Insurance excess dose identification system and device based on graph knowledge base
By introducing a graph knowledge base in the insurance overdose recognition system, a map of hospitals, pharmacies and insurance company data is built, and combining rules engines, machine learning models and anomaly detection algorithms, the existing system is solved accurately when identifying overdose behaviors. The problem of insufficient sexuality achieves higher recognition accuracy and accuracy.
Patent Information
- Application Number
- CN202411617623.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-13
- Publication Date
- 2025-02-21
AI Technical Summary
With the support of the existing insurance overdose recognition system without the support of the graph knowledge base, it is difficult to discover existing associations and rules, which affects the accuracy of identifying overdose behavior.
Adopt insurance overdose recognition system based on the graph knowledge base to collect data from hospitals, pharmacies and insurance companies through the data source layer. The graph construction layer performs data preprocessing, entity identification and extraction, graph construction and storage, and the intelligent analysis layer utilizes rules Engines, machine learning models, and anomaly detection algorithms identify behaviors overdose.
By constructing a graph knowledge base, the original data is converted into structured and analytical data, which improves the accuracy and accuracy of identification of excessive drug behaviors and enhances the ability to discover abnormal drug behaviors.
Smart Images
Figure CN119495442A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of medical overdose identification systems, and more specifically to a medical overdose identification system and device based on a graph knowledge base. Background Art
[0002] In the insurance industry, drug reimbursement is an important link, which aims to provide financial support for customers' medical expenses. Drug reimbursement refers to the insurance company's partial or full financial compensation for the cost of drugs purchased and used by customers for disease treatment according to the insurance contract. This process usually involves customers submitting medication records, prescription information and purchase receipts, and the insurance company conducts review and reimbursement.
[0003] The existing insurance overdose identification system is difficult to find the existing associations and rules without the support of the graph knowledge base, which affects the accuracy of identifying overdose behavior. Therefore, a new technical solution is needed to solve the problem. Summary of the invention
[0004] The purpose of the present invention is to provide a system and device for identifying overdose of insurance drugs based on a graph knowledge base, which solves the problem that the existing overdose identification system of insurance drugs is difficult to discover existing associations and rules without the support of a graph knowledge base, thus affecting the accuracy of identifying overdose behavior.
[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an insurance overdose identification system and device based on a graph knowledge base, comprising: an insurance overdose identification system, wherein the insurance overdose identification system comprises a data source layer, a graph construction layer, an intelligent analysis layer, an application interface layer and a user interaction layer, wherein the data source layer needs to collect hospital data, pharmacy data, insurance company data and other data sources, and the data support of multiple data sources provides strong support for subsequent graph construction and intelligent analysis;
[0006] The graph construction layer includes a data preprocessing module, an entity recognition and extraction module, a graph construction module and a graph storage and management module. The data preprocessing module cleans the data provided by the data source layer, removes duplicate, invalid and abnormal data, ensures the accuracy and consistency of the data, and converts the data into a format suitable for graph construction, ensuring that data from different sources meet unified standards in terms of format, unit and coding, so as to facilitate subsequent graph construction and analysis. The entity recognition and extraction module and the graph construction module work together to establish entity graphs, relationship graphs, comprehensive graphs, optimization graphs and monitoring and maintenance graphs through the entity recognition and extraction module, and perform node construction and edge construction through the graph construction module, using the identified entities as nodes in the graph, and assigning a unique identifier to each node. According to the relationship between entities, edges are constructed in the graph to represent the connection and interaction between entities. The graph storage and management module regularly evaluates the quality of the graph data, including the accuracy, completeness, consistency and timeliness of the data, monitors the update and change of the graph data in real time, and promptly discovers and handles data anomalies.
