A resident health diary integrated management and intelligent grading early warning method and system

Abstract

The present application relates to a kind of resident health log integrated management and intelligent grading early warning method and system, the method includes: with medical institution interface synchronization in-hospital medical data;Receive the health data that resident uploads through OCR identification outside hospital medical data and records by oneself;Data are integrated and classified;Provide the aggregation analysis of multiple categories index and data real-time sharing to outside;Add health index abnormal monitoring and intelligent early warning function, by building standard index threshold library, the health index of real-time warehousing is flow detection, identifies single index or combined index abnormal and carries out grading early warning, through multi-channel directional push to relevant person in charge, and early warning record is archived to health log, realize monitoring-early warning-disposal-trace closed-loop management, the application of the present application improves health log system, improves the initiative and practicality of health management.

Description

Technical Field

[0001] This invention relates to the field of resident health log management technology, and more specifically, to a method and system for integrated management and intelligent hierarchical early warning of resident health logs. Background Technology

[0002] When doctors see patients, they often need to take detailed medical histories and review the patient's medical records and various examination and test reports from different medical institutions on the spot, which is time-consuming and labor-intensive. At the same time, due to the correlation between diseases, if the patient fails to bring relevant medical records or does not describe the relevant information about the condition in a comprehensive manner, it can easily lead to the doctor's diagnostic errors. Such problems are even more prominent when residents consult doctors online. For a long time, due to the limitations of related technologies, it has been impossible to integrate, manage, classify, analyze and share residents' medical records, examination and test reports and home monitoring data from various medical institutions in real time, which is not conducive to improving the efficiency of clinical doctors' diagnosis and treatment and the accuracy of diagnosis. Meanwhile, most existing health management systems are limited to data collection and display, lacking automatic identification and proactive early warning mechanisms for abnormal health indicators. Residents and doctors often only discover problems after reviewing the data, making early intervention difficult. Especially in chronic disease management and home health monitoring scenarios, this lag may lead to delayed treatment of health risks, failing to meet the needs of modern proactive health management. Therefore, there is an urgent need for a method and system for integrated management of resident health logs and intelligent hierarchical early warning that can solve the above problems. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method for integrated management and intelligent hierarchical early warning of resident health logs, and to provide a system for integrated management and intelligent hierarchical early warning of resident health logs, in view of the above-mentioned defects of the prior art.

[0004] The technical solution adopted by this invention to solve its technical problem is: A method for integrated management and intelligent hierarchical early warning of resident health logs is constructed, which includes the following steps: Connect with medical institutions to simultaneously receive in-hospital medical data generated within those institutions; Receive outpatient medical data and self-recorded health data uploaded by residents using OCR recognition technology; Integrate and categorize in-hospital medical data, out-of-hospital medical data, and self-recorded health data; It provides aggregated analysis of multiple categories of indicators and real-time data sharing to external parties; It also includes steps for monitoring and issuing early warnings of abnormal health indicators: Build a standard indicator threshold library and configure normal ranges and graded early warning thresholds for different dimensions; Streaming detection is performed on real-time health indicators to determine whether the indicators exceed the threshold or experience a sudden change in trend. The warning levels are classified according to the degree of abnormality, and the warning information is pushed out through designated channels. Early warning records and handling results are archived in the health log to form a closed-loop management system of monitoring, early warning, handling, and traceability.

[0005] The resident health log integrated management and intelligent hierarchical early warning method of the present invention, wherein the in-hospital medical data generated from medical institutions includes historical data; The historical data synchronization method is as follows: Step 1: Query the maximum value of the unique identifier field in the hospital's view or database table; Step 2: Determine if the custom identifier value in the local dictionary table exists. If it does not exist, query the minimum value of the unique identifier field in the hospital view or table and assign it to the custom identifier value in the local dictionary table. If it exists and is greater than the maximum value, terminate the subsequent operation; otherwise, proceed to the next step. Step 3: Query the hospital's view or table in pages. If an error occurs during the query, terminate the subsequent operation. If the current query result is empty, proceed to step 5; otherwise, proceed to the next step. Step 4: Execute the queried results in batches using multiple threads; determine whether to save based on the unique identifier value in the local synchronization table. If it doesn't exist, save; otherwise, do nothing. The main thread will be blocked while a child thread is processing business logic. The main thread will only continue execution after all child threads have finished. If any child thread encounters an exception, the other child threads will terminate, and subsequent operations will not be executed; otherwise, proceed to the next step. Step 5: Update the custom identifier value in the local dictionary table. If the update fails, subsequent operations will not be executed; otherwise, proceed to Step 2.

