Health condition monitoring method and system based on vital sign data

Abstract

The invention discloses a health condition monitoring method and system based on vital sign data, and belongs to the technical field of computers. The method comprises the following steps: constructing a preset database comprising a rule base, a feature mode base and a semantic knowledge graph; constructing and training to obtain a target health monitoring model; after the identity of the user is confirmed through fingerprint identification, current positioning data and real-time vital sign data of the user are synchronously collected; acquiring illness conditions and historical records of the user; generating a report by adopting two paths according to the selection of the user: based on the current positioning, the real-time vital signs and the illness condition, generating a brief health condition evaluation report through preset database matching; or all data is input into the target health monitoring model, a detailed report containing health assessment, life suggestions and matters needing attention is generated, the accurate data source is ensured through identity verification, and the accuracy and personalized service level of health condition monitoring are effectively improved in combination with multiple data and two generation modes.

Description

Technical Field

[0001] This application relates to the field of health monitoring technology, and in particular to a method and system for monitoring health status based on vital sign data. Background Technology

[0002] In daily life, more and more people are using health monitoring devices for routine health monitoring. These devices can continuously or on-demand collect various vital sign data from users, such as heart rate, blood oxygen saturation, sleep quality, steps, and calorie consumption, and generate preliminary health monitoring results based on this data. This data and conclusions can help users understand their overall health trends and basic physiological status.

[0003] In actual monitoring, such systems often face interference from various internal and external factors, affecting the accurate collection of raw data. For example, the collected information may not belong to the actual user. If these interferences are not effectively identified or filtered out, they will directly lead to inaccurate data and historical records relied upon for subsequent analysis. Furthermore, the data relied upon in existing technologies is relatively singular, generally based on acquired indicator data for health assessment. Health predictions based on singular and unreliable data often result in biased assessments and recommendations, lacking personalization, and their accuracy and reliability still need further improvement. Summary of the Invention

[0004] The main objective of this application is to propose a health status monitoring method and system based on vital sign data, aiming to improve the accuracy of health status monitoring and the level of personalized services.

[0005] To achieve the above objectives, one aspect of this application proposes a health status monitoring method based on vital sign data, the method comprising: A pre-defined database is constructed, comprising a rule base, a feature pattern base, and a semantic knowledge graph; wherein the pre-defined database is automatically updated based on a pre-trained data integration model. An initial health monitoring model is constructed, and the initial health monitoring model is trained to obtain a target health monitoring model; wherein, the target health monitoring model includes a data input module, a data preprocessing module, a health assessment module, a context fusion and decision-making module, a suggestion generation module, and a data output module; Obtain the user's fingerprint and determine the user's identity based on the preset fingerprint; Once the identity is confirmed to be that of the target user, the user's current location data and real-time vital signs data are collected simultaneously. Obtain the user's medical history and historical data; wherein, the historical data includes historical vital sign data and historical detailed assessment results; Based on the user's selection, a brief health status assessment report is obtained by matching the current location data, the real-time vital signs data, and the illness status from the preset database; or, the current location data, the real-time vital signs data, the illness status, and the historical data are input into the target health monitoring model to generate a detailed health status assessment report.

[0006] In some embodiments, the construction of a preset database including a rule base, a feature pattern base, and a semantic knowledge graph includes the following steps: A rule base is constructed using a relational database, wherein the rule base is used to store explicit, urgent, or mandatory health rules; A vector database is constructed, and an approximate nearest neighbor index is established in the vector database to obtain a feature pattern library; wherein, the feature pattern library is used to store health patterns; each health pattern includes a feature vector range, a preset health assessment result corresponding to the feature vector range, and a set of preset suggestions; A semantic knowledge graph for storing relationships between medical concepts is constructed using a graph database; A preset database is constructed based on the rule base, the feature pattern base, and the semantic knowledge graph.