[0007] The intelligent analysis layer includes an overdose identification module, a risk assessment module, a decision support module and an algorithm optimization module. The overdose identification module is provided with a rule engine, a machine learning model and anomaly detection. Based on pre-set rules such as the upper limit of drug dosage, frequency of medication, etc., the prescription information in the map is matched and screened, and prescriptions that may contain overdose are preliminarily identified. The identified prescriptions are further analyzed and judged using a machine learning algorithm to improve the accuracy and precision of overdose identification. The data in the map is mined and analyzed through an anomaly detection algorithm to discover abnormal medication behavior.
[0008] As a preferred implementation of the present invention, the hospital data includes prescription information, patient medical records and drug delivery records.
[0009] As a preferred implementation of the present invention, the pharmacy data includes drug purchase records, drug sales records and source tracing records.
[0010] As a preferred implementation of the present invention, the insurance company data includes insurance claim records and customer information.
[0011] As a preferred implementation mode of the present invention, the application interface layer includes a data query interface, an overdose analysis interface, a result display interface, a security authentication interface and a log recording interface.
[0012] As a preferred implementation mode of the present invention, the user interaction layer includes login and authority management, claim review interface, history record interface and report generation interface.
[0013] As a preferred embodiment of the present invention, it includes: a body, an identification chamber is arranged inside the body and a rotating shaft is arranged on the inner bottom of the identification chamber, a placing table is arranged on the surface of the rotating shaft and the placing table is rotatably connected to the identification chamber through the rotating shaft, a plurality of recorders are arranged inside the identification chamber and a camera is arranged on the outer surface, a control panel is arranged on the surface of the body and the control panel is electrically connected to a host in the body, a fixing seat is arranged at the lower part of the identification chamber and an identity verification table and a code scanning device are arranged on the surface of the fixing seat, and a printing outlet is arranged at the lower part of the control panel.
[0014] As a preferred embodiment of the present invention, a moving mechanism is provided at the bottom of the machine body, and the moving mechanism includes a steering shaft, a mounting frame, a moving wheel and a self-locking mechanism, and an inspection door is provided on the surface of the machine body.
[0015] Compared with the prior art, the present invention has the following beneficial effects:
[0016] The insurance overdose identification system of the present invention includes a data source layer, a graph construction layer, an intelligent analysis layer, an application interface layer and a user interaction layer, wherein the data source layer needs to collect hospital data, pharmacy data, insurance company data and other data sources, and the data support of multiple data sources provides strong support for subsequent graph construction and intelligent analysis. The graph construction layer includes a data preprocessing module, an entity recognition and extraction module, a graph construction module and a graph storage and management module. The data preprocessing module cleans the data provided by the data source layer, removes duplicate, invalid and abnormal data, ensures the accuracy and consistency of the data, and converts the data into The data is converted into a format suitable for graph construction, ensuring that data from different sources meet unified standards in terms of format, unit and coding, which is convenient for subsequent graph construction and analysis. The entity recognition and extraction module works together with the graph construction module to establish entity graphs, relationship graphs, comprehensive graphs, optimization graphs and monitoring and maintenance graphs through the entity recognition and extraction module, and to build nodes and edges through the graph construction module. The identified entities are used as nodes in the graph, and each node is assigned a unique identifier. According to the relationship between entities, edges are built in the graph to represent the connection and interaction between entities. Graph storage and management The management module regularly evaluates the quality of the atlas data, including the accuracy, completeness, consistency and timeliness of the data, monitors the update and change of the atlas data in real time, and promptly discovers and handles data anomalies. The intelligent analysis layer includes the overdose identification module, risk assessment module, decision support module and algorithm optimization module. The overdose identification module is equipped with a rule engine, machine learning model and anomaly detection. Based on pre-set rules such as the upper limit of drug dosage and frequency of medication, it matches and screens the prescription information in the atlas, preliminarily identifies prescriptions that may have overdose, and uses machine learning algorithms to further analyze the identified prescriptions. Analysis and judgment can improve the accuracy and precision of over-dose identification. The data in the graph can be mined and analyzed through anomaly detection algorithms to discover abnormal medication behaviors. By constructing graph knowledge, the graph construction layer can convert the raw data provided by the data source layer into structured and analyzable graph data, providing a solid foundation for subsequent intelligent analysis and identification. The intelligent analysis layer can deeply mine and analyze the data provided by the data source layer, accurately identify over-dose behaviors, and provide decision-making support and suggestions for relevant personnel. The intelligent analysis layer can also continuously optimize and iterate algorithms to adapt to changing market demands and fraud methods. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 This is a schematic diagram of the insurance excess drug dosage identification system of the present invention;
[0018] Figure 2 This is a schematic diagram of the atlas construction module of the present invention;
[0019] Figure 3This is a schematic diagram of hospital data of the present invention;
[0020] Figure 4 This is a schematic diagram of pharmacy data of the present invention;
[0021] Figure 5 This is a schematic diagram of insurance company data of the present invention;
[0022] Figure 6 This is a schematic diagram of an overdose identification module of the present invention;
[0023] Figure 7 This is a schematic diagram of the structure of the identification device of the present invention;
[0024] Figure 8 This is a schematic diagram of the front view structure of the identification device of the present invention.