[0006] The integrated management and intelligent hierarchical early warning method for resident health logs described in this invention includes, in this invention, incremental data in the in-hospital medical data generated from medical institutions. The incremental data synchronization method is as follows: Step 1: Start a scheduled task to perform incremental data synchronization operations daily; Step 2: Query the maximum value of the unique identifier field in the hospital's view or table; Step 3: Determine if the custom identifier value in the local dictionary table is greater than the maximum value. If it is greater than the maximum value, terminate the synchronization of the current table and perform the synchronization operation of other tables; otherwise, proceed to the next step. Step 4: Update the query results to the table. If an error occurs during the current operation, terminate the synchronization of the current table. Otherwise, check if the unique identifier of the table needs to be updated. If it does not exist, save it; otherwise, update it. Step 5: Update the custom identifier value of the local dictionary table. If the update fails, terminate the synchronization of the current table and perform the synchronization operation of other tables. Otherwise, go to step 3.

[0007] The resident health log integrated management and intelligent hierarchical early warning method of the present invention includes a method for receiving outpatient medical data uploaded by residents through OCR recognition technology and self-recorded health data: OCR technology is used to structure and organize the standard fields of text-based medical records and to identify and archive the test indicators in tabular medical record reports. The aforementioned test indicators, together with the vital signs data recorded by residents in their daily lives, constitute the resident health indicator database; By aggregating and analyzing trends of individual indicators in the resident health indicator database, and using a configured standard indicator mapping rule library, the actual indicator name is fuzzily matched with the index matching words in the mapping library. The actual indicator is then classified as the one with the highest matching degree among the matched standard indicators, and the indicators are archived, trend displayed, and anomaly analyzed.

[0008] The resident health log integrated management and intelligent hierarchical early warning method of the present invention includes a method for integrating and classifying in-hospital medical data, out-of-hospital medical data, and self-recorded health data: The system integrates in-hospital medical data, out-of-hospital medical data, and self-recorded health data, and aggregates and displays all medical records generated from the same visit, categorizing them based on the visit time rather than the report time. The method further includes: It supports the classification, aggregation, and display of medical records based on multiple dimensions such as consultation time, medical record type, and medical record source, and supports refined filtering based on dimensions such as the hospital, department, medical record type, or time range. Residents can share their personal health logs with third parties by sharing QR codes, meeting the needs of online and offline medical treatment and sharing among acquaintances. When a viewer scans the QR code to view the content, the following three levels of permission checks are required: 1) Verify whether the QR code is currently enabled. If yes, continue to the next round of verification; otherwise, indicate that the health log is not visible. 2) Verify whether the WeChat user has been authorized. If so, proceed to the next round of verification; otherwise, allow the user to reapply for verification. 3) Verify whether the WeChat user's authorization has expired. If so, display the sharer's health log system information; otherwise, allow the user to reapply for permission to view the information. If all the above checks pass, the display will be based on the visibility range configured by the user for the current user. The visibility range can be divided and selected according to multiple dimensions such as log type, log source, and log time, and it also supports the configuration of composite visibility ranges.

[0009] The integrated management and intelligent hierarchical early warning method for resident health logs described in this invention, wherein the construction of a standard indicator threshold library and the configuration of normal ranges and hierarchical early warning thresholds for different dimensions specifically include: Based on medical standards, corresponding normal value ranges, warning thresholds, and danger thresholds are configured according to at least one dimension, such as age, gender, specific disease, or menstrual cycle. It supports doctors in adjusting custom thresholds for specific indicators based on individual resident differences or settings made by the resident themselves.

[0010] The resident health log integrated management and intelligent hierarchical early warning method of the present invention, wherein the step of performing streaming detection on real-time entered health indicators to determine whether the indicators exceed the threshold or experience a sudden trend change specifically includes: Real-time or near real-time streaming computation is performed on synchronous data within the hospital, OCR-recognized data, and manually entered data. Single indicator anomaly judgment: When the indicator value exceeds the preset threshold, or exceeds the normal range N times consecutively, or the change range exceeds the sudden change threshold within the set time window, it is judged as abnormal; Combined indicator anomaly detection: Preset multi-indicator joint logic rules, when multiple related indicators meet the abnormal conditions at the same time, a combined anomaly flag is triggered; The value of N in the N consecutive times exceeding the normal range can be dynamically adjusted according to the indicator type or user settings. The combined indicator joint logic rules include logical AND, logical OR and their combinations, which are used to improve the accuracy of risk identification for chronic disease complications or complex pathological states.

[0011] The integrated management and intelligent hierarchical early warning method for resident health logs described in this invention, wherein the step of classifying early warning levels according to the degree of abnormality specifically includes dividing the abnormalities into three levels: Level 1 warning: The indicator deviates slightly from the normal range and is judged as a minor abnormality; Level 2 warning: If the indicator continues to deviate from or exceed the warning threshold, it is judged as a moderate abnormality; Level 3 warning: The indicator reaches the danger threshold or shows a rapid deterioration trend, and is judged as a serious abnormality; The targeted push of early warning information through designated channels specifically includes establishing a tiered push routing strategy: Level 1 warning information is primarily sent to residents themselves, reminding them to be vigilant in their daily lives; Level 2 warning information is pushed to the resident and their assigned doctor, reminding them to have a follow-up examination; Level 3 early warning information is pushed to the resident, their attending physician, family doctor, and pre-set emergency contact, prompting emergency intervention; The push channels include at least one of the following: internal system messages, WeChat service notifications, SMS, and telephone voice messages; The push notification includes the name of the abnormal indicator, its current value, historical comparison data, warning level, suggested handling measures, and a link to directly access the health log.

