Medical and prevention fusion information management method and system based on multi-source heterogeneous data and medium

Abstract

The application provides a medical and prevention fusion information management method and system based on multi-source heterogeneous data and a medium, the method comprising: collecting multi-source heterogeneous medical and prevention data under a primary medical scene; standardizing, cleaning, format converting and data mapping the multi-source heterogeneous medical and prevention data to obtain preprocessed data, constructing medical and prevention data standard data, performing data alignment and fusion processing to obtain fusion data; intelligently analyzing, rule checking and logically quality control analyzing the fusion data based on a medical big model and a medical knowledge graph to obtain analysis results, identifying abnormal data, and correcting the abnormal data to obtain normal data; obtaining follow-up data according to medical and prevention business rules and performance evaluation indexes; outputting standardized medical and prevention business results, health evaluation reports and performance statistical data according to a multi-agent collaborative algorithm; and realizing cross-system real-time intercommunication, eliminating data barriers and greatly improving work efficiency through standardization, data mapping and normalization processing.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence and data fusion technology, specifically to a medical and prevention fusion information management method, system, and medium based on multi-source heterogeneous data. Background Technology

[0002] Primary healthcare institutions undertake core tasks such as chronic disease management, physical examinations for the elderly, resident health records, follow-ups, and public health assessments, and face the following prominent challenges:

[0003] Multi-source heterogeneous data are severely isolated and cannot be effectively integrated;

[0004] Primary healthcare systems often operate independently, including HIS, LIS, public health, physical examination, medical insurance, and smart hardware. Data standards are not unified, interfaces are not open, and paper, Excel, and business systems coexist, making it impossible to share data and creating a large number of data silos, which makes it difficult to support integrated management of medical treatment and prevention.

[0005] Manual operation is arduous, inefficient, and has high labor costs;

[0006] The work of chronic disease follow-up, elderly physical examination, record maintenance, data backfilling, and telephone notification is highly dependent on manual labor. A single follow-up / physical examination can take 5-10 minutes. During peak periods, all staff need to work overtime. Repeated data entry, manual verification, and manual statistics take up a lot of medical staff’s time, leaving a serious shortage of time for actual diagnosis and treatment services.

[0007] Poor data quality, lack of quality control, and high assessment risks;

[0008] Manual data entry is prone to omissions, errors, logical inconsistencies, and data falsification, resulting in a low rate of standardization. Hundreds to thousands of erroneous data entries need to be manually checked every month, yet it is still difficult to avoid performance deductions, fines, and poor rankings, failing to meet the requirements of public health performance and electronic health record standards.

[0009] The level of intelligence is low, and there is a lack of unified coordination and decision support.

[0010] Existing systems are mostly single-function tool software, lacking large AI models and multi-agent collaborative capabilities. They cannot automatically extract medical records, generate records, perform quality control, or automatically fill in data. Outbound notifications rely on manual dialing, resulting in high turnover rates, high training costs, and insufficient service coverage, making it difficult to meet the needs of refined health management and real-time operational decision-making. Summary of the Invention

[0011] The purpose of this application is to provide a medical and prevention integrated information management method, system and medium based on multi-source heterogeneous data. Through standardization, data mapping and normalization processing, it can realize real-time interoperability across systems, eliminate data barriers and greatly improve work efficiency.

[0012] This application embodiment also provides a medical and prevention information management method based on multi-source heterogeneous data, including: collecting multi-source heterogeneous medical and prevention data in primary medical scenarios, wherein the multi-source heterogeneous medical and prevention data includes electronic health records, outpatient diagnosis and treatment data, chronic disease follow-up data, health check-up data, laboratory test data, medical insurance settlement data and hardware monitoring data;

[0013] The multi-source heterogeneous medical and prevention data is standardized, cleaned, converted in format, and mapped to obtain preprocessed data. Medical and prevention data standard data is constructed based on the preprocessed data. The medical and prevention data standard data is then aligned and fused to obtain fused data.

[0014] Based on the medical big data model and medical knowledge graph, intelligent parsing, rule verification and logical quality control analysis are performed on the fused data to obtain analysis results. Abnormal data is identified based on the analysis results, and the abnormal data is corrected to obtain normal data.

[0015] Based on medical and preventive business rules and performance evaluation indicators, normal data is processed for tasks such as chronic disease follow-up, health check-up, file maintenance, and outbound call notification, including task scheduling, information collection, record generation, and data backfilling, to obtain follow-up data.

[0016] The follow-up data is processed using a multi-agent collaborative algorithm for data governance, business execution, decision support, and operational analysis, resulting in standardized medical and preventive business outcomes, health assessment reports, and performance statistics.

[0017] Optionally, in the medical and preventive information management method based on multi-source heterogeneous data described in the embodiments of this application, the standardization cleaning, format conversion, and data mapping of multi-source heterogeneous medical and preventive data includes:

[0018] Resident information is obtained from electronic health records. Name, ID number, contact information and address are obtained from resident information. A unique resident identifier index is established. The format of name, ID number, contact information and address are checked and normalized to obtain normalized data.

[0019] Based on multi-source heterogeneous medical and preventive data, obtain vital sign data, medication data, lifestyle data, and examination and test data;

[0020] The vital signs data, medication data, lifestyle data, and examination and test data are subjected to value range verification, logical verification, and missing data completion processing to obtain the completed data;

[0021] Normalized data and complete data are mapped to the medical and prevention data model in a unified manner.

[0022] Optionally, in the medical and preventive information management method based on multi-source heterogeneous data described in the embodiments of this application, the step of intelligently parsing, rule-verifying, and logically controlling the fused data based on a large medical model and medical knowledge graph includes:

[0023] NLP algorithms are used to extract text and perform semantic parsing on outpatient medical records, follow-up records, and physical examination reports to obtain semantic information.

[0024] Based on the guidelines for chronic disease management and public health standards, the semantic information of outpatient medical records, follow-up records, and physical examination reports is analyzed for follow-up completeness, physical examination compliance, and record standardization to obtain quality control results.

[0025] Based on the set quality control conditions, the quality control results are analyzed to extract abnormal signs, contradictory data, and out-of-range values, and early warning information is generated.

[0026] Based on the early warning information, corrective actions can be taken to correct abnormal signs, contradictory data, and out-of-range values.

