend of deviation between the actual maternal health and service fit and the ideal standard, providing a basis for further analysis and service adjustment.

[0137] In this embodiment, the second health and service standard adaptation curve, similar to the first health and service standard adaptation curve, is a curve representing the ideal health and service adaptation relationship for the fetus or infant, serving as a standard reference for assessing the health and service adaptation of the fetus or infant. Based on medical knowledge and the laws of infant health development, it is used to determine the gap between the actual health and service adaptation of the fetus or infant and the ideal state.

[0138] In this embodiment, the deviation curve between the second health and service fit curve and the second health and service standard fit curve is calculated by plotting the difference between the second health and service fit curve and the second health and service standard fit curve on the same horizontal axis. This curve visually demonstrates the degree and variation in the actual health and service fit of the fetus or infant that deviates from the ideal standard.

[0139] In this embodiment, the intelligent adaptation model is a model built using data analysis and algorithms. Based on input information, such as the deviation curve between health and service adaptation curves, it automatically adjusts and optimizes the delivery of multiple services to achieve more accurate and personalized services. This model may incorporate machine learning algorithms and artificial intelligence technologies. Through learning and analyzing large amounts of data, it identifies the intrinsic relationship between health status and service needs, thereby making intelligent service adaptation decisions. During the training process of this intelligent adaptation model:

[0140] The machine learning algorithm takes as input the deviation curve between health and service adaptation curves and outputs as manually optimized service adaptation decision rules. During training, the parameters are continuously adjusted. These parameters determine how the model processes and analyzes the input data to achieve the best service adaptation decision effect, and finally forms a model that can be used in practical applications to make intelligent service adaptation decisions based on new input information.

[0141] In this embodiment, intelligent adaptation of multiple services is driven by the first and second input curves and the intelligent adaptation model: these two deviation curves are input into the intelligent adaptation model, and the model analyzes the health and service adaptation deviations reflected by the curves according to its internal algorithms and rules, thereby intelligently adjusting and adapting multiple services such as health management services (such as electronic health record updates, remote monitoring frequency adjustments, and personalized health plan modifications) and cloud academy services (such as course recommendation adjustments and learning progress planning optimization) to meet the actual health and learning needs of maternal and child users and improve the quality and effectiveness of services.

[0142] To establish a three-tiered classification system based on the intended use and privacy sensitivity of effective multi-source data, and to dynamically respond to data access requests by incorporating relevant attributes of the user initiating the request, thereby achieving reasonable data classification management and secure access control, the following additional measures are proposed:

[0143] Based on the purpose, privacy sensitivity, and subject matter of each data block in the valid multi-source data, all data blocks are divided into a three-level classification system;

[0144] Based on the user role, domain attributes, and data ownership of the initiating user of the data access request, combined with a three-level classification system, the system dynamically responds to the data access request.

[0145] In this embodiment, each data block in the effective multi-source data refers to the various components contained in the effective multi-source data after preliminary cleaning and format standardization by the cloud computing and storage layer. These data blocks come from diverse sources, such as physiological data collected from IoT medical devices, user learning behavior data, and maternal and child health records, all of which can be considered as different data blocks in the effective multi-source data. Each data block has its own characteristics and information content, serving as the basic unit for subsequent data classification and management.

[0146] In this embodiment, the purpose, privacy sensitivity, and subject matter of each data block are defined as follows: Purpose refers to the role of each data block in maternal and child health management and education services. For example, a data block recording blood glucose monitoring data of pregnant women is used to assess their blood glucose health status. Privacy sensitivity indicates the degree to which the data involves personal privacy. Data blocks such as personal prenatal check-up reports have higher privacy sensitivity, while general health education templates have relatively lower privacy sensitivity. Subject matter refers to the entity associated with the data, which could be a specific maternal and child health user, a medical field, or the entire system. Understanding these characteristics helps in the proper classification and management of data.

[0147] In this embodiment, based on the purpose, privacy sensitivity, and subject matter of each data block in the valid multi-source data, all data blocks are divided into a three-level classification system: According to the different purposes, privacy sensitivity, and subject matter of each data block, they are divided into three levels. System data, as the first level, contains basic public knowledge, such as standard medical terminology databases and general health education templates. This type of data is accessible to all authorized users because it has a wide range of uses and low privacy sensitivity. Domain data, as the second level, is professional data specific to a particular domain, such as the diagnosis and treatment guidelines and expert courses of a hospital's obstetrics and gynecology department. Access requires the same domain permissions, and its purpose is specific to that domain, with moderate privacy sensitivity. User data, as the third level, is the user's personal private data, such as personal prenatal check-up reports, learning progress, and family records. Access is limited to the user or authorized persons because it has high privacy sensitivity and belongs to a specific user.