[0012] The integrated management and intelligent hierarchical early warning method for resident health logs described in this invention, wherein archiving early warning records and handling results into the health log specifically includes: Generate structured early warning event records, including trigger time, abnormal indicator details, early warning level, list of push recipients, and feedback status; Receive and link the handling feedback from doctors or residents, and update the handling status of the early warning event to "attention", "retested" or "visited". The system integrates complete records of early warning events with raw health data along a timeline, supporting retrospective queries based on consultation time or warning type.

[0013] A resident health log integrated management and intelligent hierarchical early warning system, applied to implement the method described above, the system comprising: The in-hospital data synchronization unit is used to connect with medical institutions and synchronize in-hospital medical data; The off-site data receiving unit is used to receive off-site data recognized by OCR and residents' self-recorded health data; The data integration and classification unit is used to clean, map, and unify the integration of multi-source heterogeneous health data; The data display unit is used to provide aggregated analysis and visualization sharing of indicators; The health indicator anomaly monitoring unit is used to perform threshold library management, real-time streaming detection, anomaly classification judgment, early warning information push and record archiving functions, including: The threshold configuration module is used to maintain standard thresholds and personalized thresholds based on multidimensional attributes; The streaming detection engine is used to monitor data inbound events and perform single indicator fluctuation calculations and multi-indicator logical correlation analysis. The early warning classification processor maps the detection results into different levels of early warning signals according to preset rules; The message push gateway connects to multiple communication interfaces to achieve tiered and targeted push notifications. The alert file manager is responsible for generating and storing alert records and indexing them in relation to health logs.

[0014] The beneficial effects of this invention are as follows: This invention further improves the application of the health log system, enabling residents' medical data in the hospital (including community health centers under the hospital's management) to be automatically uploaded in real time, and integrated and classified with the data from outside the hospital identified by OCR and the health data recorded by the residents themselves. At the same time, it supports the aggregation analysis of multiple categories of indicators and the real-time sharing of data. Doctors can clearly filter and view residents' relevant data and the trend changes and abnormalities of various indicators by type, time, specialty, institution, etc.

[0015] This invention enables the integrated display and aggregated analysis of residents' health data, as well as real-time sharing among multiple institutions and users under data security. It improves the convenience of online and offline medical visits for residents, while enhancing treatment efficiency and diagnostic accuracy. Furthermore, the newly added intelligent early warning module upgrades the system from "passive recording" to "proactive protection." This system can automatically detect potential health risks and reach relevant parties at different levels, effectively solving the problem of delayed risk detection. It is particularly suitable for the continuous management of chronic diseases such as hypertension and diabetes, significantly improving the safety of residents' health management and the efficiency of doctors' diagnostic and treatment interventions. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort: Figure 1 This is a flowchart of a preferred embodiment of the resident health log integrated management and intelligent hierarchical early warning method of the present invention; Figure 2 This is a flowchart of the in-hospital medical data synchronization process for the resident health log integrated management and intelligent hierarchical early warning method according to a preferred embodiment of the present invention; Figure 3 This is a flowchart of the health indicator aggregation and analysis method for the integrated management and intelligent hierarchical early warning of resident health logs, according to a preferred embodiment of the present invention. Figure 4 This is a flowchart illustrating the real-time sharing method for integrated management and intelligent hierarchical early warning of resident health logs, according to a preferred embodiment of the present invention. Figure 5 This is a flowchart illustrating the overall workflow of the resident health log integrated management and intelligent hierarchical early warning method for monitoring and intelligently warning of abnormal health indicators, which is a preferred embodiment of the present invention. Figure 6 This is a flowchart of the anomaly detection and hierarchical early warning method for integrated management and intelligent hierarchical early warning of resident health logs according to a preferred embodiment of the present invention; Figure 7 This is a block diagram illustrating the principle of a resident health log integrated management and intelligent hierarchical early warning system according to a preferred embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, a clear and complete description will be provided below in conjunction with the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0018] The preferred embodiment of the present invention is a method and system for integrated management and intelligent hierarchical early warning of resident health logs, such as... Figure 1 As shown, see also Figures 2-6 This includes the following steps: S01: Connect with medical institutions and receive in-hospital medical data generated within the medical institutions simultaneously; S02: Receive outpatient medical data and self-recorded health data uploaded by residents using OCR recognition technology; S03: Integrate and classify in-hospital medical data, out-of-hospital medical data, and self-recorded health data; S04: Provide aggregated analysis of multiple categories of indicators and real-time data sharing to external parties; S05: Construct a standard indicator threshold library and configure normal ranges and graded early warning thresholds for different dimensions; S06: Perform flow cytometry analysis on the health indicators entered into the database in real time to determine whether the indicators exceed the threshold or experience a sudden change in trend. S07: Classify the warning level according to the degree of abnormality, and push the warning information to designated channels; S08: Archive early warning records and handling results into the health log to form a closed-loop management of monitoring-early warning-handling-traceability; This invention further improves the application of the health log system, enabling residents' medical data at the hospital (including community health centers under the hospital's management) to be uploaded automatically in real time. It integrates and classifies the data with OCR-recognized data from outside the hospital and health data recorded by residents themselves. It also supports the aggregation analysis of multiple categories of indicators and real-time data sharing. Doctors can clearly filter and view residents' relevant data and the trend changes and abnormalities of various indicators by type, time, specialty, institution, etc.

