A medical data processing method and system based on artificial intelligence NLP technology

Abstract

The present invention discloses a medical data processing method and system based on artificial intelligence NLP technology, and relates to the field of medical data processing technology. The method includes: acquiring and aggregating medical institution data based on NLP technology to obtain medical data processing results; acquiring and aggregating government data based on NLP technology to obtain government data processing results; acquiring and processing local hospital data based on NLP technology to obtain local data processing results; and constructing a medical knowledge graph based on the medical data processing results, government data processing results, and local data processing results, in combination with a data cognitive thinking model and a knowledge graph architecture. The present invention can perform comprehensive and accurate processing and analysis of medical data, and provide effective data support for subsequent medical services.

Description

Technical Field

[0001] The present invention relates to the field of medical data processing technology, and in particular to a medical data processing method and system based on artificial intelligence (NLP) technology. Background Art

[0002] In recent years, with the advancement of my country's health information construction, in order to enhance competitiveness and facilitate patients' medical treatment, hospitals have gradually improved the information management system that interconnects banks, social insurance, and new rural insurance. The existing hospital information system includes multiple systems such as HIS (hospital information system), LIS (laboratory information management system), RIS (radiology information system) / PACS (picture archiving and communication system), and the business systems launched in each hospital come from different manufacturers and different versions. The architecture, data structure, and data standards of each system may be different. Due to the above reasons, the amount of medical data has increased dramatically, and the complexity and efficiency of multiple links such as data collection, data processing, and data analysis have also followed.

[0003] Effective management of medical data is an important part of medical management and can provide effective support for medical diagnosis and treatment. In the existing technology, medical data of a certain dimension and level are usually managed centrally, resulting in low management efficiency and effect of medical data, and unable to achieve efficient and comprehensive medical data processing, and unable to provide better support for subsequent medical diagnosis and treatment. Summary of the invention

[0004] In order to overcome the above problems or at least partially solve the above problems, the present invention provides a medical data processing method and system based on artificial intelligence NLP technology, which can comprehensively and accurately process and analyze medical data and provide effective data support for subsequent medical services.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] In a first aspect, the present invention provides a medical data processing method based on artificial intelligence (NLP) technology, comprising the following steps:

[0007] Acquire and aggregate medical institution data based on NLP technology to obtain medical data processing results;

[0008] Acquire and aggregate government data based on NLP technology to obtain government data processing results;

[0009] Acquire and process local hospital data based on NLP technology to obtain local data processing results;

[0010] Based on the results of medical data processing, government data processing and local data processing, a medical knowledge graph is constructed by combining the data cognitive thinking model and knowledge graph architecture.

[0011] The present invention achieves accurate and comprehensive data analysis and processing by combining artificial intelligence NLP (natural language processing) technology to process data from different aspects separately. Based on the data processing results from various aspects, combined with the data cognitive thinking model and knowledge graph architecture, a comprehensive and fine-grained medical knowledge graph is constructed to realize the effective application of the knowledge graph and provide more effective data support for the follow-up.

[0012] Based on the first aspect, further, the method for aggregating medical institution data includes the following steps:

[0013] S11, performing usability judgment analysis on the medical institution data at the data source end in multiple dimensions. If the medical institution data meets the preset usability conditions in multiple dimensions, proceed to step S12; otherwise, a notification of data unusability is issued;

[0014] S12. Extract actual business data from the medical institution data, determine the data conversion logic, convert and clean the medical institution data according to the preset unified data model, and perform preliminary entry of the medical institution data;

[0015] S13. Use NLP technology to analyze and deeply process the cleaned medical institution data to obtain the corresponding medical combination data;

[0016] S14. According to the preset monitoring rules, the processed medical combination data is sent to the hospital-side monitoring system for real-time analysis and screening, and the screened anomalies are uploaded to the ODS layer.

[0017] Based on the first aspect, further, the method for aggregating government data includes the following steps:

[0018] Perform data dictionary conversion and data transcoding on government data, and perform data quality control;

[0019] After the processing is completed, according to the corresponding system application needs, different government data are gathered for data theme application, and data matching is carried out with medical institution data, and all data are synchronized to the preset disease control theme library;

[0020] The data based on the disease control theme library is systematically applied and displayed according to the preset conditions of different data early warning models. Early warning information is generated based on the early warning model, and the early warning information is simultaneously pushed to various modules of the medical system for early warning response.