[0148] In this embodiment, the user initiating the data access request refers to the user who submits a data access request to the system on the one-stop learning cloud platform for women and children. This user can be the woman or child user herself, her family members, medical professionals, or other relevant personnel. These users initiate data access requests to the system for various purposes, such as viewing health records or acquiring professional knowledge.

[0149] In this embodiment, the initiating user's role, domain attribute, and data ownership relationship are defined as follows: The user role indicates the initiating user's identity category within the system, such as a medical professional, pregnant woman, or family member. Different roles have different permissions and data requirements. The domain attribute relates to the user's professional field or business scope, such as obstetrics and gynecology or pediatrics, which affects their access permissions to data in that specific field. The data ownership relationship clarifies the association between the user and the requested data, such as whether the data belongs to the user or whether the user is authorized to access the data. These factors combined determine the user's access permissions to different data blocks.

[0150] In this embodiment, the system dynamically responds to data access requests based on the user role, domain attribute, and data ownership relationship of the initiating user, combined with a three-level classification system. Upon receiving a data access request, the system first analyzes the initiating user's user role, domain attribute, and data ownership relationship, while also referencing the three-level data classification system. For example, if a medical professional (user role) from the obstetrics and gynecology domain (domain attribute) requests access to a pregnant woman's prenatal checkup report (user data, belonging to that pregnant woman), the system will decide whether to allow access based on whether the medical professional has been authorized by the pregnant woman (data ownership relationship) and the user data access rules in the three-level classification system. If authorized and in accordance with the rules, access is allowed; otherwise, access is denied. This dynamic response mechanism ensures both data security and privacy while meeting the reasonable data access needs of different users.

[0151] This invention provides an embodiment of a data management and control system based on a one-stop learning cloud platform for women and children, comprising:

[0152] The user profile building module is used to build a multimodal maternal and child user profile that includes dynamic health status tags and dynamic learning ability tags based on real-time collected maternal and child physiological data, maternal and child historical health records, and user learning behavior data.

[0153] The multidimensional horizontal comparison module is used to determine the multidimensional health horizontal comparison factors of pregnant women and fetuses or infants in each user group under each classification level of each classification attribute, based on the multimodal maternal and child user profile of all maternal and child users.

[0154] The health deviation analysis module is used to analyze the maternal and child health deviations of each user based on the multidimensional health comparison factors of pregnant women and fetuses or infants in each category at each category level of each category attribute.

[0155] The multi-directional intelligent adaptation module is used to drive the intelligent adaptation of multiple services based on the multi-dimensional health horizontal comparison factors and maternal and child health deviations of pregnant women and fetuses or infants in each category of each category attribute and at each category level.

[0156] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A data management and control method based on a one-stop learning cloud platform for women and children, characterized in that: include: Based on real-time collected maternal and child physiological data, maternal and child historical health records, and user learning behavior data, a multimodal maternal and child user profile is constructed, which includes dynamic health status tags and dynamic learning ability tags. Based on the multimodal maternal and child health user profiles of all users, this study determines the multidimensional health comparison factors for pregnant women and fetuses or infants within each user group, at each level of each classification attribute. This includes: extracting maternal health profiles and fetal or infant health profiles from the multimodal maternal and child health profiles of each user group; classifying maternal health profiles across all users based on preset maternal classification attributes and corresponding classification levels to obtain maternal health profile groups at each level of each maternal classification attribute; and classifying maternal health profiles across all users based on preset fetal or infant classification attributes and corresponding classification levels. The health profiles of fetuses or infants among young users are segmented to obtain a group of fetal or infant health profiles at each segmentation level of each fetal or infant segmentation attribute. Based on the group of fetal or infant health profiles of pregnant women and fetuses or infants among maternal and child users at each segmentation level of each segmentation attribute, the multidimensional health comparison factors of pregnant women and fetuses or infants among maternal and child users at each segmentation level of each segmentation attribute are analyzed. The multidimensional health comparison factors are used to compare and analyze multiple dimensions of maternal and child health status at each segmentation level of each segmentation attribute, so as to comprehensively understand the health differences with the group benchmark. Based on the multidimensional health comparison factors of pregnant women and fetuses or infants in each category of each classification attribute at each classification level, the maternal and child health deviations of each user are analyzed. This includes: mapping each classification level of each pregnant woman classification attribute to each classification level of each fetus or infant classification attribute to obtain each classification level of each unified classification requirement; and taking the deviation of the multidimensional health comparison factors of pregnant women and fetuses or infants in each category of each unified classification requirement as the maternal and child health deviation of each user. The maternal and child health deviation is derived from the multidimensional health comparison factor analysis of pregnant women and fetuses or infants in each category of each classification attribute at each classification level, and is the deviation calculated by comparing the factors at the corresponding levels of the two categories. Based on the multidimensional health comparison factors and maternal and child health deviations of pregnant women and fetuses or infants in each user group under each classification level of each classification attribute, the system drives the intelligent adaptation of multiple services.