[0007] In some embodiments, the data integration model includes a data parsing module, a structured processing module, a conflict handling module, an encapsulation module, and a writing module; the automatic update of the preset database includes the following steps: The data parsing module calls different interpreters to parse the data from third parties and outputs a unified raw text block. The structured processing module identifies key entities from the original text block, extracts relationships from the key entities, and performs semantic role labeling on the text of the key entities based on the extraction results to obtain a set of structured information; wherein, the key entities include vital signs, numerical ranges, medical concepts, lifestyle suggestions, and precautions; The conflict handling module performs conflict detection on the structured information set, processes the detected conflicts, and obtains a candidate healthy mode list; wherein, the conflict handling includes prioritizing authority, prioritizing newly released data, and voting processing; The encapsulation module encapsulates the candidate health pattern list to generate target data that conforms to the database format, and calculates a confidence score for the target data. The writing module writes the target data with the confidence score into the preset database, records the writing time, and marks the replaced old data as invalid, thus completing the automatic update. In some embodiments, constructing the initial health monitoring model includes the following steps: A data input module is constructed to receive all raw input data; wherein, the raw input data includes real-time vital signs data, current location data, disease status and historical data; A data preprocessing module is constructed to perform data cleaning, data standardization, and data normalization on the raw input data; A health assessment module is constructed based on recurrent neural networks or temporal convolutional networks to generate health assessment results from preprocessed data. A context fusion and decision-making module is constructed to identify the contextual relationships of preprocessed data; wherein, the contextual relationships are identified through a knowledge graph of the real-time vital signs data, the current location data, the disease status, and the historical data. A suggestion generation module is built based on a large language model to generate personalized life suggestions and precautions according to the contextual relationships. A data output module is constructed to adapt and output the health assessment results, lifestyle suggestions, and precautions in the specified formats, thereby obtaining the output results. A feedback module is constructed to obtain user feedback on the output results; The data input module, the data preprocessing module, the health assessment module, the context fusion decision module, the data output module, and the feedback module are integrated into an initial health monitoring model.

[0008] In some embodiments, training the initial health monitoring model to obtain the target health monitoring model includes the following steps: Acquire vital sign data and corresponding location data, disease status, historical data, health assessment results, lifestyle suggestions and precautions for training, and construct the original dataset; The original dataset is divided into a training dataset and a test dataset according to a certain ratio; The training dataset is input into the pre-trained initial health monitoring model for parameter learning to obtain the first health monitoring model; The first health monitoring model is optimized using the test dataset based on a combination-based cross-entropy loss function. The target health monitoring model is obtained after the loss value meets the preset conditions.

[0009] In some embodiments, obtaining a user's fingerprint and determining the user's identity based on a preset fingerprint includes the following steps: Obtain the current user's fingerprint; wherein the current user's fingerprint is collected before collecting real-time vital signs data; The current user fingerprint is compared with the user's preset fingerprint to obtain the comparison result; The user's identity is determined based on the comparison results.

[0010] In some embodiments, obtaining a brief health assessment report from the preset database based on the current location data, the real-time vital signs data, and the illness status includes the following steps: The real-time vital signs data are matched with emergency rules in the rule base; If a match is found, all the preset emergency suggestions found are returned directly; if no match is found, based on the illness and the current location data, a set of the most relevant general rules are retrieved from the rule base, and the search range of the feature pattern library is determined based on the general rules. The real-time vital signs data are preprocessed into real-time feature vectors, and the approximate nearest neighbor method of the vector database is used to retrieve candidate results within the search range. Based on the knowledge graph of semantics, combined with the disease status and the current location data, the content of the candidate results is optimized to obtain a brief health status monitoring report.

[0011] In some embodiments, inputting the current location data, the real-time vital signs data, the illness status, and the historical data into the target health monitoring model to generate a detailed health status assessment report includes the following steps: The system receives the current location data, real-time vital signs data, illness status, and historical data through the data input module. The data preprocessing module performs data cleaning, standardization, and normalization on the real-time vital signs data and the disease status to obtain the first data; the current location data and the historical data are standardized to obtain the second data. The health assessment module extracts features from the first data, identifies long-term dependencies and complex temporal patterns between features, and then identifies the health status to obtain the health assessment result; wherein, the health status includes normal, sub-healthy, warning, and abnormal. The context fusion decision module fuses the health assessment results, the current location data, and the historical data to obtain the contextual relationship; Based on the contextual relationships, the suggestion generation module automatically generates personalized life suggestions and precautions in natural language. The data output module adapts and outputs the health assessment results, lifestyle suggestions, and precautions in the correct format to obtain a detailed health status assessment report.

