Health monitoring data processing method, system and equipment and storage medium

Abstract

The embodiment of the invention provides a health monitoring data processing method, system and device and a storage medium, and the method comprises the steps: after obtaining multi-source health monitoring data through a plurality of health monitoring devices, carrying out the standardization processing of the multi-source health monitoring data, so as to obtain standard health data, and storing the standard health data in a distributed storage medium. Monitoring result data are obtained by calling the health state analysis model and taking the standard health data as input, so that the monitoring result data are stored through the data partition. According to the method, the consistency and comparability of data can be improved through standardized processing, independent data partitions and a data access mode based on multi-factor authentication are set based on user information, and the privacy protection function is enhanced. In addition, when the monitoring result data is stored, display parameters can be set to improve the compatibility between different devices, and therefore the problems that the health management system is insufficient in data accuracy, poor in privacy protection and poor in compatibility between the devices are solved.

Description

Technical Field

[0001] The present application relates to the technical field of health monitoring, and particularly to a method, system, device and storage medium for processing health monitoring data. Background Art

[0002] Health monitoring is a process of detecting physiological data through sensors that can detect information related to users' health indicators, and then analyzing and processing the physiological data to evaluate the users' health status. In order to perform health monitoring, a data integration system for intelligently monitoring health indicators is constructed. Among them, the system may include components such as wearable devices, mobile terminals, cloud platforms, and data analysis tools. Wearable devices can be used to collect users' physiological data, and components such as mobile terminals, cloud platforms, and data analysis tools are used to analyze the physiological data to provide health management suggestions and support.

[0003] As people pay more and more attention to their own health management, various products and services aimed at helping users track their health status can be constructed based on the theory of health monitoring. For example, the comprehensive health management platform provided by Company A can synchronize data with a variety of third-party health and fitness applications, detect users' physiological data such as heart rate, blood pressure, and exercise steps through wearable devices such as smart watches and smart bracelets, and analyze the detected physiological data through the comprehensive health management platform to achieve the tracking of multiple health indicators such as heart rate, sleep, and activity level.

[0004] However, since the data in the health monitoring process directly comes from users' physiological data, to a certain extent, these physiological data belong to users' privacy. And the above-mentioned health monitoring process directly uses the original data of physiological data for data processing and transmission, resulting in problems such as low data accuracy, insufficient privacy protection effect, and poor compatibility between different devices in the above-mentioned health monitoring process. Summary of the Invention

[0005] In view of this, embodiments of the present application provide a method, system, device and storage medium for processing health monitoring data to solve the problems of poor accuracy, security and compatibility in the process of processing health monitoring data.

[0006] According to one aspect of the present application, there is provided a method for processing health monitoring data, characterized in that the method includes:

[0007] Obtain multi-source health monitoring data, where the multi-source health monitoring data is data collected by multiple health monitoring devices within a preset monitoring period;

[0008] Perform standardization processing on the multi-source health monitoring data to obtain standard health data;

[0009] Store the standard health data in a distributed storage medium, and the distributed storage medium sets independent data partitions according to user information; the data partitions are set with data access permissions based on multi-factor authentication;

[0010] Invoke a health status analysis model, which is a machine learning model trained based on sample health data; the sample health data is labeled with a health pattern label and a potential risk label;

[0011] Input the standard health data into the health status analysis model to obtain the monitoring result data output by the health status analysis model, and the monitoring result data includes health pattern information and potential risk information;

[0012] Store the monitoring result data through the data partition.

[0013] According to another aspect of the present application, there is provided a health monitoring data processing system, and the system includes:

[0014] A data acquisition module for acquiring multi-source health monitoring data, where the multi-source health monitoring data is data collected by multiple health monitoring devices within a preset monitoring period;

[0015] A standardization module for performing standardization processing on the multi-source health monitoring data to obtain standard health data;

[0016] A data storage module for storing the standard health data in a distributed storage medium, and the distributed storage medium sets independent data partitions according to user information; the data partitions are set with data access permissions based on multi-factor authentication;

[0017] A model invocation module for invoking a health status analysis model, which is a machine learning model trained based on sample health data; the sample health data is labeled with a health pattern label and a potential risk label;

[0018] An analysis module for inputting the standard health data into the health status analysis model to obtain the monitoring result data output by the health status analysis model, and the monitoring result data includes health pattern information and potential risk information;

[0019] The data storage module is further configured to store the monitoring result data through the data partition.

[0020] According to another aspect of the present application, there is provided a computer device, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, and when the processor executes the program, the above-mentioned health monitoring data processing method is implemented.

[0021] According to another aspect of the present application, there is provided a storage medium having a computer program stored thereon, and when the program is executed by a processor, the above-mentioned health monitoring data processing method is implemented.

[0022] By means of the above technical solution, the embodiments of the present application provide a health monitoring data processing method, system, device and storage medium. After obtaining multi-source health monitoring data through multiple health monitoring devices, the method performs standardization processing on the multi-source health monitoring data to obtain standard health data, and stores the standard health data in a distributed storage medium. By invoking a health status analysis model and using the standard health data as input, monitoring result data is obtained, and thus the monitoring result data is stored by data partitioning. The method can improve the consistency and comparability of data through standardization processing, and enhance the privacy protection function by setting independent data partitions based on user information and a data access method based on multi-factor authentication. When storing the monitoring result data, display parameters can also be set to improve the compatibility between different devices, thereby solving the problems of insufficient data accuracy, poor privacy protection and poor compatibility between devices in the health management system.

[0023] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the specific embodiments of the present application are specifically given below. Description of the Drawings

[0024] The drawings described herein are used to provide a further understanding of the present application, and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings:

[0025] Figure 1 It is a schematic structural diagram of a data integration system provided by an embodiment of the present application;

[0026] Figure 2 It is a schematic flowchart of a health monitoring data processing method provided by an embodiment of the present application;

[0027] Figure 3 It is a schematic diagram of the hierarchical relationship of a data integration system provided by an embodiment of the present application;

[0028] Figure 4 It is a schematic flowchart of data encryption provided by an embodiment of the present application;

[0029] Figure 5 It is a schematic flowchart of setting display parameters provided by an embodiment of the present application;

[0030] Figure 6 It is a schematic flowchart of data access provided by an embodiment of the present application;

[0031] Figure 7 Schematic structural diagram of the health monitoring data processing system provided by the embodiment of the present application;

[0032] Figure 8 Schematic structural diagram of the computer device provided by the embodiment of the present application. Detailed implementation manners

[0033] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0034] In the embodiment of the present application, the health monitoring is a process of detecting physiological data through sensors, devices, medical devices, etc. for detecting information related to the user's health indicators, and then analyzing and processing the physiological data to evaluate the user's health status. Among them, physiological data refers to various quantitative information related to the physiological functions and health status of the human body. Physiological data can reflect the biological, chemical, and physical processes inside the human body, as well as the operating status of each system of the body.