[0019] This invention enables the integrated display and aggregated analysis of residents' health data, as well as real-time sharing among multiple institutions and users under data security. It improves the convenience of online and offline medical visits for residents, while enhancing treatment efficiency and diagnostic accuracy. Furthermore, the newly added intelligent early warning module upgrades the system from "passive recording" to "proactive protection." This system can automatically detect potential health risks and reach relevant parties at different levels, effectively solving the problem of delayed risk detection. It is particularly suitable for the continuous management of chronic diseases such as hypertension and diabetes, significantly improving the safety of residents' health management and the efficiency of doctors' diagnostic and treatment interventions.

[0020] It should be noted that the first and second steps mentioned above are not in any particular order and can be performed simultaneously.

[0021] Based on the views or tables provided by this medical institution, the health log system queries the required data from the corresponding views or tables and imports it into the health log system's database. The synchronized data includes both historical and incremental data. The specific synchronization process rules for these two types of data are as follows: Step 1: Query the maximum value of the unique identifier field in the hospital's view or database table; Step 2: Determine if the custom identifier value in the local dictionary table exists. If it does not exist, query the minimum value of the unique identifier field in the hospital view or table and assign it to the custom identifier value in the local dictionary table. If it exists and is greater than the maximum value, terminate the subsequent operation; otherwise, proceed to the next step. Step 3: Query the hospital's view or table in pages. If an error occurs during the query, terminate the subsequent operation. If the current query result is empty, proceed to step 5; otherwise, proceed to the next step. Step 4: Execute the queried results in batches using multiple threads; determine whether to save based on the unique identifier value in the local synchronization table. If it doesn't exist, save; otherwise, do nothing. The main thread will be blocked while a child thread is processing business logic. The main thread will only continue execution after all child threads have finished. If any child thread encounters an exception, the other child threads will terminate, and subsequent operations will not be executed; otherwise, proceed to the next step. Step 5: Update the custom identifier value in the local dictionary table. If the update fails, subsequent operations will not be executed; otherwise, proceed to Step 2. The health log system not only integrates data from both medical visits and daily health records, but also interfaces with hospital systems. The system pre-loads all medical records generated by the user within the connected medical institutions, greatly reducing the burden on users to upload their medical records.

[0022] Specifically, the technical solution for connecting health logs to medical institution visit data is as follows: The data source for patient visits provided by the hospital is usually presented in the form of a database view, which is a query result set based on one or more tables in the business database, in order to avoid affecting the hospital's internal business processes when querying data.

[0023] The health log system queries the medical record data required by the system from the above view and stores the query results in the health log system's database, which helps to improve the response speed and enhance the user experience when using the system.

[0024] The health log system uses a scheduled task approach. Every day at noon, it automatically executes a synchronization procedure according to a preset data synchronization process, synchronizing the data in the hospital's view to the health log system database. This ensures the real-time nature of the data while retaining the possibility of flexibly adjusting the data synchronization task.

[0025] The specific synchronization process rules are as follows: In the view provided by the hospital, based on the characteristics of relational databases, each medical record in the database view has a unique identifier. This identifier has an auto-incrementing attribute, meaning that the value of the identifier increases with the number of newly generated data.

[0026] Query the maximum value of the unique identifier field in the hospital view or database table; After each data synchronization task is completed, the health log system updates the dictionary table stored in the health log database with the maximum value of the unique identifier field of the synchronized data in this round. This value is managed as static fixed data and also serves as a reference in the data synchronization process.

[0027] Determine if the maximum value of the unique identifier field retrieved in step one exists in the health log system dictionary table: If it does not exist, meaning no data synchronization has been performed, query the minimum value of the unique identifier field in the hospital view and store it in the above field in the dictionary table; if it exists and is greater than or equal to the identifier value obtained in step one, meaning the health log system has been synchronized to the latest data, terminate the process; otherwise, proceed to the next step. Paginated queries of data in the hospital view involve dividing a large amount of data into fixed portions and returning a portion of the data for each query, thereby improving query efficiency.