[0012] To achieve the above objectives, another aspect of this application proposes a health status monitoring system based on vital sign data, the system comprising: The first module is used to construct a pre-set database including a rule base, a feature pattern base, and a semantic knowledge graph; wherein, the pre-set database is automatically updated based on a pre-trained data integration model; The second module is used to construct an initial health monitoring model and train the initial health monitoring model to obtain a target health monitoring model; wherein, the target health monitoring model includes a data input module, a data preprocessing module, a health assessment module, a context fusion and decision-making module, a suggestion generation module, and a data output module; The third module is used to acquire the user's fingerprint and determine the user's identity based on the preset fingerprint. The fourth module is used to collect the user's current location data and real-time vital signs data after confirming that the identity is that of the target user. The fifth module is used to obtain the user's medical history and historical data; wherein, the historical data includes historical vital sign data and historical detailed assessment results; The sixth module is used to obtain a brief health status assessment report from the preset database based on the user's selection, the current location data, the real-time vital signs data, and the illness status; or, to input the current location data, the real-time vital signs data, the illness status, and the historical data into the target health monitoring model to generate a detailed health status assessment report.

[0013] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0014] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described above.

[0015] The embodiments of this application include at least the following beneficial effects: This application provides a health status monitoring method and system based on vital sign data. This solution ensures the accuracy of data sources through identity verification and effectively improves the accuracy of health status monitoring and the level of personalized services by combining multiple data sources and two generation modes. Specifically, it combines user location data, disease status, and historical data to generate health assessment results, lifestyle suggestions, and precautions through preset database matching or target health monitoring model generation. Report generation based on more data facilitates providing personalized lifestyle suggestions and precautions while monitoring user health, thus improving user experience. This solution also identifies the user by collecting their fingerprint and matching it with preset fingerprints. Vital sign data is only collected synchronously when the user's identity is confirmed. This ensures that the collected vital sign data originates from the user, guaranteeing the accuracy of the data source and preventing the accidental collection of data from others, which could contaminate historical data and interfere with subsequent health assessments and suggestion generation, thereby improving the accuracy of health status monitoring and report generation to a certain extent. Attached Figure Description

[0016] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0017] Figure 1 This is a flowchart of a health status monitoring method based on vital sign data provided in an embodiment of this application; Figure 2 This is a flowchart illustrating the steps of a health status monitoring method based on vital sign data provided in an embodiment of this application. Figure 3 This is a schematic diagram of a health status monitoring system based on vital sign data provided in an embodiment of this application; Figure 4 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0019] Although functional modules are divided in the system diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the system or the order in the flowchart. The terms "first / S100," "second / S200," etc., in the specification, claims, and the aforementioned figures are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0020] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0021] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0022] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0024] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0025] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0026] In related technologies, monitoring systems often face interference from various internal and external factors, affecting the accurate collection of raw data. For example, the collected information may not belong to the actual user. If these interferences are not effectively identified or filtered out, they will directly lead to inaccurate data and historical records relied upon for subsequent analysis. Furthermore, existing technologies rely on relatively singular data, generally performing health assessments based on acquired indicator data. Health predictions based on singular and unreliable data often result in biased assessments and recommendations, lacking personalization, and their accuracy and reliability still need further improvement.

[0027] In view of this, this application provides a health status monitoring method and system based on vital sign data. This solution can determine that the collected vital signs are from the target user by obtaining user identifiers such as user fingerprints, thereby improving the reliability of the data collection source. Furthermore, by combining user location data, disease status, and historical data, health assessment results, lifestyle suggestions, and precautions are generated through preset database matching or target health monitoring model generation. Based on more data, report generation is beneficial for providing personalized lifestyle suggestions and precautions while monitoring user health, thus improving the user experience.

[0028] The vital sign monitoring device of this application integrates monitoring functions for heart rate, blood oxygen, microcirculation, blood pressure, respiratory rate, fatigue index, RR interval, and heart rate variability. By pressing a finger onto the device, it can collect the user's real-time vital sign data. More importantly, the device also includes a fingerprint collection function, allowing for user identification by collecting the user's fingerprint before collecting vital sign data. Furthermore, in other embodiments, the vital sign monitoring device may integrate other vital sign detection functions.

[0029] Based on the aforementioned vital sign monitoring device, this invention provides a health status monitoring method based on vital sign data, one optional step of which is as follows: Figure 1 As shown, Figure 2 This is an optional flowchart of an embodiment of this application. The method of this embodiment may include, but is not limited to, steps S100 to S600.

[0030] Step S100: Construct a preset database including a rule base, a feature pattern base, and a semantic knowledge graph; wherein the preset database is automatically updated based on a pre-trained data integration model.