[0025] In the figure: 1. Machine body; 2. Identification chamber; 3. Camera; 4. Recorder; 5. Control panel; 6. Print outlet; 7. Fixed seat; 8. Identity verification table; 9. Code scanning device; 10. Inspection door; 11. Moving mechanism; 12. Rotating axis; 13. Placement table. DETAILED DESCRIPTION
[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.
[0027] See also Figure 1-8 The present invention provides a technical solution: an insurance overdose identification system and device based on a graph knowledge base, including: an insurance overdose identification system, wherein the insurance overdose identification system includes a data source layer, a graph construction layer, an intelligent analysis layer, an application interface layer, and a user interaction layer, wherein the data source layer needs to collect hospital data, pharmacy data, insurance company data, and other data sources, and the data support of multiple data sources provides strong support for subsequent graph construction and intelligent analysis;
[0028] The graph construction layer includes a data preprocessing module, an entity recognition and extraction module, a graph construction module and a graph storage and management module. The data preprocessing module cleans the data provided by the data source layer, removes duplicate, invalid and abnormal data, ensures the accuracy and consistency of the data, and converts the data into a format suitable for graph construction, ensuring that data from different sources meet unified standards in terms of format, unit and coding, so as to facilitate subsequent graph construction and analysis. The entity recognition and extraction module and the graph construction module work together to establish entity graphs, relationship graphs, comprehensive graphs, optimization graphs and monitoring and maintenance graphs through the entity recognition and extraction module, and perform node construction and edge construction through the graph construction module, using the identified entities as nodes in the graph, and assigning a unique identifier to each node. According to the relationship between entities, edges are constructed in the graph to represent the connection and interaction between entities. The graph storage and management module regularly evaluates the quality of the graph data, including the accuracy, completeness, consistency and timeliness of the data, monitors the update and change of the graph data in real time, and promptly discovers and handles data anomalies.
[0029] The data preprocessing module performs the following steps:
[0030] Remove duplicate prescriptions: Identify and remove duplicate prescription records by comparing fields such as patient ID, doctor ID, drug name, dosage, frequency of use, and prescription date in the prescription.
[0031] Deleting invalid prescriptions: Remove prescription records that are missing key information such as patient ID, drug name, or are obviously incorrect.
[0032] Handling abnormal medication frequency: Review prescriptions with abnormally high or low medication frequencies and modify or delete them based on actual conditions.
[0033] Data format conversion: Convert the date field in the prescription information into a unified date format, such as YYYY-MM-DD.
[0034] Data standardization: Unify the coding of drug names to ensure that drug names from different sources can be correctly matched and associated. Through these data preprocessing steps, the data quality provided by the data source layer can be ensured, providing a solid foundation for subsequent graph construction and intelligent analysis.