[0028] Determine if the query returned an error: if so, terminate the process; otherwise, further verify if the query is empty. If the result is empty, proceed to step five; otherwise, continue to step four. The data obtained from paginated queries is divided into multiple batches, each containing a certain number of data items. A thread pool is created, and each child thread processes one batch of data.

[0029] The system determines whether to save a log based on the value of the unique identifier in the health log dictionary: if the unique identifier does not exist in the dictionary, it needs to be saved; otherwise, it does not need to be saved.

[0030] After all child threads have executed successfully, a save operation is performed. If any child thread encounters an exception, the other child threads will also terminate and no further operations will be performed.

[0031] After saving the partial data from the paginated query, use the maximum value of the unique identifier for this part of the data in the hospital's data view to update the identifier value of the health log system dictionary table, and continue with step 2.

[0032] Furthermore, the incremental data synchronization method is as follows: Step 1: Start a scheduled task to perform incremental data synchronization operations daily; Step 2: Query the maximum value of the unique identifier field in the hospital's view or table; Step 3: Determine if the custom identifier value in the local dictionary table is greater than the maximum value. If it is greater than the maximum value, terminate the synchronization of the current table and perform the synchronization operation of other tables; otherwise, proceed to the next step. Step 4: Update the query results to the table. If an error occurs during the current operation, terminate the synchronization of the current table. Otherwise, check if the unique identifier of the table needs to be updated. If it does not exist, save it; otherwise, update it. Step 5: Update the custom identifier value in the local dictionary table. If the update fails, terminate the synchronization of the current table and perform the synchronization operation on other tables; otherwise, go to step 3. The health log system utilizes OCR technology to not only support the structured organization of standard fields in text-based medical records, but also to recognize and archive relevant content from test indicators in tabular medical records such as laboratory reports and physical examination reports. These test indicators, along with residents' daily records of vital signs such as height, weight, blood sugar, and blood pressure, constitute the resident health indicator database. The health log system supports trend aggregation and analysis of individual indicators in all uploaded indicators. Using a pre-configured standard indicator mapping rule library, it fuzzily matches the actual indicator name with the matching terms in the mapping library, categorizing the actual indicator as the one with the highest matching degree among the matched standard indicators. The system then archives, displays trends, and performs anomaly analysis on the indicators.

[0033] The health log system supports the unified integration of structured data uploaded by residents via OCR recognition with medical data synchronously acquired by the medical institution. It aggregates and displays all medical records generated from the same visit, categorizing them primarily by visit time rather than report time. In scenarios involving medical visits and self-health management, this provides a more intuitive reflection of a resident's true health status over a specific period, enhancing the value of the integrated data.

[0034] The system supports the categorization and aggregation of medical records according to multiple dimensions such as consultation time, medical record type, and medical record source. It also supports fine-grained filtering according to dimensions such as the hospital, department, medical record type, and time range, which provides convenience for doctors to quickly locate medical records that are helpful for diagnosis and treatment by switching dimensions and filtering when seeing patients offline or online.

[0035] Preferred method for real-time sharing: Residents can share their personal health logs with third parties via QR code, meeting needs in scenarios such as online and offline medical treatment and sharing among acquaintances. To ensure the security of personal data, the health log system supports obtaining the viewer's unique WeChat identity identifier, using this identifier for multi-level authorization management. When a viewer scans the code to view the log, the following three levels of permission checks are required: 1) Verify whether the QR code is currently enabled. If yes, continue to the next round of verification; otherwise, indicate that the health log is not visible. 2) Verify whether the WeChat user has been authorized. If so, proceed to the next round of verification; otherwise, allow the user to reapply for verification. 3) Verify whether the WeChat user's authorization has expired. If so, display the sharer's health log system information; otherwise, allow the user to reapply for permission to view the information. 4) If all the above checks pass, the display will be based on the visibility range configured by the user for the current user. The visibility range can be divided and selected according to multiple dimensions such as log type, log source, and log time, and the configuration of composite visibility ranges is supported.

[0036] The specific details of the health indicator abnormality monitoring and early warning function are as follows: (1) Preset standard indicator threshold library Based on authoritative medical guidelines, a standard indicator threshold library has been established. Differentiated thresholds are pre-set for different population characteristics: Dimensional configuration: Configure corresponding normal ranges, warning thresholds, and danger thresholds according to age groups (e.g., adolescents, adults, elderly), gender, common chronic diseases (e.g., hypertension, diabetes, coronary heart disease), and special physiological cycles (e.g., pregnancy). For example, the blood pressure warning threshold for the elderly may be more lenient than that for adults.

[0037] Personalized customization: Doctors can customize thresholds for specific indicators under the guidance of a doctor, based on the patient's medical history (e.g., patients with renal insufficiency are more sensitive to creatinine levels) or the patient's own feelings. The threshold library supports dynamic hot-loading, and modifications take effect immediately.