[0021] Based on the first aspect, further, the method for processing local hospital data includes the following steps:

[0022] Build a unified data structure and standard to transfer the heterogeneous data of various hospitals to a unified data model;

[0023] Parse the clinical data of unstructured local hospital data and map it to a unified data model;

[0024] Conduct data quality control on clinical data from multiple dimensions, and record warning logs for clinical data that do not meet quality control requirements;

[0025] Monitor the resource utilization of each hospital based on local hospital data.

[0026] Based on the first aspect, further, the method for performing data quality control on clinical data from multiple dimensions includes the following steps:

[0027] Compare the converted data volume with the original data volume to determine whether there are omissions, over-recording, or wrong data in the business, and whether the data value conditions are reasonable;

[0028] Verify the source field assets of the conversion model to determine whether the field value is reasonable;

[0029] Perform duplicate and related data verification to determine whether the data is duplicated or whether there is isolated data;

[0030] Verify the field null value rate to determine whether data is lost.

[0031] Based on the first aspect, further, the method for processing local hospital data further includes the following steps:

[0032] Integrate the local data of the hospital between various business systems, pre-process the local medical data, and perform data annotation;

[0033] The preset syndrome model is trained based on the labeled local medical data, various diseases are monitored based on the trained syndrome model, and the data that conforms to the syndrome model is uploaded to the disease control subject database.

[0034] Based on the first aspect, further, the method for constructing a medical knowledge graph based on the medical data processing results, the government data processing results and the local data processing results, combined with the data cognitive thinking model and the knowledge graph architecture, includes the following steps:

[0035] Respectively identify the text data in the medical data processing results, government data processing results and local data processing results to identify the corresponding core medical entities;

[0036] Establish the relationship between core medical entities, extract relationships and events, build event graphs, integrate and reason about all medical information, and realize the recognition and understanding of medical texts;

[0037] Combine the knowledge graph and event graph to build a medical knowledge graph and provide knowledge graph services; and open the medical knowledge graph to various application modules in the medical system.

[0038] In a second aspect, the present invention provides a medical data processing system based on artificial intelligence NLP technology, including a medical institution data processing module, a government data processing module, a local data processing module and a graph construction module, wherein:

[0039] The medical institution data processing module is used to obtain and aggregate the medical institution data based on NLP technology to obtain the medical data processing results;

[0040] The government data processing module is used to obtain and aggregate government data based on NLP technology to obtain government data processing results;

[0041] The local data processing module is used to obtain and process the local data of the hospital based on NLP technology to obtain the local data processing results;

[0042] The graph construction module is used to build a medical knowledge graph based on the results of medical data processing, government data processing and local data processing, combined with the data cognitive thinking model and knowledge graph architecture.

[0043] This system realizes comprehensive and accurate processing and analysis of medical data through the cooperation of multiple modules such as medical institution data processing module, government data processing module, local data processing module and graph construction module, and provides effective data support for subsequent medical services. The present invention realizes accurate and comprehensive data analysis and processing by combining artificial intelligence NLP technology with different aspects of data for separate processing. Based on the data processing results of various aspects, combined with the data cognitive thinking model and knowledge graph architecture, a comprehensive and fine-grained medical knowledge graph is constructed to realize the effective application of knowledge graph and provide more effective data support for the follow-up.

[0044] In a third aspect, the present application provides an electronic device, comprising a memory for storing one or more programs; a processor; when the one or more programs are executed by the processor, the method as described in any one of the first aspects above is implemented.

[0045] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in any one of the first aspects above.

[0046] The present invention has at least the following advantages or beneficial effects:

[0047] The present invention provides a medical data processing method and system based on artificial intelligence NLP technology. By combining artificial intelligence NLP technology with data from different aspects for separate processing, accurate and comprehensive data analysis and processing are achieved. Based on the data processing results from various aspects, a comprehensive and fine-grained medical knowledge graph is constructed in combination with a data cognitive thinking model and a knowledge graph architecture, thereby achieving effective application of the knowledge graph and providing more effective data support for the follow-up. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0049] Figure 1 This is a flowchart of a medical data processing method based on artificial intelligence NLP technology according to an embodiment of the present invention;

[0050] Figure 2 A schematic diagram of a business process based on data cognition in an embodiment of the present invention;

[0051] Figure 3 This is a schematic diagram of the application service of the knowledge graph in an embodiment of the present invention;

[0052] Figure 4 This is a principle block diagram of a medical data processing system based on artificial intelligence NLP technology according to an embodiment of the present invention;

[0053] Figure 5 A structural block diagram of an electronic device provided by an embodiment of the present invention.