[0027] Optionally, in the medical and preventive integrated information management method based on multi-source heterogeneous data described in the embodiments of this application, the step of automatically executing task scheduling, information collection, record generation, and data backfilling according to medical and preventive business rules and performance evaluation indicators includes:

[0028] Obtain follow-up status and treatment logs, and intelligently generate follow-up tasks according to quarterly follow-up requirements;

[0029] According to the workflow robot, the diagnosis, medication, examination and test data are automatically collected based on the follow-up task, and the follow-up form and physical examination form are generated.

[0030] Obtain the generation record information of the follow-up form and physical examination form, perform full quality control processing on the generated record information, and obtain full quality control information;

[0031] Based on the full quality control information, the generated record information will be automatically backfilled into the primary healthcare information system to obtain follow-up data.

[0032] Optionally, in the medical and preventive information management method based on multi-source heterogeneous data described in the embodiments of this application, the step of performing data governance, business execution, decision support, and operational analysis on follow-up data through a multi-agent collaborative algorithm includes:

[0033] The multi-agent system includes a data agent, a business agent, and a decision agent.

[0034] Data fusion, quality control, updating, and archiving are performed based on data intelligence agents;

[0035] Follow-up visits, physical examinations, record maintenance, AI outbound calls, and health education are conducted based on business intelligence agents.

[0036] The decision-making agent uses fused data to conduct health risk assessments, referral decisions, medication recommendations, and generate performance dashboards.

[0037] Optionally, the medical and preventive information management method based on multi-source heterogeneous data described in the embodiments of this application further includes:

[0038] AI-powered outbound calls and SMS messages are used to send physical examination notifications, follow-up reminders, record information verification, and health education.

[0039] Connect to smart IoT devices, including portable blood pressure monitors, blood glucose meters, height and weight scales, and waist circumference measuring tapes;

[0040] The data is collected and automatically entered into the physical examination data in real time using smart IoT devices;

[0041] Based on the entered data, a chronic disease roster, physical examination progress, standardized management rate, and compliance rate are dynamically generated to obtain visualized statistical results.

[0042] Secondly, this application provides a medical and prevention information management system based on multi-source heterogeneous data. The system includes a memory and a processor. The memory includes a program for a medical and prevention information management method based on multi-source heterogeneous data. When the program for the medical and prevention information management method based on multi-source heterogeneous data is executed by the processor, it performs the following steps: collecting multi-source heterogeneous medical and prevention data in a primary healthcare setting. The multi-source heterogeneous medical and prevention data includes electronic health records, outpatient treatment data, chronic disease follow-up data, health check-up data, laboratory test data, medical insurance settlement data, and hardware monitoring data.

[0043] The multi-source heterogeneous medical and prevention data is standardized, cleaned, converted in format, and mapped to obtain preprocessed data. Medical and prevention data standard data is constructed based on the preprocessed data. The medical and prevention data standard data is then aligned and fused to obtain fused data.

[0044] Based on the medical big data model and medical knowledge graph, intelligent parsing, rule verification and logical quality control analysis are performed on the fused data to obtain analysis results. Abnormal data is identified based on the analysis results, and the abnormal data is corrected to obtain normal data.

[0045] Based on medical and preventive business rules and performance evaluation indicators, normal data is processed for tasks such as chronic disease follow-up, health check-up, file maintenance, and outbound call notification, including task scheduling, information collection, record generation, and data backfilling, to obtain follow-up data.

[0046] The follow-up data is processed using a multi-agent collaborative algorithm for data governance, business execution, decision support, and operational analysis, resulting in standardized medical and preventive business outcomes, health assessment reports, and performance statistics.

[0047] Optionally, in the medical and preventive integrated information management system based on multi-source heterogeneous data described in this application embodiment, the standardization cleaning, format conversion, and data mapping of multi-source heterogeneous medical and preventive data includes:

[0048] Resident information is obtained from electronic health records. Name, ID number, contact information and address are obtained from resident information. A unique resident identifier index is established. The format of name, ID number, contact information and address are checked and normalized to obtain normalized data.

[0049] Based on multi-source heterogeneous medical and preventive data, obtain vital sign data, medication data, lifestyle data, and examination and test data;

[0050] The vital signs data, medication data, lifestyle data, and examination and test data are subjected to value range verification, logical verification, and missing data completion processing to obtain the completed data;

[0051] Normalized data and complete data are mapped to the medical and prevention data model in a unified manner.

[0052] Optionally, in the medical and preventive integrated information management system based on multi-source heterogeneous data described in the embodiments of this application, the intelligent analysis, rule verification, and logical quality control of the integrated data based on the medical big data model and medical knowledge graph includes:

[0053] NLP algorithms are used to extract text and perform semantic parsing on outpatient medical records, follow-up records, and physical examination reports to obtain semantic information.

[0054] Based on the guidelines for chronic disease management and public health standards, the semantic information of outpatient medical records, follow-up records, and physical examination reports is analyzed for follow-up completeness, physical examination compliance, and record standardization to obtain quality control results.

[0055] Based on the set quality control conditions, the quality control results are analyzed to extract abnormal signs, contradictory data, and out-of-range values, and early warning information is generated.

[0056] Based on the early warning information, corrective actions can be taken to correct abnormal signs, contradictory data, and out-of-range values.

[0057] Thirdly, embodiments of this application also provide a computer-readable storage medium, which includes a medical and prevention information management method program based on multi-source heterogeneous data. When the medical and prevention information management method program based on multi-source heterogeneous data is executed by a processor, it implements the steps of the medical and prevention information management method based on multi-source heterogeneous data as described in any of the above claims.