[0031] Specifically, the pre-set database is used to perform pattern matching based on real-time vital sign data, user location data, and the user's medical condition to generate health assessment results, lifestyle suggestions, and precautions. This pre-set database is also equipped with an automatic update function, which automatically updates by periodically or irregularly fetching data from authoritative third parties.

[0032] It should be noted that the illness status refers to the user's most recently updated illness status in the software, which records whether the user has an illness and the specific name of the illness.

[0033] Furthermore, the pre-defined database includes a rule base, a feature pattern base, and a semantic knowledge graph, wherein: The rule base stores explicit, urgent, and mandatory health rules. For example, a systolic blood pressure greater than 180 indicates an urgent risk, and immediate medical attention is recommended. The rule base uses a high-performance relational database and establishes composite indexes based on disease type, location data, and urgency level, allowing queries to quickly and directly locate relevant health assessment results and recommendations.

[0034] The feature pattern library stores health patterns defined by vital sign vectors. Each pattern corresponds to a feature vector range, a health assessment result, and a set of recommendations (including lifestyle suggestions and precautions). This feature pattern library is constructed as a vector database, converting vital sign vectors into points in a high-dimensional space and establishing an approximate nearest neighbor index, enabling the system to find "similar" patterns without needing to access all the data.

[0035] The semantic knowledge graph stores the relationships between medical concepts in the form of a graph. This semantic knowledge graph is implemented using a graph database, enhancing contextual relevance and interpretability.

[0036] During querying, a three-level query mechanism is used to optimize the query logic, thereby achieving efficient querying. The three-level query is as follows: First, the monitored real-time vital signs data are matched against emergency rules in the rule base. If a match is found, all pre-defined emergency suggestions are returned directly. If no match is found, based on the user's illness and location data, a set of the most relevant general rules is retrieved from the rule base, and the search scope of the feature pattern library is determined.

[0037] Then, the real-time vital signs data are preprocessed into real-time feature vectors, and the approximate nearest neighbor method of the vector database is used to search within the above search range to obtain the most likely candidate results.

[0038] Based on semantic knowledge graphs, and combined with the user's illness status and current location data, the content of candidate results is fine-tuned as the final output.

[0039] Regarding automatic updates, this application automatically updates the preset database by constructing a data integration model.

[0040] Specifically, the data integration model includes a data parsing module, a structured processing module, a conflict handling module, an encapsulation module, and a writing module.

[0041] Furthermore, after retrieving third-party data, the data parsing module calls different parsers (such as PDF parsers, HTML parsers, JSON parsers, etc.) to perform text parsing on the obtained third-party data and output a unified raw text block.

[0042] The structured processing module is built on natural language processing technology. It obtains standardized raw text and uses a pre-trained medical NER model to identify key entities such as vital signs, numerical ranges, medical concepts, life advice and precautions from the raw text blocks. It also determines the relationships between entities through relation extraction. Furthermore, it performs semantic role labeling on the raw text blocks and outputs a set of structured information.

[0043] The conflict resolution module acquires the structured information set obtained above. Due to different sources, this information may contain duplicates or conflicts. In this embodiment, after clustering, conflict monitoring, and conflict resolution, an integrated list of candidate health patterns is output. The clustering method primarily groups information describing the same health pattern together; for example, information about "resting heart rate," "resting blood oxygen," and "normal respiratory rate" is merged into a single "healthy resting state" pattern. When different sources describe the same indicator in inconsistent ways, conflict resolution is required. This embodiment employs a combined approach of prioritizing authoritative sources, prioritizing newly released data, and voting to resolve conflicts, ultimately yielding the candidate health pattern list.

[0044] The encapsulation module retrieves the aforementioned list of candidate health patterns, maps the integrated information to the field structure of a preset database for pattern encapsulation, and generates a complete record to be written to the database. It also calculates a confidence score for each record, based on factors including the authority weight of the data source, the consistency of information across different sources, and the prediction probability of the relation extraction model. This module outputs data records that fully conform to the requirements of the preset database table, along with a confidence score.

[0045] Based on this, the working steps of the data integration model are as follows: ①~⑦: ① Triggered by a scheduled task or manually by the administrator, it begins to pull raw data from the configured authoritative third-party data source.

[0046] ② The original data is parsed into unified text using the data parsing module to obtain the original text block.

[0047] ③ The original text block is sent to the NER and structured processing module, and a series of structured information triples are output.