[0035] Physiological data can be measured or recorded through scientific methods and technical means, so as to perform quantitative or qualitative analysis related to the physiological functions, metabolic activities, health status, or disease characteristics of the human body. For physiological data, sensors for detecting information related to the user's health indicators can be used for data collection. In some embodiments, the sensors may include heart rate and cardiac activity sensors, body fluid sensors, motion and body movement sensors, physiological parameter sensors, etc. For example, photoplethysmogram sensors, sweat sensors, acceleration sensors, temperature sensors, etc. These sensors can be connected to an electronic device with data processing functions, so as to collect physiological data in multiple dimensions and uniformly summarize them to the electronic device for analysis and processing through the electronic device. For the convenience of description, in some embodiments of the present application, the sensors for detecting information related to the user's health indicators are referred to as monitoring sensors.

[0036] In some embodiments, the monitoring sensors can also be integrated into specific terminal devices to form intelligent devices with health monitoring functions, so as to realize the monitoring of the user's health status. For the convenience of description, in some embodiments of the present application, the intelligent devices with health monitoring functions are referred to as health monitoring devices. For example, the health monitoring device can be a wearable device, a mobile terminal, a virtual reality device, a healthcare device, etc. After the user wears or carries the health monitoring device, the health monitoring device can collect the user's physiological data through the built-in monitoring sensors, and based on the physiological data processing algorithm built into the intelligent device, use the physiological data for data analysis to obtain the monitoring result data.

[0037] Such asFigure 1 As shown, in some embodiments, an intelligent monitoring health index data integration system can also be constructed based on multiple health monitoring devices and data processing devices. The data integration system can integrate the health monitoring data collected by multiple health monitoring devices, so as to comprehensively analyze the user's health status based on multi-dimensional information content. For example, the data integration system can include components such as a wearable device (health monitoring device), a mobile terminal (data processing device), a cloud platform (server), and a data analysis tool. When performing health monitoring, the wearable device can be used to collect the user's physiological data, and components such as the mobile terminal, the cloud platform, and the data analysis tool are used to analyze the physiological data to obtain monitoring result data, and provide health management suggestions and support based on the monitoring result data.

[0038] In some embodiments, the cloud platform in the data integration system can be a distributed cloud server. A distributed cloud server is a computing resource based on a distributed cloud architecture, which can deploy cloud computing services to multiple different physical locations, and at the same time, the cloud service provider centrally manages the operation, governance, update, and evolution of these services. Through the distributed cloud server, the data integration system can be based on a distributed architecture, distribute computing, storage, and network resources in multiple physical locations, and connect through a network. And it is centrally managed by a single control platform to ensure a consistent user experience. It can also combine the powerful computing power of the central cloud and the low-latency characteristics of the edge cloud to achieve cloud-edge collaboration, with efficient resource allocation and task processing. Through multi-region deployment and redundancy mechanisms, high availability and redundancy design are achieved to ensure the high availability of the system.

[0039] However, since the data in the health monitoring process directly comes from the user's physiological data, to a certain extent, these physiological data belong to the user's privacy. And the above health monitoring process directly uses the original data of the physiological data for data processing and transmission, resulting in problems such as low data accuracy, insufficient privacy protection effect, and poor compatibility between different devices in the above health monitoring process.

[0040] To solve the problems of low accuracy, poor security, and poor compatibility in the process of health monitoring data processing, some embodiments of the present application provide a health monitoring data processing method, as Figure 2 shown, this method includes:

[0041] S101. Obtain multi-source health monitoring data.

[0042] To implement the health monitoring function, a data integration system for intelligent health monitoring indicators can acquire multi-source health monitoring data. Among them, the multi-source health monitoring data is the data collected by multiple health monitoring devices within a preset monitoring period. The data integration system can support various types of health monitoring devices for data collection. Therefore, the health monitoring devices include but are not limited to heart rate monitors, blood pressure monitors, blood glucose meters, sleep trackers, etc.

[0043] For this reason, the data integration system can divide into multiple system levels according to the functions it performs. In some embodiments, as Figure 3 shown, the data integration system can include a front-end device layer, a data transmission layer, a cloud platform layer, and a user interaction layer. Among them, the front-end device layer includes various health monitoring devices connected to the data integration system, such as wearable devices like smartwatches and heart rate belts, and home medical devices such as blood pressure monitors and blood glucose meters.

[0044] It should be noted that the health monitoring devices for data collection are not limited to physical devices such as wearable devices and home medical devices, but can also be applications. For example, applications on smart mobile terminals with functions such as step counting and calorie consumption, data interfaces of third-party medical service providers, etc.

[0045] The data transmission layer can use communication technologies such as Bluetooth and Wi-Fi to transmit data from the front-end devices to the cloud server. The cloud platform layer is responsible for data storage, processing, and analysis. Therefore, the cloud platform layer can include a database, computing resources, security services, and a data analysis engine. The user interaction layer can provide an intuitive interface for users through a mobile application or a web page, display the analysis results, and receive user feedback.

[0046] After multiple health monitoring devices are deployed and connected to the data integration system, they can collect the physiological data of users and transmit it to the data integration system after converting it into health monitoring data. Therefore, the multi-source health monitoring data is the data obtained after the initial processing of the original physiological data collected by multiple health monitoring devices. The initial processing of the original data can include analog-to-digital conversion, validity judgment, format conversion, etc.

[0047] To improve the timeliness of health monitoring data, multiple health monitoring devices can set corresponding monitoring periods according to the types of data they monitor. For example, when the health monitoring device is a sleep tracker for monitoring sleep quality, the monitoring period can be set according to the sleep time, such as 8h, etc. Another example is that the monitoring period is the period corresponding to the regular or event-triggered collection of physiological data by the health monitoring device, such as the heart rate per minute, the number of steps per day, etc.

[0048] Multiple health monitoring devices can collect users' physiological data in multiple dimensions, thus forming multi-source health monitoring data. The data types monitored by multiple health monitoring devices can be the same or different. For example, when a user wears a smart bracelet with a heart rate monitoring function, they can also use a medical device with a heart rate monitoring function. In some embodiments, when multiple health monitoring devices collect the same type of physiological data, the data integration system can comprehensively analyze the physiological data collected by multiple health monitoring devices to obtain more accurate physiological data.