[0058] As can be seen from the above, the medical and preventive information management method, system, and medium based on multi-source heterogeneous data provided in this application collect multi-source heterogeneous medical and preventive data in primary healthcare scenarios. This multi-source heterogeneous medical and preventive data includes electronic health records, outpatient treatment data, chronic disease follow-up data, health check-up data, laboratory test data, medical insurance settlement data, and hardware monitoring data. The multi-source heterogeneous medical and preventive data undergoes standardized cleaning, format conversion, and data mapping to obtain preprocessed data. Based on the preprocessed data, standard medical and preventive data is constructed. This standard medical and preventive data is then aligned and fused to obtain fused data. Finally, the fused data is intelligently analyzed and standardized based on a large medical model and medical knowledge graph. The system performs verification and logical quality control analysis to obtain analysis results. Based on these results, it identifies and corrects abnormal data to obtain normal data. Then, based on medical and preventative business rules and performance evaluation indicators, it processes the normal data through task scheduling, information collection, record generation, and data backfilling for chronic disease follow-up, health checkups, record maintenance, and outbound call notifications, resulting in follow-up data. Finally, it uses a multi-agent collaborative algorithm to perform data governance, business execution, decision support, and operational analysis on the follow-up data, outputting standardized medical and preventative business results, health assessment reports, and performance statistics. Through standardization, data mapping, and normalization, it achieves real-time interoperability across systems, eliminates data silos, and significantly improves work efficiency. Attached Figure Description

[0059] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 A flowchart illustrating the medical and preventive information management method based on multi-source heterogeneous data provided in this application embodiment;

[0061] Figure 2 The flowchart of the multi-source heterogeneous medical and preventive data preprocessing method for the medical and preventive integrated information management method based on multi-source heterogeneous data provided in the embodiments of this application is as follows:

[0062] Figure 3 The flowchart of the fusion data verification and quality control process of the medical and prevention fusion information management method based on multi-source heterogeneous data provided in the embodiments of this application is shown. Detailed Implementation

[0063] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0064] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0065] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a method for managing medical and preventive information based on multi-source heterogeneous data, as described in some embodiments of this application. This method is used in a terminal device and includes the following steps:

[0066] S101 collects multi-source heterogeneous medical and preventive data in primary healthcare scenarios. The multi-source heterogeneous medical and preventive data includes electronic health records, outpatient treatment data, chronic disease follow-up data, health check-up data, laboratory test data, medical insurance settlement data, and hardware monitoring data.

[0067] S102, standardize, clean, convert, and map multi-source heterogeneous medical and prevention data to obtain preprocessed data, construct standard medical and prevention data based on the preprocessed data, and perform data alignment and fusion processing on the standard medical and prevention data to obtain fused data;

[0068] S103, based on the medical big data model and medical knowledge graph, performs intelligent parsing, rule verification and logical quality control analysis on the fused data to obtain analysis results, identifies abnormal data based on the analysis results, corrects the abnormal data and obtains normal data;

[0069] S104, based on medical and preventive business rules and performance evaluation indicators, performs task scheduling, information collection, record generation and data backfilling on normal data for chronic disease follow-up, health check-up, file maintenance and outbound call notification, to obtain follow-up data;

[0070] S105 performs data governance, business execution, decision support, and operational analysis on follow-up data based on a multi-agent collaborative algorithm, and outputs standardized medical and prevention business results, health assessment reports, and performance statistics.

[0071] It should be noted that multi-source heterogeneous medical and preventive data collection is the data entry point and fundamental link in the entire integrated medical and preventive management process. Its purpose is to comprehensively, completely, and without omission acquire all-dimensional business data generated by primary healthcare and public health services. Due to the common problems of fragmented information systems, diverse data sources, and inconsistent formats in primary healthcare, data that was originally stored in isolation across different systems, devices, and media is uniformly aggregated through interface integration, RPA robot collection, direct connection to smart hardware, database synchronization, and OCR recognition. The collection scope covers residents' health information throughout their entire life cycle and data from the entire medical and preventive business chain, ensuring that subsequent data governance, intelligent analysis, and business automation have a sufficient and accurate data foundation, avoiding deviations in business execution, quality control failures, or decision-making errors due to incomplete data.

[0072] Data cleaning, transformation, mapping, alignment, and fusion address the core challenges of data silos, inconsistent standards, and lack of data sharing. First, raw data undergoes standardized cleaning, including missing value imputation, duplicate data removal, error format correction, and illegal value filtering. Next, format conversion unifies structured, semi-structured, and unstructured data into a computable and comparable format. Then, master data mapping is performed using unique resident identifiers (such as ID numbers) as the primary key, establishing a unified data standard across systems. Based on this, a medical and preventative standard data system conforming to national public health regulations is constructed, and data is aligned by time, institution, personnel, and business processes. Ultimately, deep fusion of multi-source heterogeneous data is achieved, forming a unified data view with one file per person, a single code, and full traceability, providing high-quality, standardized input for subsequent intelligent quality control.

[0073] Intelligent parsing, rule verification, logical quality control, and anomaly correction are key checkpoints to ensure data authenticity, compliance, and usability for performance evaluation. Through medical big data models and medical knowledge graphs, NLP extraction, semantic understanding, and structural transformation are performed on medical record texts, follow-up records, and physical examination information to achieve intelligent interpretation of unstructured medical information. Then, based on national basic public health service standards, chronic disease management guidelines, health record standards, and performance evaluation rules, the system performs integrity verification, value range verification, logical verification, and consistency verification on the data. The system automatically identifies abnormal data such as omissions, errors, exceeding scope, logical contradictions, and non-standard records. Data that can be automatically corrected is batch corrected and automatically completed; data that cannot be corrected is marked, alerted, and pushed to manual processing. Ultimately, the system outputs accurate, standardized, and directly usable data for business execution, preventing data errors from the source that could lead to performance deductions, fines, or lower rankings.

[0074] Task scheduling, information collection, record generation, and data backfilling are the automated implementation steps in healthcare and public health operations, moving from data to execution. Based on preset public health business rules, follow-up cycles, physical examination plans, record update requirements, and performance evaluation indicators, the system automatically schedules tasks intelligently, generates a to-do list, and assigns tasks for execution. Through RPA robots, it automatically collects business information from diagnosis, testing, medication, and hardware equipment, eliminating the need for manual data entry. It then automatically generates compliant documents such as follow-up forms, physical examination forms, and record update records according to standard form formats. Finally, after quality control approval, it automatically backfills the data into the HIS / public health system without altering existing system data or destroying historical records. This step can reduce the original 5-10 minutes / person manual operation to less than 1 minute, significantly reducing the workload, improving business efficiency and data standardization, and forming standardized follow-up / physical examination / record data that is assessable, verifiable, and traceable.

[0075] Multi-agent collaboration and result output constitute the closed-loop output and value enhancement link of the entire method. Through multi-agent collaboration algorithms, data governance agents, business execution agents, decision support agents, and operational analysis agents are scheduled to work collaboratively.