[0048] ④ The information triples are clustered and conflict resolved through the conflict handling module to form a clear and contradictory list of candidate health patterns.

[0049] ⑤ Encapsulate the candidate health pattern list into a database record format to obtain the target data, and calculate a confidence score for the target data.

[0050] ⑥ The subsequent processing of the data is determined based on the confidence score. Specifically, for high confidence records (e.g., confidence greater than 0.9): they are directly updated to the preset database; for medium and low confidence records: they are sent to the manual review queue, and after final review and confirmation by medical experts, they are entered into the database by the administrator.

[0051] ⑦ After new data is entered into the database, a new ID and time record are generated, and the old data is marked as invalid.

[0052] Step S200: Construct an initial health monitoring model and train the initial health monitoring model to obtain a target health monitoring model; wherein, the target health monitoring model includes a data input module, a data preprocessing module, a health assessment module, a context fusion and decision-making module, a suggestion generation module, and a data output module.

[0053] The health monitoring model constructed in this application mainly performs health assessments based on the acquired vital sign data, obtains health assessment results, and further generates lifestyle suggestions and precautions based on the health assessment results, current location data, disease status, and historical data.

[0054] Specifically, the health monitoring model of this application includes a data input module, a data preprocessing module, a health assessment module, a context fusion and decision-making module, a suggestion generation module, a data output module, and a feedback module. These modules are described in detail below: The data input module is used to collect and receive all raw data. The input data includes: real-time vital sign data obtained through the aforementioned vital sign monitoring equipment, the user's current location information, the user's medical history, and historical data. Specifically, the user's current location data is used to match the user's current location with climate information. Combining this climate information with the user's medical history data helps to more accurately generate customized lifestyle suggestions and precautions for the user. It should be noted that historical data includes the user's historical vital sign data and historical health assessment results.

[0055] Specifically, this module acquires real-time vital sign data and location information from the interface of the vital sign monitoring device, reads the disease information preset by the user in the software, and reads historical vital sign data and historical health assessment results stored over a period of time as input to the model.

[0056] The data preprocessing module is used to clean, standardize, and normalize the raw input data obtained from the data input module to ensure that the data quality is suitable for subsequent analysis.

[0057] Specifically, real-time vital sign data and disease status are cleaned, standardized, and normalized to obtain the first data; current location data and historical data are standardized to obtain the second data. Furthermore, during the data cleaning process, reasonable value ranges for each vital sign and location data are configured according to standard medical criteria, and invalid or abnormal data are corrected or marked.

[0058] The health assessment module is used to extract features from the first data, extracting representative features such as time-domain features, frequency-domain features, and nonlinear features. It uses recurrent neural networks (RNNs) or temporal convolutional networks to capture long-term and complex temporal patterns between features. The recurrent neural network is preferably a long short-term memory network (LSTM). The health status is divided into categories such as normal, sub-healthy, warning, and abnormal to obtain the health assessment results.

[0059] Furthermore, taking an LSTM network as an example, the health assessment module includes an input layer, multiple LSTM layers, a fully connected layer, and an output layer. The input layer receives the first data sequentially; the LSTM layers are three bidirectional LSTM layers, each with 256, 128, and 64 units respectively. A normalization and Dropout regularization layer is added after each LSTM layer to prevent overfitting; after the LSTM layers, a fully connected layer is added for classification, and the probabilities of the four classes are output through softmax.

[0060] The context fusion and decision module is used to identify the contextual relationships of the preprocessed data; the contextual relationships are identified through a knowledge graph of real-time vital signs data, current location data, disease status and historical data.

[0061] Specifically, by constructing a knowledge graph that includes health assessment results, diseases, environmental factors, lifestyle suggestions, and precautions, the complex relationships between different factors are inferred. Users' historical health status is stored as vectors, and similarity searches are used to quickly find similar historical health conditions, providing a reference for current decision-making. Furthermore, the similarity search is primarily implemented using a cosine similarity algorithm.

[0062] After processing by the context fusion decision module, the context relationship is obtained.

[0063] The suggestion generation module is built using a large language model to generate personalized life suggestions and precautions based on context.

[0064] The data output module is used to adapt and output the health assessment results, lifestyle suggestions and precautions in the correct format.

[0065] The feedback module is used to obtain user feedback on the output results and optimize the model based on the user feedback.