[0049] Since multiple health monitoring devices have different brands and models, the data forms collected and processed by the corresponding health monitoring devices also vary. Therefore, in some embodiments, the data integration system has an adaptive protocol parsing function, which can dynamically adjust protocol parsing parameters and parsing behaviors according to different network environments, device types, and application requirements, and ensure wide compatibility by automatically identifying and adapting to devices of different brands and models.

[0050] For example, the data integration system can be built with a protocol identification and conversion module, a dynamic adjustment module, an intelligent model support module, and a plugin management module. Among them, the protocol identification and conversion module can identify the communication protocols of different devices through the protocol identification process and convert them into a unified protocol format. The dynamic adjustment module can obtain network states such as the bandwidth, latency, and congestion of the current network and device requirements, and then adjust the protocol parameters in real time according to the network state or device requirements to optimize the data transmission efficiency. The intelligent model support module can use machine learning or deep learning algorithms to automatically learn protocol features and optimization strategies. The plugin management module can manage the protocol parsing framework in the form of plugins. When a protocol needs to be used, the plugin management module can provide the corresponding protocol plugin for easy expansion and update to adapt to the constantly changing protocol standards.

[0051] S102. Perform standardization processing on the multi-source health monitoring data to obtain standard health data.

[0052] After obtaining the multi-source health monitoring data, the data integration system can also process the multi-source health monitoring data so that the processed data can be used for health monitoring. Among them, processing the multi-source health monitoring data can include preprocessing and standardization processing. Therefore, the multi-source health monitoring data after processing is the standard health data.

[0053] In some embodiments, in order to preprocess multi-source health monitoring data, the data integration system may perform noise filtering on the multi-source health monitoring data according to data preprocessing items, and read the values to be processed from the multi-source health monitoring data after noise filtering. Among them, the values to be processed include missing values, outliers, and duplicate values. Then, correction processing is performed on the values to be processed according to preset correction items. Obviously, different correction processing methods can be performed for different values to be processed, that is, the preset correction items include filling in missing values, modifying or removing outliers, and deleting duplicate values.

[0054] After obtaining the multi-source health monitoring data, the data integration system preprocesses the original data, including data cleaning items such as noise filtering, outlier detection and removal, etc., to remove noise and outliers in the original data. For example, for heart rate data, a filter can be used to remove high-frequency noise, and a reasonable range can be set to remove unreasonable values.

[0055] In some embodiments, the data integration system can use digital filters such as low-pass filters to remove high-frequency noise to achieve noise filtering. And statistical methods such as Z-score and Interquartile Range (IQR) are used to identify and remove unreasonable values to achieve outlier detection and processing. A smoothing algorithm such as moving average can also be used to reduce short-term fluctuations to achieve data smoothing processing. For duplicate values in the original data, the redundant duplicate items can be deleted by comparing the duplicate values, and only one of them is retained.

[0056] After preprocessing the multi-source health monitoring data, the data integration system can perform standardization processing on the preprocessed multi-source health data. That is, in some embodiments, the data integration system can parse the data items to be unified from the multi-source health monitoring data, obtain the data types corresponding to the data items to be unified, and determine the standardization algorithm based on the data types. Among them, the standardization algorithm includes at least one of unifying the time frame, unifying the data format, and unifying the variable unit. Then, the data items to be unified are converted into a standard format according to the standardization algorithm to obtain standard health data.

[0057] For different data types, the data integration system can apply different standardization algorithms. For example, for heart rate data, different health monitoring devices may provide data in different units or precisions. That is, device A uses beats per second as the heart rate unit, and the precision can reach 0.01, that is, two decimal places after the decimal point. And device B uses beats per minute as the heart rate unit, and the precision is 0.1, that is, one decimal place after the decimal point. Then, when standardizing the heart rate data, the heart rate data can be unified in format, that is, all heart rate data should be converted to beats per minute and two decimal places should be retained, such as 72.50 bpm, etc.

[0058] Similarly, for blood pressure data, the raw data format can be expressed in mmHg or Pascal (Pa) through systolic and diastolic pressure, and may contain additional information such as pulse rate. Therefore, through data standardization, the blood pressure data can be standardized into a unified format, that is, standardized into the form of systolic and diastolic pressure, such as 120 / 80 mmHg, while recording the pulse rate as a separate field, such as 70 bpm, etc.

[0059] For blood glucose levels, the raw format of blood glucose values ​​may be expressed in mg / dL or mmol / L. Normalization can be used to unify the format into a standard unit of mg / dL and ensure that all blood glucose readings are reported in this unit, such as 100 mg / dL.

[0060] For weight and body mass index (BMI), the original format of weight may be expressed in pounds (lbs) or kilograms (kg); height may be expressed in inches (in) or centimeters (cm). After the unified format, weight is unified in kilograms and height is in centimeters, and BMI calculation follows BMI = weight / (height) 2 , keep one decimal place, such as 25.2kg / m 2 .

[0061] For sports activity data, the original format is that the step counter directly gives the number of steps, and the treadmill or other fitness equipment may provide information such as distance (kilometers / km or miles / mi), speed (kilometers per hour / km / h or miles per hour / mph). After the unified format, the number of steps remains in integer form, the distance is unified in kilometers, and the speed is in kilometers per hour, such as 10000steps, 5km, 8km / h, etc.

[0062] For sleep quality data, the original format may include information such as the time of falling asleep, the number of times you wake up, and the length of deep sleep, and each device has a different way of expressing it. After unifying the format, a set of common sleep stage classifications (such as awake, light sleep, deep sleep, REM) can be defined, and the duration of each stage can be recorded in the form of "hours: minutes", such as Deep Sleep: lh45min.

[0063] For dietary intake data, in the original format, nutritional components may be expressed in grams (g), milligrams (mg), micrograms (μg), etc., and may also include calories (cal). After unifying the format, for macronutrients (protein, fat, carbohydrates), grams are used as the unit; for micronutrients, appropriate units are selected according to specific circumstances (e.g., vitamin D is expressed in international units IU), and total calories are expressed in kilocalories (kcal), such as Protein: 75g, Vitamin D: 10IU, Total Calories: 2000kcal.