[0076] The data governance intelligent agent continuously completes data updates, archiving, and verification;

[0077] The business execution intelligent agent completes follow-up visits, physical examinations, AI outbound calls, and health education;

[0078] The decision-making support agent provides health risk assessment, medication advice, and referral decisions;

[0079] The operational analytics agent generates performance dashboards, compliance rates, progress statistics, and dynamic rosters. Ultimately, the system outputs standardized medical and preventative service results, personalized health assessment reports, and visualized performance statistics, achieving a closed-loop process encompassing data, business, management, and decision-making. This significantly improves the integrated medical and preventative service capabilities, management efficiency, and performance compliance levels at the grassroots level.

[0080] like Figure 2 As shown in the embodiment of the present invention, the standardization cleaning, format conversion, and data mapping of multi-source heterogeneous medical and prevention data include:

[0081] Resident information is obtained from electronic health records. Name, ID number, contact information and address are obtained from resident information. A unique resident identifier index is established. The format of name, ID number, contact information and address are checked and normalized to obtain normalized data.

[0082] Specifically, using electronic health records as the authoritative benchmark, the system extracts residents' core identity information, including name, ID number, contact number, residential address, and registered address, as the sole basis for data matching throughout the entire process. The system uses the resident's ID number as the core primary key to construct a globally unique identifier index, achieving unique data binding across systems, businesses, and devices, thus avoiding problems such as multiple records for one person, multiple people per record, and duplicate record creation.

[0083] Based on this, the information undergoes format validation and normalization:

[0084] The names are standardized by unifying simplified and traditional characters, full-width and half-width characters, removing special symbols, and eliminating spaces.

[0085] Perform digit verification, format verification, and legality verification on the ID number, and eliminate erroneous numbers;

[0086] The system distinguishes between landline and mobile phone numbers, completes area codes, and removes redundant characters from contact phone numbers.

[0087] Addresses are structured at the provincial, municipal, and district levels, standardized with local place names, and missing fields are filled in.

[0088] Based on multi-source heterogeneous medical and preventive data, obtain vital sign data, medication data, lifestyle data, and examination and test data;

[0089] Specifically, extract the four types of operational data that are most core, frequently used, and strongly correlated with performance evaluation in primary public health:

[0090] Vital data: blood pressure, blood sugar, heart rate, height, weight, BMI, waist circumference, dorsalis pedis artery pulsation, etc.

[0091] Medication data: drug name, dosage, frequency of use, medication adherence, adverse reactions, etc.

[0092] Lifestyle data: smoking, drinking, exercise, salt intake, dietary habits, mental state, etc.

[0093] Test results include: complete blood count, urinalysis, biochemistry, glycated hemoglobin, electrocardiogram, and ultrasound results.

[0094] The vital signs data, medication data, lifestyle data, and examination and test data are subjected to value range verification, logical verification, and missing data completion processing to obtain the completed data;

[0095] Specifically, the extracted business data undergoes triple intelligent verification to ensure its authenticity, reasonableness, and usability for medical and preventive healthcare operations:

[0096] Value range verification judges whether the value is within a reasonable range according to medical common sense and public health standards, for example:

[0097] Normal range for systolic blood pressure and normal range for diastolic blood pressure;

[0098] Reasonable ranges for fasting blood glucose and glycated hemoglobin;

[0099] Age, weight, height, and gender matching verification. Automatically marks out-of-range, extreme values, and obviously erroneous data.

[0100] Logical checks examine whether there are medical inconsistencies or business conflicts between the data, for example:

[0101] Diagnosed with hypertension but no blood pressure data, diabetes but no blood glucose record, mismatch between age and type of physical examination, and mismatch between medication and diagnosis.

[0102] Missing data items that are reasonably missing and can be automatically filled in according to rules:

[0103] Historical data inheritance completion / same visit related data completion / standard default value completion and logical deduction completion.

[0104] The final result is complete, reasonable, consistent, and error-free high-quality business data, which is called data completion.

[0105] Normalized data and complete data are mapped to the medical and prevention data model in a unified manner.

[0106] It should be noted that the above-mentioned normalized basic data and supplementary business data are uniformly mapped to the pre-constructed medical and prevention integration standard data model in accordance with the national basic public health service standards, primary medical and prevention data standards, and electronic health record standards.

[0107] This data model has the following characteristics:

[0108] Unified fields, unified standards, unified coding, and unified structure;

[0109] Compatible with HIS, public health systems, physical examination systems, medical insurance systems, and smart hardware data;

[0110] It supports the entire process of subsequent data alignment, fusion, quality control, analysis, and backfilling;

[0111] It meets the requirements for performance appraisal, report statistics, health assessment, and regulatory reporting.

[0112] like Figure 3 As shown in the embodiment of the present invention, intelligent parsing, rule verification, and logical quality control of fused data based on a large medical model and a medical knowledge graph include:

[0113] NLP algorithms are used to extract text and perform semantic parsing on outpatient medical records, follow-up records, and physical examination reports to obtain semantic information.

[0114] Specifically, a large amount of information in primary healthcare exists in free text, semi-structured, and unstructured forms (such as handwritten medical records, colloquial descriptions, physical examination summaries, and follow-up notes), which cannot be directly recognized and used by the system. Relying on a large-scale medical model and medical NLP algorithms, this approach analyzes text data such as outpatient medical records, chronic disease follow-up records, health check-up reports, and laboratory test results.

[0115] Key information extraction: Automatically extracts core elements such as symptoms, signs, diagnosis, medication, lifestyle, test results, and doctor's advice;

[0116] Semantic understanding: Normalization of medical terminology, merging of synonyms, and understanding of context;

[0117] Structured transformation: converting fragmented text into structured semantic information that the system can recognize, compute, and compare.

[0118] This process enables machine-readable unstructured medical texts, providing a computable data foundation for subsequent rule-based quality control and logical judgment.

[0119] Based on the guidelines for chronic disease management and public health standards, the semantic information of outpatient medical records, follow-up records, and physical examination reports is analyzed for follow-up completeness, physical examination compliance, and record standardization to obtain quality control results.