[0066] Based on this, the training process of the health monitoring model includes: acquiring vital sign data and corresponding location data, disease status, historical data, health assessment results, lifestyle suggestions and precautions for training, and constructing an original dataset; dividing the original dataset into training dataset and test dataset according to the proportion; inputting the training dataset into the pre-trained initial health monitoring model for parameter learning to obtain the first health monitoring model; using the test dataset to optimize the first health monitoring model using a combination-based cross-entropy loss function, and obtaining the target health monitoring model after the loss value meets the preset conditions.

[0067] Step S300: Obtain the user's fingerprint and determine the user's identity based on the preset fingerprint.

[0068] Specifically, in the health monitoring embodiment of this application, upon first use, the target user is guided to register their fingerprint features as a preset fingerprint. Each subsequent use involves collecting the fingerprint through the aforementioned vital sign monitoring device and comparing it with the preset fingerprint to confirm that the collected data originates from the target user. This reduces human interference in data collection without requiring further user intervention. If fingerprint matching fails, a mismatch notification is provided to the user.

[0069] Furthermore, in some embodiments, a single vital sign monitoring device can serve multiple target users, categorizing the collected vital sign data based on fingerprint recognition results. When no matching fingerprint is detected, the user is prompted to register their fingerprint. Subsequent health status assessment reports are generated solely based on the historical and location data of the currently matched user.

[0070] Step S400: Upon confirming the identity of the target user, simultaneously collect the user's current location data and real-time vital signs data.

[0071] By using step S400, it can be confirmed that the collected vital sign data comes from the user, ensuring the accuracy of the data source, preventing the accidental collection of data from other people, contaminating historical data, interfering with subsequent health assessments and suggestion generation, and improving the accuracy of health status monitoring and report generation to a certain extent.

[0072] Step S500: Obtain the user's medical history and historical data; wherein, the historical data includes historical vital sign data and historical detailed assessment results.

[0073] Disease status can be obtained from the application terminal using the monitoring method of this application. Historical data includes vital sign data and detailed historical assessment results; specifically, after each test, relevant data is recorded in the aforementioned application terminal to form historical data.

[0074] Step S600: Based on the user's selection, a brief health status assessment report is obtained by matching the current location data, real-time vital signs data, and illness status from a preset database; or, the current location data, real-time vital signs data, illness status, and historical data are input into the target health monitoring model to generate a detailed health status assessment report.

[0075] Specifically, a brief evaluation result is generated by default, and users can trigger a more detailed evaluation result.

[0076] When generating a brief assessment result, the brief assessment result is matched from a preset database based on real-time vital signs data, current location data, and the latest disease status.

[0077] When generating detailed assessment results, real-time vital signs data, current location data, the latest disease status, and historical data (including historical detailed assessment data and historical vital signs data) are input into the health status monitoring model for evaluation to obtain detailed assessment results.

[0078] In summary, the embodiments of this application have at least the following beneficial effects: 1. Integrating fingerprint detection functionality into monitoring devices that collect vital sign data via fingers can verify the accuracy of the data source without requiring additional user intervention. This prevents the device from mistakenly collecting data from other individuals, which could interfere with subsequent health assessments and recommendations, thus improving the accuracy of the generated data to some extent.

[0079] 2. By combining user location data, disease status, and historical data, health assessment results, lifestyle suggestions, and precautions are generated through preset database matching or target health monitoring models. The generation of reports based on more data is conducive to providing personalized lifestyle suggestions and precautions while monitoring users' health, thereby improving the user experience.

[0080] 3. It is divided into brief assessment and detailed assessment, and generates reports according to user needs, providing customized health status monitoring and improving data processing speed.

[0081] Please see Figure 3 This application also provides a health status monitoring system based on vital sign data, which can implement the above-mentioned health status monitoring method based on vital sign data. The system includes: The first module 101 is used to construct a preset database including a rule base, a feature pattern base, and a semantic knowledge graph; wherein the preset database is automatically updated based on a pre-trained data integration model; The second module 102 is used to construct an initial health monitoring model and train the initial health monitoring model to obtain a target health monitoring model; wherein, the target health monitoring model includes a data input module, a data preprocessing module, a health assessment module, a context fusion and decision-making module, a suggestion generation module, and a data output module; The third module 103 is used to acquire the user's fingerprint and determine the user's identity based on the preset fingerprint. The fourth module 104 is used to collect the user's current location data and real-time vital signs data after confirming that the identity is that of the target user. The fifth module 105 is used to acquire the user's medical history and historical data; wherein, the historical data includes historical vital sign data and historical detailed assessment results; The sixth module 106 is used to obtain a brief health status assessment report from the preset database based on the user's selection, the current location data, the real-time vital signs data, and the illness status; or, input the current location data, the real-time vital signs data, the illness status, and the historical data into the target health monitoring model to generate a detailed health status assessment report.