[0054] Explanation of the accompanying drawings: 100, medical institution data processing module; 200, government data processing module; 300, local data processing module; 400, graph construction module; 101, memory; 102, processor; 103, communication interface. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0056] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0058] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0059] In the description of the embodiments of the present invention, "plurality" means at least 2.

[0060] Example:

[0061] like Figure 1 As shown, in the first aspect, an embodiment of the present invention provides a medical data processing method based on artificial intelligence NLP technology, comprising the following steps:

[0062] S1. Acquire and aggregate medical institution data based on NLP technology to obtain medical data processing results; the medical institution data mainly refers to the business data of each medical institution.

[0063] S2. Acquire and aggregate government data based on NLP technology to obtain government data processing results; the government data mainly refers to non-medical diagnosis and treatment data, such as some personnel management data.

[0064] S3. Acquire and process the local hospital data based on NLP technology to obtain local data processing results; the local hospital data includes basic disease data information of each hospital and basic hospital information data.

[0065] S4. Based on the results of medical data processing, government data processing and local data processing, a medical knowledge graph is constructed by combining the data cognitive thinking model and the knowledge graph architecture.

[0066] The present invention achieves accurate and comprehensive data analysis and processing by combining artificial intelligence NLP (natural language processing) technology to process data from different aspects separately. Based on the data processing results from various aspects, combined with the data cognitive thinking model and knowledge graph architecture, a comprehensive and fine-grained medical knowledge graph is constructed to realize the effective application of the knowledge graph and provide more effective data support for the follow-up.

[0067] Based on the first aspect, in some embodiments of the present invention, further, the method for aggregating medical institution data includes the following steps:

[0068] S11, performing usability judgment analysis on the medical institution data at the data source end in multiple dimensions. If the medical institution data meets the preset usability conditions in multiple dimensions, proceed to step S12; otherwise, a notification of data unusability is issued;

[0069] S12. Extract actual business data from the medical institution data, determine the data conversion logic, convert and clean the medical institution data according to the preset unified data model, and perform preliminary entry of the medical institution data;

[0070] S13. Use NLP technology to analyze and deeply process the cleaned medical institution data to obtain the corresponding medical combination data;

[0071] S14. According to the preset monitoring rules, the processed medical combination data is sent to the hospital-side monitoring system for real-time analysis and screening, and the screened anomalies are uploaded to the ODS layer.

[0072] Conduct general quality control on the data at the data source end, and determine whether the data meets the requirements in terms of timeliness, data integrity, etc. If the data does not meet the requirements, notify in time to ensure the usability of the data. Based on historical project experience, sort out the data model that is common to disease control projects and clarify the data processing goals. Query actual business data, confirm the data conversion logic, and complete the initial entry of data. Sort out the hospital's dictionary and organize the fields that are actually used. As the hospital's electronic medical record information may be encrypted, contact the manufacturer to decrypt the data. For large text data, use NLP capabilities to complete the parsing of large texts. Considering data security, private data (name, ID number, contact information) is securely encrypted and stored in accordance with security technology that meets national standards. At present, PDX has contacted the Bureau of Political and Digital Affairs to ensure that the entire security storage system complies with security policies. Complete data quality control to ensure that the processed data is available. Use NLP technology to deeply process the data to meet the implementation of subsequent complex monitoring logic. The center formulates monitoring rules for each topic and sends them to the hospital-side monitoring system. The monitoring system performs real-time calculations and synchronizes the data that meets the requirements to the center through a secure channel, or feeds the data back to the hospital itself according to the center's requirements to form a closed loop on the link.