[0064] During the implementation of the standardization process, the data integration system can implement a time synchronization mechanism, that is, to ensure that data from different sources can be compared within a unified time frame. For example, all timestamps of multi-source health monitoring data are converted to Coordinated Universal Time (UTC) time to ensure the consistency of data on the time axis. And, based on the clock synchronization protocol, the Network Time Protocol (NTP) or a similar protocol can be used to synchronize the time of the front-end device and the server regularly.

[0065] It can be seen that in the above embodiments, by preprocessing and standardizing the multi-source health monitoring data, the consistency and comparability of all input multi-source health monitoring data can be ensured, which can not only achieve the unified analysis and processing of data, but also improve device compatibility.

[0066] S103. Store the standard health data in a distributed storage medium.

[0067] After preprocessing and standardizing the health monitoring data to obtain the standard health data, the data integration system can also store the standard health data, that is, store the standard health data in a distributed storage medium. To improve the security during data storage, independent data partitions can be set in the distributed storage medium according to user information, and data access permissions based on multi-factor authentication are set for the data partitions.

[0068] To achieve cloud storage and management of health monitoring data, the data integration system can use a highly available cloud server to store users' health data and adopt a distributed file system to improve data access speed. During data storage, a distributed database can be constructed based on the principle of distributed data storage, and a large amount of user data can be stored in the storage medium based on the distributed database. During the storage process, each user has an independent data partition to ensure data isolation and privacy protection.

[0069] In some embodiments, the data integration system may apply a specific encryption standard to encrypt sensitive information, ensuring security during data transmission and storage. That is, as Figure 4 shown, when storing the standard health data in the distributed storage medium, sensitive information in the multi-source health monitoring data can be marked first, then the data type corresponding to the sensitive information can be obtained, and the encryption algorithm can be determined according to the data type. The encryption complexity of the encryption algorithm is positively correlated with the security requirements corresponding to the data type. That is, the higher the security requirements of the data, the higher the encryption complexity of the corresponding encryption algorithm used. Then, based on the encryption algorithm, the health monitoring data and standard health data corresponding to the sensitive information are encrypted to generate encrypted data, and based on the distributed file storage standard, the encrypted data is stored in the distributed storage medium.

[0070] For example, after the data integration system generates the standard health data, sensitive information such as names and ID numbers in the standard health data can be marked to determine that the information is sensitive. Then, the data type corresponding to the sensitive information is obtained to determine the encryption algorithm. For sensitive information such as names and ID numbers, the AES-256-bit encryption algorithm can be used to encrypt the sensitive information to generate encrypted data to ensure data security. In addition, the data integration system can also perform key management, that is, implement a secure key generation, distribution, and update mechanism to ensure the security of the keys.

[0071] After generating the encrypted data, the data integration system can also store the encrypted data in the distributed storage medium based on the distributed file storage standard. In some embodiments, to store the encrypted data, the data integration system can extract user information from the multi-source health monitoring data and delimit data partitions in the distributed storage medium according to the user information. The data partitions include a storage space and a backup space with the same user label. The user label is a label value obtained by performing a hash process on the user information using a hash function. The storage space and the backup space belong to different storage node devices respectively.

[0072] The storage space and the backup space, as independent data partitions, can be based on sharding by user, that is, the data is divided into different database instances or nodes according to the user ID or other unique identifiers. All relevant data of each user is stored in the same shard. Then, by selecting an appropriate hash function to perform a hash process on the user ID, it is determined which shard the data should be placed in. For example, using the consistent hashing algorithm can effectively reduce the amount of data that needs to be reallocated when adding or removing nodes. Logically implemented at the application layer, the data integration system can calculate the correct shard location according to the user ID and direct read and write requests to the corresponding database instance.

[0073] Since selecting a good partitioning key is crucial for performance and scalability, and in the embodiments of the present application, the user ID can ensure that data of the same user is always located on the same shard. Therefore, when selecting a partitioning key, the user ID can be the most direct choice. The selection of the partitioning key can ensure that the selected partitioning key has good distribution characteristics and avoid hot spot problems, that is, some shards bear more load than other shards. If it is found that the data access frequency of certain specific users is much higher than that of other users, it may be necessary to further optimize the partitioning strategy.

[0074] The storage space and backup space can be physically and logically isolated. Among them, physical isolation provides the highest level of isolation and improves data security by providing completely independent physical storage spaces for each user. Logical isolation, on the other hand, creates logically independent partitions on shared physical resources through software-level design, making the data storage process more flexible and efficient.

[0075] Corresponding access control and permission management policies can also be set for the defined data partitions. That is, the data integration system can achieve fine-grained permission control. Even within the same database instance, security can be enhanced by setting access permissions at the table, row, or even column level. And based on encrypted storage, on the basis of logical isolation, sensitive fields are encrypted and stored, and only authorized application programs can decrypt and read this information.

[0076] The data integration system can also set backup and recovery policies for the defined data partitions, that is, by regularly performing full and incremental backups and formulating a detailed disaster recovery plan to ensure that the data of affected users can be quickly restored in case of a failure, and fully considering cross-regional replication solutions to improve data availability and disaster tolerance.

[0077] In addition, the data integration system can also monitor and audit the defined data partitions. That is, by implementing a comprehensive logging and monitoring mechanism to track all operations on user data in order to timely detect potential security threats. By regularly reviewing system logs, check whether the database has abnormal behaviors or unauthorized access attempts.

[0078] For example, for the scenario of health monitoring data processing, a health monitoring platform supporting millions of users can be built. To this end, a distributed database such as Cassandra or MongoDB can be used for the storage database to take advantage of the natural support for horizontal scaling and sharding functions of the distributed database to achieve distributed storage of health monitoring data. When designing the user table, include the primary key user_id and use this field as the partition key. The application layer code of the data integration system can be responsible for calculating the location of the target shard based on the incoming user_id and performing corresponding Create, Read, Update, Delete (CRUD) operations. For particularly sensitive information such as medical records, encrypt it before writing to the database. And configure appropriate indexing strategies to ensure high query efficiency without affecting data isolation.

[0079] After partitioning the data in the distributed storage medium according to the user information, the data integration system can calculate the hash value of the encrypted data corresponding to the user information and store the encrypted data in the storage space and the backup space with the same tag value as the hash value respectively.

[0080] It can be seen that in the above embodiment, the data integration system introduces a redundant backup strategy by delimiting the storage space and the backup space, and the storage space and the backup space belong to different node devices respectively. When the health monitoring data in the storage space is lost due to a hardware failure, the data integration system can also obtain and maintain the health monitoring data from the backup space, further improving the data security.

[0081] S104. Invoke the health status analysis model.