[0038] (2) Real-time streaming anomaly detection The system incorporates a streaming computing engine (such as Flink or Kafka Streams) to process continuously incoming health metrics at the millisecond level. Data sources include laboratory data synchronized with the hospital's HIS system, external report data recognized by OCR, and home monitoring data manually entered by residents (such as smart bracelet step counts and home blood pressure monitor readings).

[0039] Single indicator detection: When the indicator value exceeds the preset threshold; or deviates from the normal range N times consecutively (N is configurable, default is 3 times); or the value changes drastically within a set short time window (such as 1 hour) (such as blood sugar dropping by 30%), an abnormality marker is triggered.

[0040] Combined indicator detection: A built-in rule engine supports setting complex logical combinations. For example, a rule can be defined as: "IF (blood pressure > 140 / 90 mmHg) AND (heart rate > 100 bpm) AND (symptoms include 'chest pain') THEN Trigger a high-risk cardiovascular alert." This multi-factor association analysis effectively reduces false positives and accurately captures complex pathological states.

[0041] (3) Anomaly classification and intelligent push To avoid alert fatigue, the system classifies anomalies into three levels and establishes a precise hierarchical push route: Level 1 Warning (Mild): Indicators are slightly abnormal but pose no immediate threat (e.g., occasionally slightly elevated fasting blood glucose). Target audience: Resident only. Channel: In-system pop-up or WeChat notification; content suggestion: "Pay attention to your diet and retest the next day."

[0042] Level 2 Warning (Moderate): Indicators remain above the standard or show a significant worsening trend (e.g., high blood pressure for 3 consecutive days). Target audience: Resident and their registered family doctor / attending physician. Channels: WeChat service notification + SMS, containing specific numerical comparisons and the instruction to "return for a follow-up appointment within 3 days".

[0043] Level 3 Warning (Severe): Indicators reach the red line or vital signs are extremely unstable (e.g., blood oxygen saturation <90%). Target audience: Residents + Attending physician + Emergency contact (family member). Channels: High-frequency SMS + telephone voice call, emphasizing urgency and providing a quick access point to the emergency center with one click.

[0044] (4) Early warning record archiving and closed loop The system automatically generates standardized early warning event files, including: trigger timestamp, abnormal snapshot, early warning level, push serial number, and recipient list. Once a doctor or patient responds (e.g., a doctor issues a prescription, a patient confirms a visit), the response can be uploaded back to the system, updating the event status to "handled." All early warning files are seamlessly embedded into residents' health logs along a timeline. When reviewing historical medical records, doctors can clearly trace the entire process of each health fluctuation, forming a complete data loop of "monitoring-early warning-intervention-review."

[0045] The early warning module uses a health indicator database to normalize indicators, ensuring consistent detection benchmarks; it relies on a data integration unit to acquire real-time data streams, eliminating the need to repeatedly build data collection channels; and it follows a QR code permission system to ensure that the scope of sharing early warning information is strictly controlled, preventing unauthorized personnel from viewing sensitive early warning history.

[0046] A resident health log integrated management and intelligent hierarchical early warning system is applied to the resident health log integrated management and intelligent hierarchical early warning method described above. Figure 7 As shown, the system includes an in-hospital data synchronization unit 100, an out-of-hospital data receiving unit 101, a data integration and classification unit 102, and a data display unit 103. The in-hospital data synchronization unit 100 is used to connect with medical institutions and synchronously receive in-hospital medical data generated within the medical institutions. The off-site data receiving unit 101 is used to receive off-site medical data uploaded by residents through OCR recognition technology and health data recorded by themselves. The data integration and classification unit 102 is used to integrate and classify in-hospital medical data, out-of-hospital medical data, and self-recorded health data; Data display unit 103 is used to provide aggregated analysis of multiple categories of indicators and real-time data sharing to external parties; It also includes a health indicator abnormality monitoring unit 104, which specifically includes: Threshold configuration module: Provides a graphical interface for administrators to maintain medical threshold parameters; Streaming detection engine: A persistent background service responsible for metric calculation and rule matching; Early warning classification processor: the decision-making center, which transforms detection results into specific early warning instructions; Message push gateway: encapsulates the communication protocols for connecting to WeChat, SMS, and voice interfaces; Warning File Manager: Responsible for CRUD operations on warning records and their associated indexes with the main log; This invention further improves the application of the health log system, enabling residents' medical data in the hospital (including community health centers under the hospital's management) to be uploaded automatically in real time, and integrated and classified with OCR-recognized data from outside the hospital and health data recorded by residents themselves. At the same time, it supports the aggregation analysis of multiple categories of indicators and real-time data sharing. Doctors can clearly filter and view residents' relevant data and the trend changes and abnormalities of various indicators by type, time, specialty, institution, etc. Furthermore, the newly added intelligent early warning module upgrades the system from "passive recording" to "proactive protection." This system can automatically capture potential health risks and reach relevant parties at different levels, effectively solving the problem of delayed risk detection. It is particularly suitable for the continuous management of chronic diseases such as hypertension and diabetes, significantly improving the safety of residents' health management and the efficiency of doctors' diagnosis and treatment intervention.