[0073] Based on the first aspect, in some embodiments of the present invention, further, the method for aggregating government data includes the following steps:

[0074] Perform data dictionary conversion and data transcoding on government data, and perform data quality control;

[0075] After the processing is completed, according to the corresponding system application needs, different government data are gathered for data theme application, and data matching is carried out with medical institution data, and all data are synchronized to the preset disease control theme library;

[0076] The data based on the disease control theme library is systematically applied and displayed according to the preset conditions of different data early warning models. Early warning information is generated based on the early warning model, and the early warning information is simultaneously pushed to various modules of the medical system for early warning response.

[0077] After government data is collected and synchronized to the ODS layer, it is governed according to specific topics and usage standards, and a specific governance process.

[0078] Take the monitoring of staff in key places as an example:

[0079] The commerce department provides daily punch-in record information of key venue staff and key venue work, and pushes data to the data front-end machine every day through the big data platform of the Bureau of Political and Digital Affairs; the intelligent multi-point trigger platform obtains data update information on the front-end machine every day, and synchronizes the data to the central data processing layer for data dictionary conversion and data transcoding, and simultaneously performs data quality control to verify data integrity and consistency; after the central data processing layer completes the data processing work, it aggregates different data for data theme application according to system application needs. For example, the key venue staff information provided by the commerce department is split into port terminal staff, quarantine hotel staff, airport entry staff, and cold chain cold storage staff list information, and is synchronized with the monitoring hospital information, fever clinic information, and nucleic acid testing information provided by the health department, and the pharmacy drug purchase information provided by the drug supervision department for data matching, and the data is synchronized to the theme library; the theme library data is applied to the system according to different data warning models (threshold) conditions and data is displayed, and the warning information is pushed to each module of the system simultaneously, and provided to the center for warning response through different warning level signals, forming a closed loop simultaneously.

[0080] Based on the first aspect, in some embodiments of the present invention, further, the method for processing local hospital data comprises the following steps:

[0081] Build a unified data structure and standard to transfer the heterogeneous data of various hospitals to a unified data model;

[0082] Parse the clinical data of unstructured local hospital data and map it to a unified data model;

[0083] Conduct data quality control on clinical data from multiple dimensions, and record warning logs for clinical data that do not meet quality control requirements;

[0084] Monitor the resource utilization of each hospital based on local hospital data.

[0085] The hospital information system includes multiple systems such as HIS, LIS, RIS / PACS, etc., and the business systems launched in each hospital come from different manufacturers and different versions. The architecture, data structure, and data standards of each system may be different. In order to facilitate the subsequent recognition of data, it is necessary to uniformly process the data of each business system, build a unified data structure and standard, and convert the heterogeneous data of each hospital into a unified data model. Parse the unstructured clinical data, automatically segment the text data, identify the content of each paragraph of text, add field labels, and map it to a unified data model. The data of the monitoring point hospital is controlled by the ETL platform, and the data quality is controlled according to the configured normative indicators, integrity indicators, consistency indicators, and timeliness indicators. Warning logs are recorded for data that does not meet the quality control requirements. Monitor each monitoring hospital at the center end, and display and alarm the resource usage of each testing hospital (such as CPU, memory, disk, network, etc.). Prometheus is used for system monitoring and alarm. Data is collected through exporter. Each hospital front-end machine reports data through pushgateway. Grafana on the central end can view the status chart of each hospital front-end machine, record the CPU, memory, disk usage and other conditions in detail, set alarm thresholds, and issue alarms via SMS.

[0086] Based on the first aspect, in some embodiments of the present invention, further, the method for performing data quality control on clinical data from multiple dimensions includes the following steps:

[0087] Compare the converted data volume with the original data volume to determine whether the business has any omissions, excess or errors, and whether the data acquisition conditions are reasonable; compare the converted data volume of the front-end original business data with the main table business data of the original business database to find out whether the business has any omissions, excess or errors, and whether the data acquisition conditions are reasonable.

[0088] Perform asset verification of the source fields of the conversion (asset verification of the source fields before conversion to the unified data model) to determine whether the field values ​​are reasonable; verify whether each field of the conversion is taken from the source field of the original business model to find out whether the field values ​​are reasonable.

[0089] Perform duplicate and associated data verification to determine whether the data is duplicated or whether there is isolated data; compare the primary key duplication and foreign key association of CDM data within the organization to ensure that there is no duplicate data and isolated data exists.