[0120] Specifically, based on the National Basic Public Health Service Standards, Guidelines for the Management of Chronic Diseases such as Hypertension / Diabetes, Health Record Standards, and Performance Evaluation Rules, mandatory compliance checks are conducted on semantic information across three dimensions:

[0121] The follow-up integrity quality control check includes whether the number of follow-ups, follow-up period, required fields, vital signs data, medication records, and lifestyle guidance are complete and whether they meet the requirements for quarterly follow-ups and annual assessments.

[0122] The compliance and quality control of physical examinations include checking whether the physical examination items for the elderly and children are complete, whether the auxiliary examinations are complete, whether the physical examination process is in accordance with regulations, and whether the results can be used for public health statistics.

[0123] The standardization and quality control of resident health records include checking the completeness of fields, logical consistency, format correctness, and timeliness of updates to ensure that the coverage rate of standardized electronic records is met.

[0124] The system compares each item, automatically scores it, and determines whether it is qualified. Finally, it outputs quality control results information including completeness, compliance, and standardization, providing a basis for anomaly identification.

[0125] Based on the set quality control conditions, the quality control results are analyzed to extract abnormal signs, contradictory data, and out-of-range values, and early warning information is generated.

[0126] Specifically, based on preset medical rules, business rules, numerical ranges, and logical constraints, the quality control results are analyzed in depth to automatically identify three types of typical problematic data:

[0127] Abnormal physical signs such as blood pressure, blood sugar, heart rate, BMI, and waist circumference exceeding the medically reasonable range indicate health risks or data entry errors.

[0128] Logical errors include contradictions between diagnostic and physical signs data, contradictions between medication and disease, contradictions between age and project data, and conflicts between historical and current data.

[0129] Values ​​outside the range (e.g., blood pressure > 250 mmHg, blood glucose > 30 mmol / L, age > 120 years, etc., which are obvious input errors).

[0130] The system automatically marks, locates, and labels the aforementioned issues, generating traceable, viewable, and processable early warning information to enable early detection and alerts for data problems.

[0131] Based on the early warning information, corrective actions can be taken to correct abnormal signs, contradictory data, and out-of-range values.

[0132] It should be noted that, based on the early warning information, and following medical common sense, historical data, business rules, and public health logic, intelligent repair is performed on data that can be automatically corrected without human intervention.

[0133] Obvious input errors, such as misplaced values, incorrect symbols, incorrect units, and incorrect decimal points, will be automatically corrected according to the rules.

[0134] Logical conflict: Automatically corrects contradictory items based on diagnosis, age, and gender;

[0135] Missing derivable items: automatically completed based on historical records or data from the same period;

[0136] Incorrect format: Automatically standardized to the standard format.

[0137] The corrected data is re-entered into the quality control process for verification to ensure that it is qualified and can be directly used for business execution. For complex issues that cannot be automatically corrected (such as missing information or unclear diagnosis), warnings are issued and manual review is pushed to ensure that the original data is not modified at will, thus guaranteeing data security and traceability.

[0138] According to an embodiment of the present invention, task scheduling, information collection, record generation, and data backfilling are automatically executed based on medical and preventive business rules and performance evaluation indicators, including:

[0139] Obtain follow-up status and treatment logs, and intelligently generate follow-up tasks according to quarterly follow-up requirements;

[0140] According to the workflow robot, the diagnosis, medication, examination and test data are automatically collected based on the follow-up task, and the follow-up form and physical examination form are generated.

[0141] Specifically, the RPA workflow robot is activated, and according to the requirements of follow-up and physical examination tasks, it automatically retrieves data from the HIS system, outpatient medical records, prescription data, laboratory testing systems, and smart hardware devices without manual operation, system switching, or copying and pasting.

[0142] Medical data: chief complaint, diagnosis, physical signs, and treatment recommendations;

[0143] Medication data: drug name, usage, dosage, and compliance;

[0144] Test results: blood glucose, blood pressure, biochemistry, electrocardiogram, and ultrasound results;

[0145] Lifestyle data: smoking, drinking, exercise, and salt intake.

[0146] The robot automatically fills in, calculates, and selects according to the national public health standard template, generating standardized, complete, and directly applicable chronic disease follow-up forms and health check-up forms with one click. The entire process does not change the doctor's original writing habits and does not increase the workload of medical staff.

[0147] Obtain the generation record information of the follow-up form and physical examination form, perform full quality control processing on the generated record information, and obtain full quality control information;

[0148] Based on the full quality control information, the generated record information will be automatically backfilled into the primary healthcare information system to obtain follow-up data.

[0149] It should be noted that, based on the full volume of quality control information, follow-up records and physical examination records that pass quality control are automatically, accurately, and silently backfilled into primary healthcare information systems (such as public health systems, HIS systems, and health record systems) through a secure interface. This process does not modify existing historical data, damage the system structure, generate redundant data, or leave any traces of operation. After backfilling, standardized follow-up data that is searchable, statistically verifiable, assessable, and monitorable is generated, serving as the official basis for performance evaluation, health management, and public health reporting, thus achieving a complete closed loop for medical and preventive services from data to execution to archiving.

[0150] According to embodiments of the present invention, a multi-agent collaborative algorithm is used to perform data governance, business execution, decision support, and operational analysis on follow-up data, including:

[0151] Multi-agent systems include data agents, business agents, and decision agents;

[0152] Data fusion, quality control, updating, and archiving are performed based on data intelligence agents;

[0153] Follow-up visits, physical examinations, record maintenance, AI outbound calls, and health education are conducted based on business intelligence agents.

[0154] The decision-making agent uses fused data to conduct health risk assessments, referral decisions, medication recommendations, and generate performance dashboards.

[0155] It should be noted that the data intelligence agent, as the core of the system's data processing, continuously manages follow-up data, physical examination data, and resident record data throughout their entire lifecycle.

[0156] Data fusion: Align and integrate multi-source business data to form a complete health profile of residents;

[0157] Intelligent quality control: Continuous inspections are conducted according to public health rules to dynamically detect and correct abnormal and contradictory data;

[0158] Real-time updates: Synchronize the latest data from diagnosis, follow-up, physical examinations, and hardware monitoring to keep the data up-to-date;

[0159] Secure archiving: Standardized archiving of completed business transactions ensures traceability, auditability, and verifiability. Through data intelligence, it ensures that system data is always accurate, complete, standardized, and usable for performance evaluation.