[0082] After storing the standard health data in the data partition of the distributed storage medium, the data integration system can invoke the health status analysis model to perform analysis and processing of the health monitoring data. Among them, the health status analysis model is a machine learning model trained based on sample health data; the sample health data is labeled with a health pattern label and a potential risk label. Then, based on the health status analysis model trained with the sample health data, machine learning algorithms can be used to perform pattern recognition on the user's health monitoring data and historical data to discover potential health trends or risk factors.

[0083] In some embodiments, before invoking the health status analysis model, the data integration system can also perform model training. That is, the data integration system can first invoke a trained machine learning model and obtain sample health data. Among them, the trained machine learning model can be a machine learning model such as a random forest or a support vector machine. The sample health data is training data obtained by annotating multi-source health monitoring data with labels. For example, the sample health data can include heart rate data, blood pressure data, blood glucose data, and sleep data, etc. The health trends and potential risks under different heart rates, blood pressures, blood glucose levels, and sleep conditions can be annotated based on manual annotation or machine automatic annotation and manual verification.

[0084] Then, the sample health data is input into the trained machine learning model, and the health analysis result output by the trained machine learning model is obtained. Among them, the health analysis result is the classification probability of the sample health data for health trends and potential risks. Then, the health analysis result is compared with the annotated label, so as to calculate the error value between the health analysis result and the annotated label. This error value can be obtained based on the loss function set by the machine learning model.

[0085] After obtaining the error value, the error value is judged based on a preset classification error threshold. If the error value is greater than the classification error threshold, it means that the current machine learning model has not been trained to convergence. Therefore, the error value can be used for backpropagation, that is, the model parameters of the machine learning model are adjusted, and iterative training continues.

[0086] If the error value is less than or equal to the classification error threshold, or the number of iterations of the machine learning model reaches the set iteration number threshold, it means that the current machine learning model has been trained to convergence. Therefore, the model parameters of the machine learning model at this time can be output to obtain the health status analysis model.

[0087] S105. Input the standard health data into the health status analysis model to obtain the monitoring result data output by the health status analysis model.

[0088] After invoking the health status analysis model, the data integration system can input the standard health data into the health status analysis model to analyze and process the standard health data through the health status analysis model, so as to obtain the classification probability of the standard health monitoring data for health patterns and potential risks. The health pattern and potential risk with the highest classification probability are used as the output data, that is, the monitoring result data. Obviously, the monitoring result data includes health pattern information and potential risk information.

[0089] When the health status analysis model performs feature extraction, it can extract useful features from the original data, such as heart rate changes, activity intensity, etc. And use supervised learning or unsupervised learning algorithms to train the model, such as random forest, clustering algorithm, etc., so that the health status analysis model can identify health patterns and risk factors. It can be seen that the data integration system can use machine learning algorithms to perform pattern recognition on the user's historical data, discover potential health trends or risk factors, and thus enhance the data analysis ability.

[0090] In some embodiments, the health status analysis model can also use time series analysis algorithms such as Autoregressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), etc. to predict future health trends. That is, a module for regression analysis and data prediction can also be provided in the health status analysis model to perform regression analysis based on standard health monitoring data and historical health monitoring data, and use time series analysis to predict future trends, such as heart rate change trends, changes in sleep quality, etc.

[0091] In some embodiments, the data integration system can also provide personalized health suggestions based on the standard health monitoring data and historical health monitoring data. For example, customize a personalized health improvement plan according to factors such as the user's age, gender, and lifestyle, and provide targeted health guidance.

[0092] To achieve personalized health suggestions, the data integration system can also build a user profile, that is, build a user profile including information such as age, gender, and lifestyle. And based on the user profile and health data, use a recommendation algorithm to generate a personalized health improvement plan. It can also dynamically adjust health suggestions according to the user's feedback and behavior changes.

[0093] The monitoring result data output by the health status analysis model can also be used to form a visualization report, that is, generate an easy-to-understand and personalized health report to help users better understand their own status and take corresponding actions. For this purpose, the data integration system can also automatically generate a health report including key indicators, trend analysis, and personal suggestions. And allow users to select the indicators and time periods of concern to generate a customized report. The data integration system can also provide functions for report export and sharing, facilitating users to share health information with doctors or family members.

[0094] To present a visual report, the data integration system can also build a simple and intuitive user interface. In terms of UI design, a modern and simple design style can be adopted to ensure that the user interface is easy to understand and operate. And visual charts such as bar charts, line charts, and pie charts can be used to display key health indicators to help users quickly understand the data.

[0095] S106. Store the monitoring result data through the data partitioning.

[0096] After obtaining the monitoring result data output by the health status analysis model, the data integration system can also store the output monitoring result data. Similar to the standard health monitoring data in the above embodiments, the detection result data can also be based on independent data partitions and complete data storage in the storage space and backup space respectively. For example, a distributed database is used to store the monitoring result data, and each user has an independent data partition to ensure data isolation and privacy protection. And the AES-256 bit encryption algorithm is used to encrypt the sensitive information in the monitoring result data to ensure the security of the monitoring result data, which will not be elaborated here.

[0097] In some embodiments, the distributed database of the data integration system can adopt an open interface design, that is, provide a standardized data interface to support the data access of third-party devices and service providers and promote cross-platform data sharing. For example, for the application programming interface (API) design of the data integration system, standard interfaces such as RESTful API can be provided for third-party devices and service providers to access. A software development kit (SDK) can also be provided to facilitate developers to integrate new devices or services. And the data integration system can support plug-in extension, allowing new device drivers or data processors to be added.

[0098] To improve device compatibility, the data integration system can support cross-platform access, that is, support mobile applications and web pages, so that users can obtain a consistent experience regardless of which platform they are on. In this regard, the data integration system can achieve multi-terminal support, that is, by developing iOS, Android applications and web pages, users can access their health data on any device.

[0099] In some embodiments, the data integration system can be based on responsive design to ensure that the display effect of the application is consistent on different screen sizes and devices. Therefore, as Figure 5As shown, when storing the monitoring result data through the data partition, the data integration system can also obtain the data to be displayed, which is the data extracted from the monitoring result data and / or the multi-source health monitoring data. Then, metadata is generated based on the data to be displayed, and display parameters are set for the metadata. The display parameters include percentage layout parameters, maximum display size, custom style, and display mode label. Then, the metadata is stored in the data partition.