[0090] Verify the field null rate to determine whether data is lost; calculate the vacancy rate of important fields, check those with relatively low vacancy rates, analyze the reasons, and ensure that data is not lost without reason.

[0091] Based on the first aspect, in some embodiments of the present invention, further, the method for processing local hospital data further includes the following steps:

[0092] Integrate the local data of the hospital between various business systems, pre-process the local medical data, and perform data annotation;

[0093] The preset syndrome model is trained based on the labeled local medical data, various diseases are monitored based on the trained syndrome model, and the data that conforms to the syndrome model is uploaded to the disease control subject database.

[0094] In the process of monitoring and extracting data from the monitored hospitals, information personnel, medical experts, and clinical research experts jointly participate and confirm to build syndrome models for various diseases, and upload the original data of suspected cases that meet the syndrome model to the disease control subject database for subsequent early warning and prediction applications. The medical team formulates disease judgment rules based on the available medical knowledge and combines rule definition specifications to increase the speed and accuracy of the system's disease judgment. Extract the medical data obtained locally, complete the data fusion between various business systems of the hospital through information such as the patient's main index, and perform unified format conversion, data model mapping, semantic understanding and other pre-processing operations on these data to prepare for subsequent disease prediction. Then, train the syndrome model based on the labeled data to achieve automatic and rapid disease monitoring and upload the original data. In order to adapt to the ever-changing data description form and the newly added medical data, improve the robustness of the system, and continuously iterate and update the model so that it can adapt to the ever-changing external environment.

[0095] Based on the first aspect, in some embodiments of the present invention, further, the method for constructing a medical knowledge graph based on medical data processing results, government data processing results and local data processing results, combined with a data cognitive thinking model and a knowledge graph architecture, includes the following steps:

[0096] Respectively identify the text data in the medical data processing results, government data processing results and local data processing results to identify the corresponding core medical entities;

[0097] Establish the relationship between core medical entities, extract relationships and events, build event graphs, integrate and reason about all medical information, and realize the recognition and understanding of medical texts;

[0098] Combine the knowledge graph and event graph to build a medical knowledge graph and provide knowledge graph services; and open the medical knowledge graph to various application modules in the medical system.

[0099] Use relevant methods of natural language processing to structure the unstructured text in medical data. First, identify the core medical entities in the text data, then establish the relationship between these medical entities, and then aggregate and extract the event graph. Complete the integration and reasoning of medical information, and then realize the cognition and understanding of medical texts. At the same time, in order to improve the consistency of structured results and supplement the hidden or omitted information in medical texts. The system uses medical knowledge graphs to complete the information in the event graph and map related attributes to standard descriptors. In addition, in order to improve the intelligent processing capabilities of disease control products and maximize the use of medical information, the system opens the knowledge graph to various application modules. This not only makes it easier for each system module to use existing medical knowledge to improve the accuracy of the output results, but also absorbs new medical knowledge from each system module and expands the coverage of the knowledge graph.

[0100] Application services based on data cognition + knowledge graph of the present invention include: Figure 2-Figure 3 The list of cognitive + knowledge graph service functions is shown in Table 1:

[0101] Table 1

[0102]

[0103]

[0104] In some embodiments of the present invention, natural language processing (NLP) related technologies are used to process unstructured text data and convert it into structured graph data, namely event graphs. Other systems can make more full use of medical record data on this basis and further realize intelligent analysis of medical record data.

[0105] The main technologies used in data cognition systems include named entity recognition (NER), entity relationship extraction (Extract Entity Relation), dependency analysis (Dependency Analysis), event extraction, etc.

[0106] Medical knowledge graph can be regarded as the bottom layer, which can eliminate entity ambiguity and integrate knowledge from multiple data sources; with the natural support of graph database for relational data and the characteristics of graph topology, it can provide data support for upper-level business. The construction of graph includes: ontology construction, knowledge extraction, knowledge integration, knowledge storage, knowledge reasoning, etc. According to the organization form of data, the data in the data source can be divided into: structured data, semi-structured data, and unstructured data. Medical knowledge graph includes knowledge graph construction module and knowledge graph service module. Medical knowledge graph contains various types of data in the medical field, focusing on "disease", "clinical findings", "drugs", "operation", "observation operation", "body parts and structures", "treatment plan", etc. The knowledge graph construction module extracts the required knowledge from the above multiple types of data sources, organizes it into the required form, and stores it in the graph database. The knowledge graph service module provides web service, processes requests, and returns the data queried in the knowledge base.