[0047] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A method for integrated management and intelligent hierarchical early warning of resident health logs, characterized in that, Includes the following steps: Connect with medical institutions to simultaneously receive in-hospital medical data generated within those institutions; Receive outpatient medical data and self-recorded health data uploaded by residents using OCR recognition technology; Integrate and categorize in-hospital medical data, out-of-hospital medical data, and self-recorded health data; It provides aggregated analysis of multiple categories of indicators and real-time data sharing to external parties; It also includes steps for monitoring and issuing early warnings of abnormal health indicators: Build a standard indicator threshold library and configure normal ranges and graded early warning thresholds for different dimensions; Streaming detection is performed on real-time health indicators to determine whether the indicators exceed the threshold or experience a sudden change in trend. The warning levels are classified according to the degree of abnormality, and the warning information is pushed out through designated channels. Early warning records and handling results are archived in the health log to form a closed-loop management system of monitoring, early warning, handling, and traceability.

2. The method for integrated management and intelligent hierarchical early warning of resident health logs according to claim 1, characterized in that, The in-hospital medical data generated within the medical institution includes historical data; The historical data synchronization method is as follows: Step 1: Query the maximum value of the unique identifier field in the hospital's view or database table; Step 2: Determine if the custom identifier value in the local dictionary table exists. If it does not exist, query the minimum value of the unique identifier field in the hospital view or table and assign it to the custom identifier value in the local dictionary table. If it exists and is greater than the maximum value, terminate the subsequent operation; otherwise, proceed to the next step. Step 3: Query the hospital's view or table in pages. If an error occurs during the query, terminate the subsequent operation. If the current query result is empty, proceed to step 5; otherwise, proceed to the next step. Step 4: Execute the query results in batches using multiple threads; The system determines whether to perform a save operation based on the unique identifier value in the local synchronization table. If the identifier does not exist, the operation is saved; otherwise, no action is taken. While a child thread is processing business logic, the main thread will be blocked. The main thread will only continue execution after all child threads have finished processing. If any child thread encounters an exception, all other child threads will terminate, and subsequent operations will not be executed; otherwise, the next step will proceed. Step 5: Update the custom identifier value in the local dictionary table. If the update fails, subsequent operations will not be executed; otherwise, proceed to Step 2.

3. The method for integrated management and intelligent hierarchical early warning of resident health logs according to claim 2, characterized in that, The in-hospital medical data generated within medical institutions also includes incremental data; The incremental data synchronization method is as follows: Step 1: Start a scheduled task to perform incremental data synchronization operations daily; Step 2: Query the maximum value of the unique identifier field in the hospital's view or table; Step 3: Determine if the custom identifier value in the local dictionary table is greater than the maximum value. If it is greater than the maximum value, terminate the synchronization of the current table and perform the synchronization operation of other tables; otherwise, proceed to the next step. Step 4: Update the query results to the table. If an error occurs during the current operation, terminate the synchronization of the current table. Otherwise, check if the unique identifier of the table needs to be updated. If it does not exist, save it; otherwise, update it. Step 5: Update the custom identifier value of the local dictionary table. If the update fails, terminate the synchronization of the current table and perform the synchronization operation of other tables. Otherwise, go to step 3.

4. The method for integrated management and intelligent hierarchical early warning of resident health logs according to claim 1, characterized in that, The method for receiving outpatient medical data and self-recorded health data uploaded by residents using OCR recognition technology includes: OCR technology is used to structure and organize the standard fields of text-based medical records and to identify and archive the test indicators in tabular medical record reports. The aforementioned test indicators, together with the vital signs data recorded by residents in their daily lives, constitute the resident health indicator database; By aggregating and analyzing trends of individual indicators in the resident health indicator database, and using a configured standard indicator mapping rule library, the actual indicator name is fuzzily matched with the index matching words in the mapping library. The actual indicator is then classified as the one with the highest matching degree among the matched standard indicators, and the indicators are archived, trend displayed, and anomaly analyzed.

5. The method for integrated management and intelligent hierarchical early warning of resident health logs according to claim 1, characterized in that, The method for integrating and classifying in-hospital medical data, out-of-hospital medical data, and self-recorded health data includes: The system integrates in-hospital medical data, out-of-hospital medical data, and self-recorded health data, and aggregates and displays all medical records generated from the same visit, categorizing them based on the visit time rather than the report time. The method further includes: It supports the classification, aggregation, and display of medical records based on multiple dimensions such as consultation time, medical record type, and medical record source, and supports refined filtering based on dimensions such as the hospital, department, medical record type, or time range. Residents can share their personal health logs with third parties by sharing QR codes, meeting the needs of online and offline medical treatment and sharing among acquaintances. When a viewer scans the QR code to view the content, the following three levels of permission checks are required: 1) Verify whether the QR code is currently enabled. If yes, continue to the next round of verification; otherwise, indicate that the health log is not visible. 2) Verify whether the WeChat user has been authorized. If so, proceed to the next round of verification; otherwise, allow the user to reapply for verification. 3) Verify whether the WeChat user's authorization has expired. If so, display the sharer's health log system information; otherwise, allow the user to reapply for permission to view the information. If all the above checks pass, the display will be based on the visibility range configured by the user for the current user. The visibility range can be divided and selected according to multiple dimensions such as log type, log source, and log time, and it also supports the configuration of composite visibility ranges.