[0100] The goal of responsive design is to ensure that websites or applications can provide a consistent user experience on various devices such as desktop computers, tablets, and smartphones. To this end, technologies such as flexible grid layouts, flexible images, and media queries can be used to adapt to different screen sizes. The data integration system can utilize a fluid grid layout, that is, a percentage-based layout. Instead of using fixed pixel values to define widths, percentages are used to set the widths of elements, enabling the page layout to be automatically adjusted according to the screen size. The data integration system can utilize Cascading Style Sheets (CSS) frameworks such as Bootstrap or Foundation, which provide predefined classes and components, to quickly build responsive layouts.

[0101] The data integration system can also set a flexible view display strategy, such as Flexible Images and Media. Set the maximum display size in the properties of the CSS framework, such as the maximum display width max-width. By setting max-width: 100% for the metadata of all image types, images will be automatically scaled according to the display space but will not exceed their original size. When setting the flexible view display strategy, relative units such as vw and vh can also be used to define the size of media elements, enabling them to change according to the viewport size.

[0102] The data integration system can also define display styles based on the principle of Media Queries, that is, use the @media rule in the CSS framework to apply specific styles for different screen sizes. By detecting device characteristics such as width, height, and orientation, style sheets can be customized for different devices. Then, display style notification is implemented based on the viewport meta tag. For example, add a (meta) tag to the head of a HyperText Markup Language (HTML) document to inform the browser how to control the scale and size of the page.

[0103] After setting the display parameters, the data integration system can also perform testing and optimization based on the set display parameters. That is, through cross-browser testing, ensure that it can work properly on mainstream browsers and different versions of browsers. When performing testing and optimization, the device mode in the developer tools can be used to check the performance under different screen sizes. And consider the loading time, optimize the image size, reduce HyperText Transfer Protocol (HTTP) requests, and use a Content Delivery Network (CDN) to accelerate resource loading, etc.

[0104] In some embodiments, the data integration system can also build an instant feedback mechanism to implement real-time notification and recommendation engine functions. When an abnormal situation is detected, the user is reminded in a timely manner through push notifications. And based on the user's current status and historical data, provide instant health advice and action guidelines.

[0105] Furthermore, as a refinement and extension of the specific implementation manner of the above embodiments, in order to fully illustrate the specific implementation process of this embodiment, some embodiments of this application also provide a method for processing health monitoring data, as Figure 6 shown, this method includes:

[0106] S201. Obtain a data access request.

[0107] To support data access, the data integration system can receive a data access request sent by an electronic device, that is, the data access request is a request sent by an electronically device that has passed multi-factor authentication. That is, in some embodiments, the data integration system can ensure that only authorized users can access their personal data based on a multi-factor authentication scheme. Among them, the multi-factor authentication includes various combinations of knowledge factor authentication, possession factor authentication, and biometric factor authentication.

[0108] The knowledge factor (Something You Know) is to authenticate the identity through factors such as passwords and Personal Identification Numbers (PINs). As the first layer of protection, users need to create a strong password and update it regularly to ensure security. In some cases, a 4-6 digit personal identification number, that is, a PIN, can be used as an alternative or supplement to the password.

[0109] The "Something You Have" factor implements identity authentication through one-time verification codes, token information, message authentication codes, etc. Among them, the one-time verification code is a one-time password (OTP) generated by electronic devices such as mobile terminals. By installing authentication-related applications, such as ×× Authenticator, on the mobile terminal, a one-time verification code is generated. This dynamically generated code needs to be entered each time when logging in. Token information is achieved through physical hardware token devices, such as USB-form security keys like YubiKey, to complete the verification process after connecting to the electronic device. The message authentication code is a text message or email verification code, that is, the authentication system can send a temporary verification code to the user's registered mobile phone number or email address, and the user needs to enter this verification code within the specified time to complete the verification.

[0110] The "Something You Are" factor, i.e., biometric factor, is the process of implementing identity authentication through the recognition of fingerprints, faces, irises, etc. Among them, fingerprint recognition uses the built-in fingerprint sensor of the device for identity verification, which is applicable to smartphones and tablets that support this function. Facial recognition captures the user's face image through the front camera and compares it to confirm the identity, which is suitable for devices with high-resolution cameras. Iris scanning is applicable to some electronic devices that can provide iris recognition function to provide higher security.

[0111] When the data integration system performs multi-factor authentication, it can first perform identity registration. That is, when the user first registers an account, in addition to setting a username and password, at least one additional authentication method needs to be selected and configured. For example, download and configure a one-time password (OTP) generator application, or bind a mobile phone number to receive verification codes. When the user attempts to log in, the username and password are first entered. Then the authentication system can prompt the user to complete the second-step verification. This may include opening the OTP application on the mobile phone to obtain the current valid code, or waiting to receive the verification code in the text message / email and entering it into the login interface. If biometric authentication is enabled, according to the selected method (such as fingerprint, facial recognition, etc.), the user needs to complete the last-step verification through the corresponding biometric scan.

[0112] When the data integration system performs multi-factor authentication, it can also detect abnormal behaviors. That is, the system can monitor login activities. Once abnormal behaviors such as logging in from an infrequently used device or geographical location are detected, additional verification steps are automatically triggered, and the account may even be temporarily locked until the identity is further verified.

[0113] Obviously, the data integration system can also provide a recovery mechanism, that is, provide an alternative method to enable users to reset their account access rights even when the primary authentication tool is lost. For example, when a user changes their mobile phone and cannot access the original OTP application, identity recovery can be completed using pre-set security questions, alternative email addresses, or phone numbers, etc.

[0114] S202. In response to the data access request, extract the target data from the distributed storage medium.

[0115] After obtaining the data access request, the data integration system can respond to the data access request, that is, the data integration system can extract the data information to be accessed from the data access request and extract the target data from the distributed storage medium according to the data information to be accessed, where the target data includes the standard health data and / or the monitoring result data.

[0116] According to the different purposes of data access, the data integration system can respectively obtain data from the standard health data and / or the monitoring result data. For example, when the data access request indicates accessing original physiological data such as blood pressure data and heart rate data, the data integration system can extract blood pressure data and heart rate data from the standard health data in the distributed storage medium. When the data access request indicates accessing monitoring result data such as health patterns and potential risks, the data integration system can extract data related to health patterns and potential risks from the monitoring result data in the distributed storage medium.

[0117] S203. Invoke the anonymization processing algorithm according to the field type corresponding to the sensitive field in the target data.

[0118] To improve the security of data during transmission, after extracting the target data from the distributed storage medium, the data integration system can read the sensitive fields of the target data and the field types corresponding to the sensitive fields to invoke the anonymization processing algorithm according to the field types. The anonymization processing algorithm can be used to anonymize the target data. Through anonymization processing, information that can directly or indirectly identify an individual can be removed or hidden, thereby protecting user privacy without affecting the value of data analysis.