[0082] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0083] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned health status monitoring method based on vital sign data. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0084] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0085] Please see Figure 4 , Figure 4 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 301 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 302 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 302 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 302 and is called and executed by the processor 301. Input / output interface 303 is used to implement information input and output; The communication interface 304 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 305 transmits information between various components of the device (e.g., processor 301, memory 302, input / output interface 303, and communication interface 304); The processor 301, memory 302, input / output interface 303, and communication interface 304 are connected to each other within the device via bus 305.

[0086] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0087] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0088] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0089] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0090] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0091] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0093] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0094] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0095] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, apparatuses, or units, and may be electrical, mechanical, or other forms.

[0096] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0097] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0098] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0099] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for monitoring health status based on vital sign data, characterized in that, Includes the following steps: A pre-defined database is constructed, comprising a rule base, a feature pattern base, and a semantic knowledge graph; wherein the pre-defined database is automatically updated based on a pre-trained data integration model; An initial health monitoring model is constructed, and the initial health monitoring model is trained to obtain a target health monitoring model; wherein, the target health monitoring model includes a data input module, a data preprocessing module, a health assessment module, a context fusion and decision-making module, a suggestion generation module, and a data output module; Obtain the user's fingerprint and determine the user's identity based on the preset fingerprint; Once the identity is confirmed to be that of the target user, the user's current location data and real-time vital signs data are collected simultaneously. Obtain the user's medical history and historical data; wherein, the historical data includes historical vital sign data and historical detailed assessment results; Based on the user's selection, a brief health status assessment report is obtained by matching the current location data, the real-time vital signs data, and the illness status from the preset database; or, the current location data, the real-time vital signs data, the illness status, and the historical data are input into the target health monitoring model to generate a detailed health status assessment report.

2. The method according to claim 1, characterized in that, The construction of the pre-defined database, which includes a rule base, a feature pattern base, and a semantic knowledge graph, includes the following steps: A rule base is constructed using a relational database, wherein the rule base is used to store explicit, urgent, or mandatory health rules; A vector database is constructed, and an approximate nearest neighbor index is established in the vector database to obtain a feature pattern library; wherein, the feature pattern library is used to store health patterns; each health pattern includes a feature vector range, a preset health assessment result corresponding to the feature vector range, and a set of preset suggestions; A semantic knowledge graph for storing relationships between medical concepts is constructed using a graph database; A preset database is constructed based on the rule base, the feature pattern base, and the semantic knowledge graph.

3. The method according to claim 1, characterized in that, The data integration model includes a data parsing module, a structured processing module, a conflict handling module, an encapsulation module, and a writing module; the automatic update of the preset database includes the following steps: The data parsing module calls different interpreters to parse the data from third parties and outputs a unified raw text block. The structured processing module identifies key entities from the original text block, extracts relationships from the key entities, and performs semantic role labeling on the text of the key entities based on the extraction results to obtain a set of structured information; wherein, the key entities include vital signs, numerical ranges, medical concepts, lifestyle suggestions, and precautions; The conflict handling module performs conflict detection on the structured information set, processes the detected conflicts, and obtains a candidate health mode list; wherein, the conflict handling includes prioritizing authority, prioritizing newly released data, and voting processing; The encapsulation module encapsulates the candidate health pattern list to generate target data that conforms to the database format, and calculates a confidence score for the target data. The writing module writes the target data with the confidence score into the preset database, records the writing time, and marks the replaced old data as invalid, thus completing the automatic update.

4. The method according to claim 1, characterized in that, The construction of the initial health monitoring model includes the following steps: A data input module is constructed to receive all raw input data; wherein, the raw input data includes real-time vital signs data, current location data, disease status and historical data; A data preprocessing module is constructed to perform data cleaning, data standardization, and data normalization on the raw input data; A health assessment module is constructed based on recurrent neural networks or temporal convolutional networks to generate health assessment results from preprocessed data. A context fusion and decision-making module is constructed to identify the contextual relationships of preprocessed data; wherein, the contextual relationships are identified through a knowledge graph of the real-time vital signs data, the current location data, the disease status, and the historical data. A suggestion generation module is built based on a large language model to generate personalized life suggestions and precautions according to the contextual relationships. A data output module is constructed to adapt and output the health assessment results, lifestyle suggestions, and precautions in the specified formats, thereby obtaining the output results. A feedback module is constructed to obtain user feedback on the output results; The data input module, the data preprocessing module, the health assessment module, the context fusion decision module, the data output module, and the feedback module are integrated into an initial health monitoring model.