[0107] 1. Entity recognition: Identify the core medical objects in the medical record text, such as diagnosis, surgery, medicine, examination, and test, to help the system "recognize" the medical record text.

[0108] 2. Dependency analysis: Establish the relationship between entities in the text. These relationships are mainly semantic relationships and simple medical relationships, so as to achieve the understanding of the medical record text on the basis of NER and further realize the recognition of medical record data.

[0109] 3. Entity relationship extraction: Identifying the medical relationships between entities in medical record texts is a supplement and extension of dependency analysis, and enables further understanding of medical record texts.

[0110] 4. Information assembly: Based on the rule base, the results of NER, entity relationship extraction and dependency analysis are combined to achieve further information aggregation, and obtain various events that can directly represent the various events that occurred during the patient's hospitalization, as well as the attributes and correlations of each event.

[0111] 5. Node query: Determine whether a node exists in the knowledge graph based on the query conditions. If it exists, return the main information of the node, otherwise return nothing.

[0112] 6. Terminology normalization: Convert non-standard medical terms into standard medical terms based on the knowledge graph.

[0113] Second, as Figure 4As shown, an embodiment of the present invention provides a medical data processing system based on artificial intelligence NLP technology, including a medical institution data processing module 100, a government data processing module 200, a local data processing module 300 and a graph construction module 400, wherein:

[0114] The medical institution data processing module 100 is used to obtain and aggregate the medical institution data based on NLP technology to obtain medical data processing results;

[0115] The government data processing module 200 is used to obtain and aggregate government data based on NLP technology to obtain government data processing results;

[0116] The local data processing module 300 is used to obtain and process the local data of the hospital based on the NLP technology to obtain the local data processing results;

[0117] The graph construction module 400 is used to construct a medical knowledge graph based on the medical data processing results, government data processing results and local data processing results, combined with the data cognitive thinking model and the knowledge graph architecture.

[0118] This system realizes comprehensive and accurate processing and analysis of medical data through the cooperation of multiple modules such as medical institution data processing module 100, government data processing module 200, local data processing module 300 and graph construction module 400, and provides effective data support for subsequent medical services. The present invention realizes accurate and comprehensive data analysis and processing by combining artificial intelligence NLP technology with different aspects of data for separate processing, and constructs a comprehensive and fine-grained medical knowledge graph based on the data processing results of various aspects combined with the data cognitive thinking model and knowledge graph architecture, so as to realize the effective application of knowledge graph and provide more effective data support for the follow-up.

[0119] like Figure 5 As shown, in a third aspect, an embodiment of the present application provides an electronic device, which includes a memory 101 for storing one or more programs and a processor 102. When the one or more programs are executed by the processor 102, any method in the first aspect described above is implemented.

[0120] The memory 101, the processor 102 and the communication interface 103 are directly or indirectly electrically connected to each other to achieve data transmission or interaction. For example, these elements can be electrically connected to each other through one or more communication buses or signal lines. The memory 101 can be used to store software programs and modules, and the processor 102 executes various functional applications and data processing by executing the software programs and modules stored in the memory 101. The communication interface 103 can be used to communicate signaling or data with other node devices.

[0121] Among them, the memory 101 can be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc.

[0122] The processor 102 may be an integrated circuit chip with signal processing capability. The processor 102 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0123] In the embodiments provided in the present application, it should be understood that the disclosed method and system can also be implemented in other ways. The method and system embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the method and system, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0124] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0125] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by the processor 102, a method as in any one of the first aspects described above is implemented. If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or part of the contribution to the prior art or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk.