[0119] In some embodiments, the anonymization processing algorithm includes one or more combinations of data desensitization, data generalization, differential privacy, k-anonymity processing, and data aggregation.

[0120] Among them, data masking refers to processing sensitive fields through algorithms such as replacement, obfuscation, and encryption to remove the sensitivity of sensitive fields. When performing replacement, fictional but seemingly real data can be used to replace the sensitive fields in the original data. For example, replace the real name with a randomly generated name. When performing obfuscation, some fields can be partially masked, such as showing the prefix of an email address while hiding the domain name part, or only showing some digits of the ID number. When performing encryption, irreversible encryption algorithms such as hash functions can be applied to sensitive fields to hide their real content while retaining the data format.

[0121] When performing data masking, the data integration system can first determine which fields need to be masked, and then select the appropriate masking method according to the field type, so as to perform the masking operation before the data is exported or shared.

[0122] Generalization refers to converting specific numerical values into broader categories or ranges. This method is particularly applicable to numerical data such as age, income, etc. For this purpose, the data integration system can first define the generalization hierarchy. For example, age can be grouped by intervals such as "0 - 18 years old", "19 - 35 years old", etc. Then for each sensitive field, map its value to the corresponding generalization level according to the predefined rules.

[0123] Differential privacy is a mathematically rigorous privacy protection framework that ensures that the presence or absence of a single record does not significantly affect the overall statistical results by adding noise to the query results. When performing anonymization, the data integration system can first determine the types of statistical data to be published, such as average values, counts, etc. Then calculate the appropriate amount of noise, which can be based on the trade-off between the required privacy protection level and the dataset size, and add the calculated noise before publishing the statistical data.

[0124] k-anonymity processing, that is, K-Anonymity, requires that in any set of attribute combinations that may be used to re-identify an individual, at least K records have the same value. In this way, even if an attacker knows certain specific attributes, they cannot uniquely determine which record belongs to whom. For this purpose, the data integration system can analyze all possible quasi-identifiers in the dataset that can be used to re-identify an individual, such as gender, date of birth, postal code, etc. Then apply generalization or other techniques to make each set of quasi-identifiers meet the requirements of K-anonymity.

[0125] Data aggregation forms an overall trend or pattern by summarizing data points of multiple individuals, avoiding the exposure of specific details of individuals. To this end, a data integration system can collect a sufficient number of relevant individual data. Aggregate the data according to logical grouping methods such as by region, age group, etc. Publish the summarized statistical results instead of the original data.

[0126] S204. Perform anonymization processing on the sensitive fields using the anonymization processing algorithm to convert the target data into data to be transmitted.

[0127] After determining the anonymization processing algorithm, the data integration system can use the determined anonymization processing algorithm to perform anonymization processing on the sensitive fields to convert the target data into data to be transmitted. For example, when a data integration system wants to share a batch of users' health data for medical research while ensuring that user privacy is not violated. The data integration system can first completely delete or replace direct identifiers such as names and ID numbers. For potential quasi-identifiers such as birthdays, genders, and postal codes, the K-anonymity principle can be applied to ensure that each group of quasi-identifier combinations corresponds to at least K users. When it comes to specific numerical values such as blood pressure readings and weights, generalization or differential privacy techniques can be used so that the exact data of a single user cannot be deduced. And only publish the data summary report after the above processing, rather than the original records.

[0128] S205. Based on the transmission encryption algorithm corresponding to the data transmission protocol, perform encryption processing on the data to be transmitted, and send the encrypted data to be transmitted to the electronic device.

[0129] After converting the target data into data to be transmitted, the data integration system can determine the data transmission protocol between the electronic device and the system, such as the Secure Sockets Layer (SSL) / Transport Layer Security (TLS) security protocol, etc., and then based on the transmission encryption algorithm corresponding to the data transmission protocol, perform encryption processing on the data to be transmitted. For example, perform SSL / TLS encryption on the transmission process of the data to be transmitted to prevent man-in-the-middle attacks.

[0130] By applying the technical solutions of the above embodiments, the health monitoring data processing method provided in the above embodiments can improve the consistency and comparability of data through standardized processing, set up independent data partitions based on user information and a data access method based on multi-factor authentication, and enhance the privacy protection function. It can also set display parameters when storing the monitoring result data to improve compatibility between different devices, thereby solving the problems of insufficient data accuracy, poor privacy protection, and poor compatibility between devices in the health management system.

[0131] Further, as a specific implementation of the health monitoring data processing method in the above embodiments, some embodiments of the present application further provide a health monitoring data processing system, as Figure 7 shown. The system includes:

[0132] A data acquisition module, configured to acquire multi-source health monitoring data, where the multi-source health monitoring data is data collected by multiple health monitoring devices within a preset monitoring period;

[0133] A standardization module, configured to perform standardization processing on the multi-source health monitoring data to obtain standard health data;

[0134] A data storage module, configured to store the standard health data in a distributed storage medium, where the distributed storage medium sets independent data partitions according to user information; the data partitions are set with data access permissions based on multi-factor authentication;

[0135] A model invocation module, configured to invoke a health status analysis model, where the health status analysis model is a machine learning model trained based on sample health data; the sample health data is labeled with a health pattern label and a potential risk label;

[0136] An analysis module, configured to input the standard health data into the health status analysis model to obtain monitoring result data output by the health status analysis model, where the monitoring result data includes health pattern information and potential risk information;

[0137] The data storage module is further configured to store the monitoring result data through the data partitions.

[0138] By applying the technical solutions of the above embodiments, the above embodiments provide a health monitoring data processing system. After acquiring multi-source health monitoring data through multiple health monitoring devices, the system performs standardization processing on the multi-source health monitoring data to obtain standard health data, and stores the standard health data in a distributed storage medium. By invoking the health status analysis model and using the standard health data as input, monitoring result data is obtained, and then the monitoring result data is stored through data partitions. The system can improve the consistency and comparability of data through standardization processing, and set independent data partitions and data access methods based on multi-factor authentication according to user information, enhancing the privacy protection function. It can also set display parameters when storing the monitoring result data to improve the compatibility between different devices, thereby solving the problems of insufficient data accuracy, poor privacy protection, and poor compatibility between devices in the health management system.