2. The data management and control method based on the one-stop learning cloud platform for women and children according to claim 1, characterized in that, Based on real-time collected maternal and child physiological data, historical health records, and user learning behavior data, a multimodal maternal and child user profile is constructed, including dynamic health status tags and dynamic learning ability tags, comprising: Through a multi-protocol adaptation engine, various physiological data of maternal and child users obtained from monitoring by various IoT medical devices are collected in real time as maternal and child physiological data, and maternal and child historical health records and user learning behavior data are also collected at the same time. The real-time collected maternal and child physiological data, maternal and child historical health records, and user learning behavior data are transmitted to the cloud computing and storage layer for preliminary cleaning and format standardization to obtain effective multi-source data. A multimodal profile of maternal and child health users, including dynamic health status labels and dynamic learning ability labels, is constructed based on effective multi-source data.

3. The data management and control method based on the one-stop learning cloud platform for women and children according to claim 2, characterized in that, A multimodal user profile of women and children is constructed based on effective multi-source data, including dynamic health status tags and dynamic learning ability tags, comprising: By performing threshold determination and trend analysis on valid physiological data and valid historical health data from valid multi-source data, dynamic health status labels are generated. Cluster analysis is performed on effective learning behavior data from effective multi-source data to generate dynamic learning ability tags that include preferred course types, learning abilities, and content interaction frequency; By aggregating the dynamic health status tags and dynamic learning ability tags of each maternal and child health user, a multimodal maternal and child health user profile is constructed.

4. The data management and control method based on the one-stop learning cloud platform for women and children according to claim 1, characterized in that, Based on multidimensional health comparison factors and maternal and child health biases of pregnant women and fetuses or infants in each user group across different classification levels and attributes, the system drives intelligent adaptation of multiple services, including: Based on the multidimensional health comparison factors of pregnant women and fetuses or infants in all maternal and child users at each level of each classification attribute, the baseline range of each dimension of the horizontal comparison factors of pregnant women at each level of each classification attribute, and the baseline range of each dimension of the horizontal comparison factors of fetuses or infants at each level of each classification attribute are calculated. Based on the baseline range of horizontal comparison factors of pregnant women and fetuses or infants at each level of each classification attribute, multidimensional health horizontal comparison factors and maternal and child health deviations, the two-dimensional feature coordinates of health and service adaptation of pregnant women and fetuses or infants in each maternal and child user at each level of each classification attribute are determined. Based on the pregnant women, fetuses, or infants among the users of each maternal and child health service, the two-dimensional feature coordinates of health and service adaptation under each classification level of each classification attribute drive the intelligent adaptation of multiple services.

5. The data management and control method based on the one-stop learning cloud platform for women and children according to claim 4, characterized in that, Based on the baseline range of horizontal comparison factors for pregnant women and fetuses or infants at each level of each classification attribute, multidimensional health horizontal comparison factors, and maternal and child health bias, the two-dimensional feature coordinates of health and service adaptation for pregnant women and fetuses or infants among each maternal and child user are determined at each level of each classification attribute, including: The proportion of single-dimensional comparative factors that did not exceed the corresponding benchmark range among the multidimensional health comparative factors of pregnant women in each category of each category of each attribute was statistically analyzed, and this proportion was used as the health attainment of pregnant women in each category of each category of each attribute at each level. Based on the ratio of the maternal and child health deviation of pregnant women in each category of each category at each category level to the maximum maternal and child health deviation of pregnant women in each category at all categories at all categories, the service adaptation priority of pregnant women in each category at each category level is determined. The proportion of single-dimensional comparative factors that did not exceed the corresponding benchmark range in the multidimensional health comparative factors of fetuses or infants in each category of each category of each attribute was statistically analyzed. This proportion was used as the health attainment of fetuses or infants in each category of each category of each attribute in each category of each maternal and child health user. Based on the ratio of the maternal and child health deviation of the fetus or infant in each user group under each classification level of each classification attribute to the maximum maternal and child health deviation of the fetus or infant in each user group under all classification levels of all classification attributes, the service adaptation priority of the fetus or infant in each user group under each classification level of each classification attribute is determined. Based on the health attainment and service adaptation priority of pregnant women in each category of each attribute at each level among the maternal and child health users, two-dimensional feature coordinates of health and service adaptation for the corresponding pregnant women in each category of each attribute at each level are determined. Similarly, based on the health attainment and service adaptation priority of fetuses or infants in each category of each attribute at each level among the maternal and child health users, two-dimensional feature coordinates of health and service adaptation for the corresponding fetuses or infants in each category of each attribute at each level are determined.