5. The method according to claim 1, characterized in that, The process of training the initial health monitoring model to obtain the target health monitoring model includes the following steps: Acquire vital sign data and corresponding location data, disease status, historical data, health assessment results, lifestyle suggestions and precautions for training, and construct the original dataset; The original dataset is divided into a training dataset and a test dataset according to a certain ratio; The training dataset is input into the pre-trained initial health monitoring model for parameter learning to obtain the first health monitoring model; The first health monitoring model is optimized using the test dataset based on a combination-based cross-entropy loss function. The target health monitoring model is obtained after the loss value meets the preset conditions.

6. The method according to claim 1, characterized in that, The process of obtaining a user's fingerprint and determining the user's identity based on a preset fingerprint includes the following steps: Obtain the current user's fingerprint; wherein the current user's fingerprint is collected before collecting real-time vital signs data; The current user fingerprint is compared with the user's preset fingerprint to obtain the comparison result; The user's identity is determined based on the comparison results.

7. The method according to claim 1, characterized in that, The step of obtaining a brief health assessment report by matching the current location data, the real-time vital signs data, and the illness status from the preset database includes the following steps: The real-time vital signs data are matched with emergency rules in the rule base; If a match is found, all the preset emergency suggestions found are returned directly; if no match is found, based on the illness and the current location data, a set of the most relevant general rules are retrieved from the rule base, and the search range of the feature pattern library is determined based on the general rules. The real-time vital signs data are preprocessed into real-time feature vectors, and the approximate nearest neighbor method of the vector database is used to retrieve candidate results within the search range. Based on the knowledge graph of semantics, combined with the disease status and the current location data, the content of the candidate results is optimized to obtain a brief health status monitoring report.

8. The method according to claim 1, characterized in that, The process of inputting the current location data, real-time vital sign data, disease status, and historical data into the target health monitoring model to generate a detailed health status assessment report includes the following steps: The system receives the current location data, the real-time vital signs data, the illness status, and the historical data through the data input module. The data preprocessing module performs data cleaning, standardization, and normalization on the real-time vital signs data and the disease status to obtain the first data; the current location data and the historical data are standardized to obtain the second data. The health assessment module extracts features from the first data, identifies long-term dependencies and complex temporal patterns between features, and then identifies the health status to obtain a health assessment result; wherein, the health status includes normal, sub-healthy, warning, and abnormal. The context fusion decision module fuses the health assessment results, the current location data, and the historical data to obtain the contextual relationship; Based on the contextual relationships, the suggestion generation module automatically generates personalized life suggestions and precautions in natural language. The data output module adapts and outputs the health assessment results, lifestyle suggestions, and precautions in the correct format to obtain a detailed health status assessment report.

9. A health status monitoring system based on vital sign data, characterized in that, include: The first module is used to construct a pre-set database including a rule base, a feature pattern base, and a semantic knowledge graph; wherein, the pre-set database is automatically updated based on a pre-trained data integration model; The second module is used to construct an initial health monitoring model and train the initial health monitoring model to obtain a target health monitoring model; wherein, the target health monitoring model includes a data input module, a data preprocessing module, a health assessment module, a context fusion and decision-making module, a suggestion generation module, and a data output module; The third module is used to acquire the user's fingerprint and determine the user's identity based on the preset fingerprint. The fourth module is used to collect the user's current location data and real-time vital signs data after confirming that the identity is that of the target user. The fifth module is used to obtain the user's medical history and historical data; wherein, the historical data includes historical vital sign data and historical detailed assessment results; The sixth module is used to obtain a brief health status assessment report from the preset database based on the user's selection, the current location data, the real-time vital signs data, and the illness status; or, to input the current location data, the real-time vital signs data, the illness status, and the historical data into the target health monitoring model to generate a detailed health status assessment report.

10. An electronic device, characterized in that, Including the processor and memory; The memory is used to store programs; The processor executes the program to implement the method as described in any one of claims 1 to 8.

Images