[0139] It should be noted that the health monitoring data processing system provided in the embodiments of the present application and the data integration system described in the above embodiments may be the same system or different systems. That is, in some embodiments, the health monitoring data processing system is a subsystem of the data integration system. On the basis of including the health monitoring data processing system, the data integration system may further include an electronic device serving as a user terminal, etc. Moreover, for other corresponding descriptions of the various functional units involved in the health monitoring data processing system provided in the embodiments of the present application, reference may be made to the corresponding descriptions in the health monitoring data processing method provided in the above embodiments, which will not be elaborated here.

[0140] As Figure 8 shown, the embodiments of the present application further provide a computer device, which may specifically be a personal computer, a server, a network device, etc. The computer device includes a bus, a processor, a memory, and a communication interface, and may further include an input / output interface and a display device. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store location information. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps in the method embodiments are implemented.

[0141] Those skilled in the art can understand that the structure of the above computer device is only a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components, or combine certain components, or have different component arrangements.

[0142] In one embodiment, a computer-readable storage medium is further provided. The computer-readable storage medium may be non-volatile or volatile, and stores a computer program. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0143] In one embodiment, a computer program product is further provided, including a computer program. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0144] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties.

[0145] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods.

[0146] Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random access memories (ReRAMs), magnetoresistive random access memories (MRAMs), ferroelectric random access memories (FRAMs), phase change memories (PCMs), graphene memories, etc.

[0147] Volatile memories can include random access memory (RAM) or external cache memories, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0148] The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, graphics processors, digital signal processors, programmable logics, data processing logics based on quantum computing, etc., without limitation.

[0149] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0150] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A health monitoring data processing method, characterized in that: The method comprises: Acquire multi-source health monitoring data, where the multi-source health monitoring data is data collected by multiple health monitoring devices within a preset monitoring period; Performing standardization processing on the multi-source health monitoring data to obtain standard health data; The standard health data is stored in a distributed storage medium, wherein the distributed storage medium is provided with independent data partitions according to user information; the data partitions are provided with data access rights based on multi-factor authentication; Calling a health status analysis model, wherein the health status analysis model is a machine learning model trained based on sample health data; the sample health data carries a health pattern label and a potential risk label; Inputting the standard health data into the health status analysis model to obtain monitoring result data output by the health status analysis model, wherein the monitoring result data includes health pattern information and potential risk information; The monitoring result data is stored through the data partitions.

2. The method according to claim 1, characterized in that Performing standardization processing on the multi-source health monitoring data to obtain standard health data, including: performing noise filtering on the multi-source health monitoring data; Reading a value to be processed from the multi-source health monitoring data after noise filtering, wherein the value to be processed includes missing values, abnormal values, and repeated values; Correction processing is performed on the to-be-processed values ​​according to preset correction items, wherein the preset correction items include filling in missing values, modifying or eliminating abnormal values, and deleting duplicate values.

3. The method according to claim 2, characterized in that Performing standardization processing on the multi-source health monitoring data to obtain standard health data, including: Parsing the data items to be unified from the multi-source health monitoring data; Acquire the data type corresponding to the data item to be unified, and determine a standardization algorithm based on the data type, wherein the standardization algorithm includes at least one of a unified time frame, a unified data format, and a unified variable unit; The data items to be unified are converted into a standard format according to the standardization algorithm to obtain standard health data.

4. The method according to claim 1, characterized in that: Storing the standard health data in a distributed storage medium includes: marking sensitive information in the multi-source health monitoring data; Acquire a data type corresponding to the sensitive information, and determine an encryption algorithm according to the data type, wherein the encryption complexity of the encryption algorithm is positively correlated with the security requirement corresponding to the data type; Encrypting the health monitoring data and the standard health data corresponding to the sensitive information based on the encryption algorithm to generate encrypted data; Based on a distributed file storage standard, the encrypted data is stored in the distributed storage medium.

5. The method according to claim 4, characterized in that Based on a distributed file storage standard, storing the encrypted data in the distributed storage medium includes: extracting user information from the multi-source health monitoring data; Delimiting data partitions in a distributed storage medium according to the user information, the data partitions comprising a storage space and a backup space with the same user tag; the user tag is a tag value obtained by performing a hash process on the user information using a hash function; the storage space and the backup space belong to different storage node devices respectively; Calculate a hash value of the user information corresponding to the encrypted data; According to the hash value, the encrypted data is stored in the storage space and the backup space respectively, which are provided with a label value identical to the hash value.

6. The method according to claim 1, characterized in that The method further comprises: Obtaining a data access request, where the data access request is a request sent by an electronic device through multi-factor authentication, where the multi-factor authentication includes a combination of knowledge factor authentication, possession factor authentication, and biometric factor authentication; In response to the data access request, extracting target data from the distributed storage medium, the target data including the standard health data and / or the monitoring result data; According to the field type corresponding to the sensitive field in the target data, an anonymization processing algorithm is called, wherein the anonymization processing algorithm includes one or more combinations of data desensitization, data generalization, differential privacy, k-anonymity processing, and data aggregation; Performing anonymization processing on the sensitive field using the anonymization processing algorithm to convert the target data into data to be transmitted; Based on the transmission encryption algorithm corresponding to the data transmission protocol, the data to be transmitted is encrypted, and the encrypted data to be transmitted is sent to the electronic device.

7. The method according to claim 1, characterized in that Storing the monitoring result data through the data partitions includes: Acquire data to be displayed, where the data to be displayed is data extracted from the monitoring result data and / or the multi-source health monitoring data; generating metadata according to the data to be displayed; Setting display parameters for the metadata, the display parameters including percentage layout parameters, maximum display size, custom style, and display mode label; The metadata is stored in the data partition.

8. A health monitoring data processing system, characterized in that: The system comprises: A data acquisition module, used to acquire multi-source health monitoring data, where the multi-source health monitoring data is data collected by multiple health monitoring devices within a preset monitoring period; A standardization module, used for performing standardization processing on the multi-source health monitoring data to obtain standard health data; A data storage module, used for storing the standard health data in a distributed storage medium, wherein the distributed storage medium is provided with independent data partitions according to user information; the data partitions are provided with data access rights based on multi-factor authentication; A model calling module is used to call a health status analysis model, wherein the health status analysis model is a machine learning model obtained by training based on sample health data; the sample health data carries a health mode label and a potential risk label; An analysis module, used for inputting the standard health data into the health status analysis model to obtain monitoring result data output by the health status analysis model, wherein the monitoring result data includes health mode information and potential risk information; The data storage module is also used to store the monitoring result data through the data partitions.

9. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images