6. The data management and control method based on the one-stop learning cloud platform for women and children according to claim 4, characterized in that, Based on the pregnant women and fetuses or infants among each group of maternal and child health users, the system uses two-dimensional feature coordinates for health and service adaptation at each level of each attribute classification to drive intelligent adaptation of multiple services, including: Based on the preset sorting rules, the two-dimensional feature coordinates of health and service adaptation of pregnant women in each maternal and child health user under all classification levels of all classification attributes are sorted and curve fitted to obtain the first health and service adaptation curve. Based on the preset sorting rules, the two-dimensional feature coordinates of health and service adaptation of fetuses or infants in each maternal and child user under all classification levels of all classification attributes are sorted and curve fitted to obtain the second health and service adaptation curve. Generate the deviation curve between the first health and service adaptation curve and the first health and service standard adaptation curve as the first input curve, and use the deviation curve between the second health and service adaptation curve and the second health and service standard adaptation curve as the second input curve. Intelligent adaptation of multiple services is driven by the first input curve, the second input curve, and the intelligent adaptation model.

7. The data management and control method based on the one-stop learning cloud platform for women and children according to claim 2, characterized in that, Also includes: Based on the purpose, privacy sensitivity, and subject matter of each data block in the valid multi-source data, all data blocks are divided into a three-level classification system; Based on the user role, domain attributes, and data ownership of the initiating user of the data access request, combined with a three-level classification system, the system dynamically responds to the data access request.

8. A data management and control system based on a one-stop learning cloud platform for women and children, characterized in that: include: The user profile building module is used to build a multimodal maternal and child user profile that includes dynamic health status tags and dynamic learning ability tags based on real-time collected maternal and child physiological data, maternal and child historical health records, and user learning behavior data. The multidimensional horizontal comparison module is used to determine the multidimensional health horizontal comparison factors for pregnant women and fetuses or infants in each user group based on the multimodal maternal and child user profiles of all users, at each classification level of each classification attribute. This includes: extracting maternal health profiles and fetal or infant health profiles from the multimodal maternal and child user profiles of each user; classifying the maternal health profiles of all users based on preset multiple maternal classification attributes and corresponding classification levels to obtain maternal health profile groups at each classification level of each maternal classification attribute; and classifying the maternal health profiles based on preset multiple fetal or infant classification attributes and corresponding classification levels. Next, the health profiles of fetuses or infants among all maternal and child health users are segmented to obtain a group of fetal or infant health profiles at each segmentation level of each fetal or infant segmentation attribute. Based on the group of fetal or infant health profiles of pregnant women and fetuses or infants among all maternal and child health users at each segmentation level of each segmentation attribute, the multidimensional health comparison factors of pregnant women and fetuses or infants among all maternal and child health users at each segmentation level of each segmentation attribute are analyzed. The multidimensional health comparison factors are used to compare and analyze multiple dimensions of maternal and child health status at each segmentation level of each segmentation attribute, so as to comprehensively understand the health differences with the group benchmark. The health deviation analysis module is used to analyze the maternal and child health deviations of each user based on the multidimensional health comparison factors of pregnant women and fetuses or infants under each classification level of each classification attribute. This includes: mapping each classification level of each pregnant woman classification attribute to each classification level of each fetus or infant classification attribute to obtain the classification levels of each unified classification requirement; and taking the deviation of the multidimensional health comparison factors of pregnant women and fetuses or infants under each classification level of each unified classification requirement as the maternal and child health deviation of each user. The maternal and child health deviation is derived from the multidimensional health comparison factor analysis of pregnant women and fetuses or infants under each classification level of each classification attribute, and is calculated by comparing the factors at the corresponding levels of the two. The multi-directional intelligent adaptation module is used to drive the intelligent adaptation of multiple services based on the multi-dimensional health horizontal comparison factors and maternal and child health deviations of pregnant women and fetuses or infants in each category of each category attribute and at each category level.

Images