[0160] The decision-making intelligence agent is the system's AI brain, providing doctors and administrators with real-time, scientific, and actionable decision support based on comprehensive data fusion, large-scale medical models, and knowledge graphs.

[0161] Health risk assessment: Risk classification and trend prediction for hypertension, diabetes, and the elderly;

[0162] Intelligent referral assessment: Automatically determines whether a referral is needed and the referral priority based on clinical guidelines;

[0163] Personalized medication recommendations: Provide adjustment plans and monitoring requirements based on physical signs, medical history, and adherence;

[0164] Performance dashboard generation: Automatically calculates follow-up completion rate, physical examination rate, standardized management rate, and target achievement rate, generating a visual dashboard, dynamic roster, and performance reports. The intelligent decision-making system makes management quantifiable, services precise, and decisions data-driven, significantly improving the level of primary healthcare management.

[0165] According to an embodiment of the present invention, it further includes:

[0166] AI-powered outbound calls and SMS messages are used to send physical examination notifications, follow-up reminders, record information verification, and health education.

[0167] Specifically, relying on medical big data models and intelligent voice interaction capabilities, the system uses a combination of AI-powered outbound calls and simultaneous SMS notifications to reach key groups such as the elderly, patients with chronic diseases, and those undergoing physical examinations, providing comprehensive public health services across all scenarios.

[0168] Physical examination notification: Automatically calls groups of people who should be examined, such as the elderly, those with chronic diseases, and children, to inform them of the examination time, location, and precautions, thereby improving the attendance rate.

[0169] Follow-up reminders: Automatic reminders for individuals whose follow-up appointments are about to expire or have already exceeded their deadlines, ensuring that follow-up tasks are completed on time.

[0170] Document information verification: For missing or non-standard information such as addresses, telephone numbers, and contact persons, the system automatically verifies and completes the information through outbound calls, thereby improving the standardization rate of documents.

[0171] Health education: Based on disease types such as hypertension and diabetes, personalized information on diet, exercise, medication, and monitoring is automatically pushed to enhance health intervention. The entire process requires no manual dialing or scripted responses, is stable, efficient, traceable, and statistically verifiable, significantly reducing the workload of medical staff on telephone and improving the coverage and completion rate of public health services.

[0172] It connects to smart IoT devices, including portable blood pressure monitors, blood glucose meters, height and weight scales, and waist circumference measuring tapes.

[0173] The data is collected and automatically entered into the physical examination data in real time using smart IoT devices;

[0174] Specifically, during physical examinations or follow-ups, medical staff use smart IoT devices to measure data, which is then automatically uploaded to the system in real time. This eliminates the need for manual input of values, selection of units, or format verification. The system automatically receives vital signs data such as blood pressure, blood sugar, height, weight, BMI, and waist circumference uploaded from the devices, automatically performing format conversion, value verification, unit standardization, and anomaly identification, resulting in standardized, accurate data that can be directly used for business operations. This process involves zero manual input, zero errors, and high efficiency, significantly shortening the time required for individual physical examinations / follow-ups and improving the authenticity and accuracy of the data.

[0175] Based on the entered data, a chronic disease roster, physical examination progress, standardized management rate, and compliance rate are dynamically generated to obtain visualized statistical results.

[0176] It should be noted that, based on real-time data collected by smart devices, and combined with comprehensive business data such as follow-up visits, physical examinations, and records, core public health management indicators are automatically calculated, dynamically updated, and displayed in real time.

[0177] Dynamic chronic disease roster: Real-time display of the number of people under management for hypertension and diabetes, their classification and grading, follow-up status, and physical examination results.

[0178] Physical examination progress statistics: Real-time display of the number of people who should be examined, the number of people who have been examined, the number of people who have not been examined, the completion rate, and the number of people who have exceeded the time limit.

[0179] Standardized Management Rate: Automatically calculates indicators such as standardized management, regular follow-up, and complete records according to public health assessment rules.

[0180] Chronic disease control compliance rate: Automatically tracks the number of people who have achieved blood pressure and blood sugar targets, the compliance rate, and trend changes. The final result is visualized, filterable, exportable, and usable for performance evaluation, providing real-time data support for primary public health management, performance evaluation, and work scheduling, achieving visualized management, data-driven decision-making, and transparent performance evaluation.

[0181] Secondly, this application provides a medical and prevention information management system based on multi-source heterogeneous data. The system includes a memory and a processor. The memory includes a program for a medical and prevention information management method based on multi-source heterogeneous data. When the program for the medical and prevention information management method based on multi-source heterogeneous data is executed by the processor, it performs the following steps: collecting multi-source heterogeneous medical and prevention data in a primary healthcare setting. The multi-source heterogeneous medical and prevention data includes electronic health records, outpatient treatment data, chronic disease follow-up data, health check-up data, laboratory test data, medical insurance settlement data, and hardware monitoring data.

[0182] Standardized cleaning, format conversion and data mapping of multi-source heterogeneous medical and prevention data are performed to obtain preprocessed data. Based on the preprocessed data, standard medical and prevention data is constructed. The standard medical and prevention data is then aligned and fused to obtain fused data.

[0183] Based on the medical big data model and medical knowledge graph, intelligent parsing, rule verification and logical quality control analysis are performed on the fused data to obtain analysis results. Abnormal data is identified based on the analysis results, and the abnormal data is corrected to obtain normal data.

[0184] Based on medical and preventive business rules and performance evaluation indicators, normal data is processed for tasks such as chronic disease follow-up, health check-up, file maintenance, and outbound call notification, including task scheduling, information collection, record generation, and data backfilling, to obtain follow-up data.

[0185] The follow-up data is processed using a multi-agent collaborative algorithm for data governance, business execution, decision support, and operational analysis, resulting in standardized medical and preventive business outcomes, health assessment reports, and performance statistics.

[0186] According to embodiments of the present invention, standardization cleaning, format conversion, and data mapping of multi-source heterogeneous medical and preventive data include:

[0187] Resident information is obtained from electronic health records. Name, ID number, contact information and address are obtained from resident information. A unique resident identifier index is established. The format of name, ID number, contact information and address are checked and normalized to obtain normalized data.

[0188] Based on multi-source heterogeneous medical and preventive data, obtain vital sign data, medication data, lifestyle data, and examination and test data;

[0189] The vital signs data, medication data, lifestyle data, and examination and test data are subjected to value range verification, logical verification, and missing data completion processing to obtain the completed data;

[0190] Normalized data and complete data are mapped to the medical and prevention data model in a unified manner.