6. The method for integrated management and intelligent hierarchical early warning of resident health logs according to claim 1, characterized in that, The construction of a standard indicator threshold library, configuring normal ranges and tiered early warning thresholds for different dimensions, specifically includes: Based on medical standards, corresponding normal value ranges, warning thresholds, and danger thresholds are configured according to at least one dimension, such as age, gender, specific disease, or menstrual cycle. It supports doctors in adjusting custom thresholds for specific indicators based on individual resident differences or settings made by the resident themselves.

7. The method for integrated management and intelligent hierarchical early warning of resident health logs according to claim 1, characterized in that, The process of performing streaming detection on real-time health indicators to determine whether the indicators exceed thresholds or experience sudden trend changes specifically includes: Real-time or near real-time streaming computation is performed on synchronous data within the hospital, OCR-recognized data, and manually entered data. Single indicator anomaly judgment: When the indicator value exceeds the preset threshold, or exceeds the normal range N times consecutively, or the change range exceeds the sudden change threshold within the set time window, it is judged as abnormal; Combined indicator anomaly detection: Preset multi-indicator joint logic rules, when multiple related indicators meet the abnormal conditions at the same time, a combined anomaly flag is triggered; The value of N in the N consecutive times exceeding the normal range can be dynamically adjusted according to the indicator type or user settings. The combined indicator joint logic rules include logical AND, logical OR and their combinations, which are used to improve the accuracy of risk identification for chronic disease complications or complex pathological states.

8. The method for integrated management and intelligent hierarchical early warning of resident health logs according to claim 1, characterized in that, The method of classifying early warning levels based on the degree of anomaly specifically includes dividing anomalies into three levels: Level 1 warning: The indicator deviates slightly from the normal range and is judged as a minor abnormality; Level 2 warning: If the indicator continues to deviate from or exceed the warning threshold, it is judged as a moderate abnormality; Level 3 warning: The indicator reaches the danger threshold or shows a rapid deterioration trend, and is judged as a serious abnormality; The targeted push of early warning information through designated channels specifically includes establishing a tiered push routing strategy: Level 1 warning information is primarily sent to residents themselves, reminding them to be vigilant in their daily lives; Level 2 warning information is pushed to the resident and their assigned doctor, reminding them to have a follow-up examination; Level 3 early warning information is pushed to the resident, their attending physician, family doctor, and pre-set emergency contact, prompting emergency intervention; The push channels include at least one of the following: internal system messages, WeChat service notifications, SMS, and telephone voice messages; The push notification includes the name of the abnormal indicator, its current value, historical comparison data, warning level, suggested handling measures, and a link to directly access the health log.

9. The method for integrated management and intelligent hierarchical early warning of resident health logs according to claim 1, characterized in that, The process of archiving early warning records and handling results into the health log specifically includes: Generate structured early warning event records, including trigger time, abnormal indicator details, early warning level, list of push recipients, and feedback status; Receive and link the handling feedback from doctors or residents, and update the handling status of the early warning event to "followed", "retested" or "visited". The system integrates complete records of early warning events with raw health data along a timeline, supporting retrospective queries based on consultation time or warning type.

10. A resident health log integrated management and intelligent hierarchical early warning system, applied to implement the method as described in any one of claims 1-9, the system comprising: The in-hospital data synchronization unit is used to connect with medical institutions and synchronize in-hospital medical data; The off-site data receiving unit is used to receive off-site data recognized by OCR and residents' self-recorded health data; The data integration and classification unit is used to clean, map, and unify the integration of multi-source heterogeneous health data; The data display unit is used to provide aggregated analysis and visualization sharing of indicators; The health indicator anomaly monitoring unit is used to perform threshold library management, real-time streaming detection, anomaly classification judgment, early warning information push and record archiving functions, including: The threshold configuration module is used to maintain standard thresholds and personalized thresholds based on multidimensional attributes; The streaming detection engine is used to monitor data inbound events and perform single indicator fluctuation calculations and multi-indicator logical correlation analysis. The early warning classification processor maps the detection results into different levels of early warning signals according to preset rules; The message push gateway connects to multiple communication interfaces to achieve tiered and targeted push notifications. The alert file manager is responsible for generating and storing alert records and indexing them in relation to health logs.

Images