[0035] The intelligent analysis layer includes an overdose identification module, a risk assessment module, a decision support module and an algorithm optimization module. The overdose identification module is provided with a rule engine, a machine learning model and anomaly detection. Based on pre-set rules such as the upper limit of drug dosage, frequency of medication, etc., the prescription information in the map is matched and screened, and prescriptions that may contain overdose are preliminarily identified. The identified prescriptions are further analyzed and judged using a machine learning algorithm to improve the accuracy and precision of overdose identification. The data in the map is mined and analyzed through an anomaly detection algorithm to discover abnormal medication behavior.
[0036] Further improvements, such as Figure 5 As shown: the hospital data includes prescription information, patient medical records and drug delivery records. The prescription information includes patient ID, doctor ID, drug name, dosage, frequency of use, and prescription date. The prescription information directly reflects the drugs and dosages prescribed by the doctor for the patient; the patient medical records include the patient's medical history, allergy history, and past surgical records. This information is helpful for analyzing whether the patient needs to take certain drugs for a long time or in large quantities, thereby determining whether there is a risk of overdose; drug delivery records record the inventory status of hospital drugs, the quantity and date of drugs issued to patients, and can be used to verify the authenticity of prescription information and assist in identifying potential fraud.
[0037] Further improvements, such as Figure 5 As shown: The pharmacy data includes drug purchase records, drug sales records and source tracing records. The drug purchase records include patient ID, drug name, purchase quantity, purchase date, and pharmacy ID. These records are important data sources for identifying excessive drug dosage, because patients may purchase drugs through multiple pharmacies to bypass prescription restrictions. Drug sales records record the sales quantity and sales trends of various types of drugs. These data are helpful in analyzing whether there are abnormal sales of specific drugs and thus discover potential fraud.
[0038] Further improvements, such as Figure 5 As shown: The insurance company data includes insurance claim records and customer information. The insurance claim records include claim application ID, patient ID, drug costs, claim date, and claim status. These data are the key basis for the system to identify over-dose fraud, because fraudsters may submit false claims to obtain improper benefits. Customer information includes basic customer information such as name, age, gender, contact information, and insurance records. This information helps the system build a customer profile and assists in analyzing the authenticity of the claim application.
[0039] Further improvements, such as Figure 5As shown: the application interface layer includes a data query interface, an overdose analysis interface, a result display interface, a security authentication interface and a log recording interface. The data query interface queries relevant information of a specific insurance claim application, including patient information, drug purchase records, and prescription information; the overdose analysis interface performs overdose analysis on a specific insurance claim application and returns the analysis results, including the existence of overdose, the name of the overdose drug, and the exceeded dosage; the result display interface displays the results of the overdose analysis in a graphical manner, which is convenient for users to understand and process; the security authentication interface authenticates the users who access the system to ensure the security and privacy of the data. Sensitive data, including medical records, personal identity information and prescription information, will be encrypted using advanced encryption technology during transmission and storage, such as the AES-256 encryption algorithm, to ensure that the data is not accessed by unauthorized third parties. The system has a strict access control mechanism, and different users are granted different data access rights according to their roles and permissions. Only authorized administrators can access complete sensitive data, while ordinary users can only access desensitized data. Access and operations to sensitive data will be recorded in detail, and security audits will be conducted regularly to ensure compliance with data use. The logging interface records the system's access and operation logs for subsequent analysis and auditing.
[0040] Further improvements, such as Figure 5 As shown: the user interaction layer includes login and permission management, claim review interface, history record interface and report generation interface. Login and permission management provides a user login portal, supports user name, password and possible two-factor authentication login method, and assigns different access rights according to user roles to ensure data security and compliance. The claim review interface displays the claim application to be reviewed, including the patient's basic information, application time, and application amount. Click on the claim application to display a detailed analysis report of the application, including the possibility score of over-dose, abnormal drug purchase records, and related evidence chain. The history record interface displays the claim application records that the user has reviewed, supports filtering and sorting by time, status and other conditions, and can view the audit results, analysis reports and any notes at the time; the report generation interface provides a variety of report templates for users to choose from to meet different occasions and needs, allowing users to filter data according to specific conditions and configure the content, format and style of the report.