[0191] According to embodiments of the present invention, intelligent analysis, rule verification, and logical quality control of fused data based on a large medical model and a medical knowledge graph are performed, including:

[0192] NLP algorithms are used to extract text and perform semantic parsing on outpatient medical records, follow-up records, and physical examination reports to obtain semantic information.

[0193] Based on the guidelines for chronic disease management and public health standards, the semantic information of outpatient medical records, follow-up records, and physical examination reports is analyzed for follow-up completeness, physical examination compliance, and record standardization to obtain quality control results.

[0194] Based on the set quality control conditions, the quality control results are analyzed to extract abnormal signs, contradictory data, and out-of-range values, and early warning information is generated.

[0195] Based on the early warning information, corrective actions can be taken to correct abnormal signs, contradictory data, and out-of-range values.

[0196] In summary, the present invention has the following significant beneficial effects:

[0197] Completely break down data silos and achieve integrated integration of medical and preventive data; unify the collection of multi-source heterogeneous data such as electronic health records, outpatient treatment, chronic disease follow-up, physical examination, laboratory tests, medical insurance, and smart hardware, and achieve real-time interoperability across systems through standardization, data mapping, and normalization processing, eliminate data barriers, and build a complete health profile of residents.

[0198] Full-process automation significantly improves work efficiency and frees up manpower.

[0199] The system automatically assesses follow-up status, collects treatment data, generates records, performs automatic data quality control, and automatically backfills data. The time for a single follow-up / physical examination is reduced from 5-10 minutes to less than 1 minute, reducing manual operations by more than 90%. This can save hundreds of thousands of minutes of manpower per year, allowing medical staff to focus on providing high-value medical services.

[0200] Data quality has been comprehensively improved, ensuring that assessment standards are met and avoiding deductions and fines.

[0201] Based on medical big data models, knowledge graphs, and NLP, full-field automatic quality control is achieved, with real-time early warning and automatic correction of abnormal data. The coverage rate of standardized electronic health records can be increased from about 50% to over 90%. The data is traceable and monitorable, and can stably meet the requirements of public health performance evaluation.

[0202] AI empowers the business closed loop, supporting refined medical and prevention management;

[0203] It provides integrated capabilities such as chronic disease management, physical examinations for the elderly and children, record quality control, AI outbound calling, health education, dynamic roster, and data dashboard. It supports both central physical examinations and mobile physical examinations in rural areas, adapts to different medical and nursing resource configurations, and significantly improves service coverage and management accuracy.

[0204] Multi-agent collaboration reduces costs and enhances residents' sense of gain. Natural language interaction enables automated processing with a single sentence, while multi-agent collaboration completes data governance, business execution, and decision support. AI outbound calls replace human telephone calls, resulting in lower costs, wider coverage, and more stable responses, effectively improving the rate of physical examination visits, follow-up completion rates, and resident satisfaction.

[0205] A third aspect of the present invention provides a computer-readable storage medium, the storage medium including a medical and prevention information management method program based on multi-source heterogeneous data, wherein when the medical and prevention information management method program based on multi-source heterogeneous data is executed by a processor, it implements the steps of the medical and prevention information management method based on multi-source heterogeneous data as described above.

[0206] This invention discloses a method, system, and medium for managing medical and preventive information based on multi-source heterogeneous data. It collects multi-source heterogeneous medical and preventive data from primary healthcare settings, including electronic health records, outpatient treatment data, chronic disease follow-up data, health check-up data, laboratory test data, medical insurance settlement data, and hardware monitoring data. The multi-source heterogeneous medical and preventive data undergoes standardized cleaning, format conversion, and data mapping to obtain preprocessed data. Based on the preprocessed data, standard medical and preventive data is constructed. This standard data is then aligned and fused to obtain fused data. Finally, the fused data is intelligently analyzed and validated using a large-scale medical model and medical knowledge graph. Logical quality control analysis yields results, which are used to identify and correct abnormal data, resulting in normal data. Based on medical and preventative business rules and performance evaluation indicators, normal data is processed through task scheduling, information collection, record generation, and data backfilling for chronic disease follow-up, health checkups, record maintenance, and outbound call notifications, resulting in follow-up data. A multi-agent collaborative algorithm is used to perform data governance, business execution, decision support, and operational analysis on the follow-up data, outputting standardized medical and preventative business results, health assessment reports, and performance statistics. Through standardization, data mapping, and normalization, real-time cross-system communication is achieved, eliminating data silos and significantly improving work efficiency.

[0207] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0208] 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. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0209] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0210] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0211] Alternatively, if the integrated units of the present invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for integrated medical and preventive information management based on multi-source heterogeneous data, characterized in that, include: Collect multi-source heterogeneous medical and preventive data in primary healthcare settings. The multi-source heterogeneous medical and preventive data includes electronic health records, outpatient treatment data, chronic disease follow-up data, health check-up data, laboratory test data, medical insurance settlement data, and hardware monitoring data. The multi-source heterogeneous medical and prevention data is standardized, cleaned, converted in format, and mapped to obtain preprocessed data. Medical and prevention data standard data is constructed based on the preprocessed data. The medical and prevention data standard data is then aligned and fused to obtain fused data. Based on a medical big data model and medical knowledge graph, intelligent parsing, rule verification and logical quality control analysis are performed on the fused data to obtain analysis results. Based on the analysis results, abnormal data is identified, and the abnormal data is corrected to obtain normal data. Based on medical and preventive business rules and performance evaluation indicators, normal data is processed for chronic disease follow-up, health check-up, record maintenance, outbound call notification task scheduling, information collection, record generation and data backfilling to obtain follow-up data. The follow-up data is processed using a multi-agent collaborative algorithm for data governance, business execution, decision support, and operational analysis, resulting in standardized medical and preventive business outcomes, health assessment reports, and performance statistics.