[0126] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0127] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present application. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A medical data processing method based on artificial intelligence NLP technology, characterized in that: The following steps are involved: Acquire and aggregate medical institution data based on NLP technology to obtain medical data processing results; Acquire and aggregate government data based on NLP technology to obtain government data processing results; Acquire and process local hospital data based on NLP technology to obtain local data processing results; According to the results of medical data processing, government data processing and local data processing, a medical knowledge graph is constructed in combination with the data cognitive thinking model and the knowledge graph architecture, including: identifying the text data in the medical data processing results, government data processing results and local data processing results respectively to identify the corresponding core medical entities; establishing the relationship between each core medical entity, and extracting the relationship and events, constructing an event graph, integrating and reasoning all medical information, and realizing the cognition and understanding of medical texts; combining the knowledge graph and the event graph to construct a medical knowledge graph and provide knowledge graph services; and opening the medical knowledge graph to various application modules in the medical system; The method for aggregating medical institution data comprises the following steps: S11, performing usability judgment analysis on the medical institution data at the data source end in multiple dimensions. If the medical institution data meets the preset usability conditions in multiple dimensions, proceed to step S12; otherwise, a notification of data unusability is issued; S12. Extract actual business data from the medical institution data, determine the data conversion logic, convert and clean the medical institution data according to the preset unified data model, and perform preliminary entry of the medical institution data; S13. Use NLP technology to analyze and deeply process the cleaned medical institution data to obtain the corresponding medical combination data; S14. Send the processed medical combination data to the hospital-side monitoring system according to the preset monitoring rules for real-time analysis and screening, and upload the screened anomalies to the ODS layer; The method for aggregating and processing government affairs data comprises the following steps: Perform data dictionary conversion and data transcoding on government data, and perform data quality control; After the processing is completed, according to the corresponding system application needs, different government data are gathered for data theme application, and data matching is carried out with medical institution data, and all data are synchronized to the preset disease control theme library; The data based on the disease control theme library is systematically applied and displayed according to the preset conditions of different data early warning models. Early warning information is generated based on the early warning model, and the early warning information is simultaneously pushed to various modules of the medical system for early warning response.

2. According to claim 1, a medical data processing method based on artificial intelligence NLP technology is characterized in that: The method for processing local hospital data comprises the following steps: Build a unified data structure and standard to transfer the heterogeneous data of various hospitals to a unified data model; Parse the clinical data of unstructured local hospital data and map it to a unified data model; Conduct data quality control on clinical data from multiple dimensions, and record warning logs for clinical data that do not meet quality control requirements; Monitor the resource utilization of each hospital based on local hospital data.

3. A medical data processing method based on artificial intelligence NLP technology according to claim 2, characterized in that: The method for performing data quality control on clinical data from multiple dimensions comprises the following steps: Compare the converted data volume with the original data volume to determine whether there are omissions, over-recording, or wrong data in the business, and whether the data value conditions are reasonable; Verify the source field assets of the conversion model to determine whether the field value is reasonable; Perform duplicate and related data verification to determine whether the data is duplicated or whether there is isolated data; Verify the field null value rate to determine whether data is lost.

4. The medical data processing method based on artificial intelligence NLP technology according to claim 1 is characterized in that: The method for processing local hospital data also includes the following steps: Integrate the local data of the hospital between various business systems, pre-process the local medical data, and perform data annotation; The preset syndrome model is trained based on the labeled local medical data, various diseases are monitored based on the trained syndrome model, and the data that conforms to the syndrome model is uploaded to the disease control subject database.

5. A medical data processing system based on artificial intelligence NLP technology that executes the medical data processing method based on artificial intelligence NLP technology as described in any one of claims 1 to 4, characterized in that: It includes medical institution data processing module, government data processing module, local data processing module and graph construction module, among which: The medical institution data processing module is used to obtain and aggregate the medical institution data based on NLP technology to obtain the medical data processing results; The government data processing module is used to obtain and aggregate government data based on NLP technology to obtain government data processing results; The local data processing module is used to obtain and process the local data of the hospital based on NLP technology to obtain the local data processing results; The graph construction module is used to construct a medical knowledge graph based on the medical data processing results, government data processing results and local data processing results, combining the data cognitive thinking model and the knowledge graph architecture, including: identifying the text data in the medical data processing results, government data processing results and local data processing results respectively to identify the corresponding core medical entities; establishing the relationship between each core medical entity, and extracting relationships and events, constructing an event graph, integrating and reasoning all medical information, and realizing the cognition and understanding of medical texts; combining the knowledge graph and the event graph to construct a medical knowledge graph and provide knowledge graph services; and opening the medical knowledge graph to various application modules in the medical system.

6. An electronic device, characterized in that: include: A memory for storing one or more programs; processor; When the one or more programs are executed by the processor, the method according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images