[0041] Further improvements, such as Figure 7 , 8As shown: it includes: a body 1, an identification chamber 2 is arranged inside the body 1, and a rotating shaft 12 is arranged on the inner bottom of the identification chamber 2, a placement table 13 is arranged on the surface of the rotating shaft 12, and the placement table 13 is rotatably connected to the identification chamber 2 through the rotating shaft 12, a plurality of recorders 4 are arranged inside the identification chamber 2, and a camera 3 is arranged on the outer surface, a control panel 5 is arranged on the surface of the body 1, and the control panel 5 is electrically connected to the host in the body 1, a fixing seat 7 is arranged at the lower part of the identification chamber 2, and an identity authentication table 8 and a code scanning device 9 are arranged on the surface of the fixing seat 7, and a printing outlet 6 is arranged at the lower part of the control panel 5.
[0042] Further improvements, such as Figure 7 As shown: a moving mechanism 11 is arranged at the bottom of the machine body 1 and the moving mechanism 11 includes a steering shaft, a mounting frame, moving wheels and a self-locking mechanism, and an inspection door 10 is arranged on the surface of the machine body 1.
[0043] Working principle: The insurance overdose identification system includes a data source layer, a graph construction layer, an intelligent analysis layer, an application interface layer, and a user interaction layer. The data source layer needs to collect hospital data, pharmacy data, insurance company data, and other data sources. The data support of multiple data sources provides strong support for subsequent graph construction and intelligent analysis. The graph construction layer includes a data preprocessing module, an entity recognition and extraction module, a graph construction module, and a graph storage and management module. The data preprocessing module cleans the data provided by the data source layer, removes duplicate, invalid, and abnormal data, ensures the accuracy and consistency of the data, and converts the data into The data is converted into a format suitable for graph construction, ensuring that data from different sources meet unified standards in terms of format, unit and coding, which is convenient for subsequent graph construction and analysis. The entity recognition and extraction module works together with the graph construction module to establish entity graphs, relationship graphs, comprehensive graphs, optimization graphs and monitoring and maintenance graphs through the entity recognition and extraction module, and to build nodes and edges through the graph construction module. The identified entities are used as nodes in the graph, and each node is assigned a unique identifier. According to the relationship between entities, edges are built in the graph to represent the connection and interaction between entities. Graph storage and management The management module regularly evaluates the quality of the atlas data, including the accuracy, completeness, consistency and timeliness of the data, monitors the update and change of the atlas data in real time, and promptly discovers and handles data anomalies. The intelligent analysis layer includes the overdose identification module, risk assessment module, decision support module and algorithm optimization module. The overdose identification module is equipped with a rule engine, machine learning model and anomaly detection. Based on pre-set rules such as the upper limit of drug dosage and frequency of medication, it matches and screens the prescription information in the atlas, preliminarily identifies prescriptions that may have overdose, and uses machine learning algorithms to further analyze the identified prescriptions. Analysis and judgment can improve the accuracy and precision of over-dose identification. The data in the graph can be mined and analyzed through anomaly detection algorithms to discover abnormal medication behaviors. By constructing graph knowledge, the graph construction layer can convert the raw data provided by the data source layer into structured and analyzable graph data, providing a solid foundation for subsequent intelligent analysis and identification. The intelligent analysis layer can deeply mine and analyze the data provided by the data source layer, accurately identify over-dose behaviors, and provide decision-making support and suggestions for relevant personnel. The intelligent analysis layer can also continuously optimize and iterate algorithms to adapt to changing market demands and fraud methods.
[0044] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached 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 are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.
[0045] Finally, a few points should be explained: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, and can be mechanical or electrical connections, or internal connectivity between two components, or direct connections. "Up", "down", "left", "right", etc. are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may change.
[0046] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. 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.