2. The medical and preventive information management method based on multi-source heterogeneous data according to claim 1, characterized in that, The standardization, cleaning, format conversion, and data mapping of multi-source heterogeneous medical and prevention data include: Resident information is obtained from electronic health records. Name, ID number, contact information and address are obtained from resident information. A unique resident identifier index is established. The format of name, ID number, contact information and address are checked and normalized to obtain normalized data. Based on multi-source heterogeneous medical and preventive data, obtain vital sign data, medication data, lifestyle data, and examination and test data; The vital signs data, medication data, lifestyle data, and examination and test data are subjected to value range verification, logical verification, and missing data completion processing to obtain the completed data; Normalized data and complete data are mapped to the medical and prevention data model in a unified manner.

3. The medical and preventive information management method based on multi-source heterogeneous data according to claim 2, characterized in that, The intelligent analysis, rule verification, and logical quality control of the fused data based on the medical big data model and medical knowledge graph include: NLP algorithms are used to extract text and perform semantic parsing on outpatient medical records, follow-up records, and physical examination reports to obtain semantic information. Based on the guidelines for chronic disease management and public health standards, the semantic information of outpatient medical records, follow-up records, and physical examination reports was analyzed for follow-up completeness, physical examination compliance, and record standardization to obtain quality control results. Based on the set quality control conditions, the quality control results are analyzed to extract abnormal signs, contradictory data, and out-of-range values, and early warning information is generated. Based on the early warning information, corrective actions can be taken to correct abnormal signs, contradictory data, and out-of-range values.

4. The medical and preventive information management method based on multi-source heterogeneous data according to claim 3, characterized in that, The automatic execution of task scheduling, information collection, record generation, and data backfilling based on medical and preventive business rules and performance evaluation indicators includes: Obtain follow-up status and treatment logs, and intelligently generate follow-up tasks according to quarterly follow-up requirements; According to the workflow robot, the diagnosis, medication, examination and test data are automatically collected based on the follow-up task, and the follow-up form and physical examination form are generated. Obtain the generation record information of follow-up forms and physical examination forms, perform full quality control processing on the generated record information, and obtain full quality control information; Based on the full quality control information, the generated record information will be automatically backfilled into the primary healthcare information system to obtain follow-up data.

5. The medical and preventive information management method based on multi-source heterogeneous data according to claim 4, characterized in that, The method of using a multi-agent collaborative algorithm to perform data governance, business execution, decision support, and operational analysis on follow-up data includes: The multi-agent system includes a data agent, a business agent, and a decision agent. Data fusion, quality control, updating, and archiving are performed based on data intelligence agents; Follow-up visits, physical examinations, record maintenance, AI outbound calls, and health education are conducted based on business intelligence agents. The decision-making agent uses fused data to conduct health risk assessments, referral decisions, medication recommendations, and generate performance dashboards.

6. The medical and preventive information management method based on multi-source heterogeneous data according to claim 5, characterized in that, Also includes: AI-powered outbound calls and SMS messages are used to send physical examination notifications, follow-up reminders, record information verification, and health education. Connect to smart IoT devices, including portable blood pressure monitors, blood glucose meters, height and weight scales, and waist circumference measuring tapes; The data is collected and automatically entered in real time using smart IoT devices to obtain the data entry. Based on the entered data, a chronic disease roster, physical examination progress, standardized management rate, and compliance rate are dynamically generated to obtain visualized statistical results.

7. A medical and preventive information management system based on multi-source heterogeneous data, characterized in that, The system includes a memory and a processor. The memory contains a program for a medical and preventive information management method based on multi-source heterogeneous data. When the program for the medical and preventive information management method based on multi-source heterogeneous data is executed by the processor, it performs the following steps: Collect multi-source heterogeneous medical and preventive data in primary healthcare settings. The multi-source heterogeneous medical and preventive data includes electronic health records, outpatient treatment data, chronic disease follow-up data, health check-up data, laboratory test data, medical insurance settlement data, and hardware monitoring data. The multi-source heterogeneous medical and prevention data is standardized, cleaned, converted in format, and mapped to obtain preprocessed data. Medical and prevention data standard data is constructed based on the preprocessed data. The medical and prevention data standard data is then aligned and fused to obtain fused data. Based on a medical big data model and medical knowledge graph, intelligent parsing, rule verification and logical quality control analysis are performed on the fused data to obtain analysis results. Based on the analysis results, abnormal data is identified, and the abnormal data is corrected to obtain normal data. Based on medical and preventive business rules and performance evaluation indicators, normal data is processed for chronic disease follow-up, health check-up, record maintenance, outbound call notification task scheduling, information collection, record generation and data backfilling to obtain follow-up data. The follow-up data is processed using a multi-agent collaborative algorithm for data governance, business execution, decision support, and operational analysis, resulting in standardized medical and preventive business outcomes, health assessment reports, and performance statistics.

8. The medical and preventive information management system based on multi-source heterogeneous data according to claim 7, characterized in that, The standardization, cleaning, format conversion, and data mapping of multi-source heterogeneous medical and prevention data include: Resident information is obtained from electronic health records. Name, ID number, contact information and address are obtained from resident information. A unique resident identifier index is established. The format of name, ID number, contact information and address are checked and normalized to obtain normalized data. Based on multi-source heterogeneous medical and preventive data, obtain vital sign data, medication data, lifestyle data, and examination and test data; The vital signs data, medication data, lifestyle data, and examination and test data are subjected to value range verification, logical verification, and missing data completion processing to obtain the completed data; Normalized data and complete data are mapped to the medical and prevention data model in a unified manner.

9. The medical and preventive information management system based on multi-source heterogeneous data according to claim 8, characterized in that, The intelligent analysis, rule verification, and logical quality control of the fused data based on the medical big data model and medical knowledge graph include: NLP algorithms are used to extract text and perform semantic parsing on outpatient medical records, follow-up records, and physical examination reports to obtain semantic information. Based on the guidelines for chronic disease management and public health standards, the semantic information of outpatient medical records, follow-up records, and physical examination reports was analyzed for follow-up completeness, physical examination compliance, and record standardization to obtain quality control results. Based on the set quality control conditions, the quality control results are analyzed to extract abnormal signs, contradictory data, and out-of-range values, and early warning information is generated. Based on the early warning information, corrective actions can be taken to correct abnormal signs, contradictory data, and out-of-range values.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a medical and prevention information management method program based on multi-source heterogeneous data. When the medical and prevention information management method program based on multi-source heterogeneous data is executed by a processor, it implements the steps of the medical and prevention information management method based on multi-source heterogeneous data as described in any one of claims 1 to 6.

Images