Claims
1. A system and device for identifying excessive drug dosage based on a graph knowledge base, characterized in that: include: Insurance overdose identification system, which includes a data source layer, a graph construction layer, an intelligent analysis layer, an application interface layer, and a user interaction layer. The data source layer needs to collect hospital data, pharmacy data, insurance company data, and other data sources. The data support of multiple data sources provides strong support for subsequent graph construction and intelligent analysis; The graph construction layer includes a data preprocessing module, an entity recognition and extraction module, a graph construction module and a graph storage and management module. The data preprocessing module cleans the data provided by the data source layer, removes duplicate, invalid and abnormal data, ensures the accuracy and consistency of the data, and converts the data into a format suitable for graph construction, ensuring that data from different sources meet unified standards in terms of format, unit and coding, so as to facilitate subsequent graph construction and analysis. The entity recognition and extraction module and the graph construction module work together to establish entity graphs, relationship graphs, comprehensive graphs, optimization graphs and monitoring and maintenance graphs through the entity recognition and extraction module, and perform node construction and edge construction through the graph construction module, using the identified entities as nodes in the graph, and assigning a unique identifier to each node. According to the relationship between entities, edges are constructed in the graph to represent the connection and interaction between entities. The graph storage and management module regularly evaluates the quality of the graph data, including the accuracy, completeness, consistency and timeliness of the data, monitors the update and change of the graph data in real time, and promptly discovers and handles data anomalies. The intelligent analysis layer includes an overdose identification module, a risk assessment module, a decision support module and an algorithm optimization module. The overdose identification module is provided with a rule engine, a machine learning model and anomaly detection. Based on pre-set rules such as the upper limit of drug dosage, frequency of medication, etc., the prescription information in the map is matched and screened, and prescriptions that may contain overdose are preliminarily identified. The identified prescriptions are further analyzed and judged using a machine learning algorithm to improve the accuracy and precision of overdose identification. The data in the map is mined and analyzed through an anomaly detection algorithm to discover abnormal medication behavior.
2. A system and device for identifying excessive drug dosage based on a graph knowledge base according to claim 1, characterized in that: The hospital data includes prescription information, patient medical records and drug delivery records.
3. The system and device for identifying excessive drug dosage based on a graph knowledge base according to claim 1, characterized in that: The pharmacy data includes drug purchase records, drug sales records and source tracing records.
4. The system and device for identifying excessive drug dosage based on a graph knowledge base according to claim 1, characterized in that: The insurance company data includes insurance claim records and customer information.
5. The system and device for identifying excessive drug dosage based on a graph knowledge base according to claim 1, characterized in that: The application interface layer includes a data query interface, an overdose analysis interface, a result display interface, a security authentication interface and a log recording interface.
6. The system and device for identifying excessive drug dosage based on a graph knowledge base according to claim 1, characterized in that: The user interaction layer includes login and authority management, claim review interface, history record interface and report generation interface.
7. The main control device of the insurance excess drug dosage identification system based on the graph knowledge base according to claim 1 is characterized by: include: A machine body (1), wherein an identification chamber (2) is arranged inside the machine body (1) and a rotating shaft (12) is arranged at the inner bottom of the identification chamber (2), a placement table (13) is arranged on the surface of the rotating shaft (12) and the placement table (13) is rotatably connected to the identification chamber (2) through the rotating shaft (12), a plurality of recorders (4) are arranged inside the identification chamber (2) and a camera (3) is arranged on the outer surface, a control panel (5) is arranged on the surface of the machine body (1) and the control panel (5) is electrically connected to a host in the machine body (1), a fixing seat (7) is arranged at the bottom of the identification chamber (2) and an identity verification table (8) and a code scanning device (9) are arranged on the surface of the fixing seat (7), and a printing outlet (6) is arranged at the bottom of the control panel (5).
8. The main control device of the insurance excess drug dosage identification system based on the graph knowledge base according to claim 7 is characterized by: A moving mechanism (11) is arranged at the bottom of the machine body (1), and the moving mechanism (11) comprises a steering shaft, a mounting frame, moving wheels and a self-locking mechanism, and an inspection door (10) is arranged on the surface of the machine body (1).