Medical insurance data checking method and system based on big data, electronic equipment and medium

Abstract

The invention discloses a medical insurance data checking method and system based on big data, electronic equipment and a medium, and relates to the technical field of data processing. The method comprises the following steps: acquiring position information and qualification information of a plurality of medical institutions in a target area and medical information of insured persons, constructing a medical institution space association network, and constructing an insured person-medical institution secondary association network according to the medical information of each insured person; calculating a reference medical-seeing radius, and screening out a cross-mechanism medical-seeing path exceeding the reference medical-seeing radius; determining a plurality of medical institution nodes meeting a preset frequent mutual access condition, obtaining medical insurance settlement data of each medical institution node, and calculating a corresponding collusion risk score; and taking the medical institution node of which the collusion risk score is greater than a preset threshold as a target node, and generating a medical insurance data check report corresponding to the target node and the insured group associated with the target node. By implementing the technical scheme provided by the invention, the effect of improving the accuracy of medical insurance data verification is achieved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and specifically to a medical insurance data verification method, system, electronic device and medium based on big data. Background Art

[0002] In the current healthcare field, with the aging population and the increase in medical needs, the pressure on the medical insurance system continues to rise. Effectively managing and monitoring the use of medical insurance funds to ensure that they are not abused or defrauded is a key task to maintain the sustainability of the entire healthcare system. Due to the large number of participants involved in medical insurance, frequent transactions, and huge amounts, the management of medical insurance funds faces great challenges.

[0003] At present, the existing medical insurance data verification methods mainly use staff verification and some integrated systems to conduct automated verification for different medical institutions. However, in actual applications, since medical insurance data involves many medical institutions and often have mutual influence, the existing method of conducting separate data verification for different medical institutions often easily ignores the risk issues between medical institutions, thereby reducing the accuracy of medical insurance data verification. Summary of the invention

[0004] The present application provides a medical insurance data verification method, system, electronic device and medium based on big data, which can improve the accuracy of medical insurance data verification.

[0005] In the first aspect, the present application provides a medical insurance data verification method based on big data, comprising: Obtain location information, qualification information, and medical information of insured persons of multiple medical institutions in the target area; Based on the location information and qualification information of each of the medical institutions, a spatial association network of medical institutions in the target area is constructed, and based on the medical information of each of the insured persons, a secondary association network of insured persons and medical institutions in the target area is constructed; Calculate the benchmark medical radius corresponding to the medical institution spatial association network, and screen out the cross-institutional medical paths that exceed the benchmark medical radius in the insured person-medical institution secondary association network; Determine multiple medical institution nodes that meet preset frequent mutual visit conditions in the cross-institution medical treatment path, and obtain medical insurance settlement data of each of the medical institution nodes; Combine the medical insurance settlement data of each of the medical institution nodes to calculate the corresponding collusion risk score; The medical institution node whose collusion risk score is greater than a preset threshold is taken as the target node, and a medical insurance data verification report corresponding to the target node and the insured person group associated with the target node is generated.

[0006] In a second aspect of the present application, a medical insurance data verification system based on big data is provided, the system comprising: The information acquisition module is used to obtain the location information, qualification information and medical information of multiple medical institutions in the target area as well as the medical information of the insured persons; An association network determination module is used to construct a spatial association network of medical institutions in the target area based on the location information and qualification information of each medical institution, and to construct a secondary association network of insured persons and medical institutions in the target area based on the medical information of each insured person; A path determination module is used to calculate the reference medical radius corresponding to the medical institution spatial association network, and screen out the cross-institution medical paths that exceed the reference medical radius in the insured person-medical institution secondary association network; The data verification module is used to determine multiple medical institution nodes that meet the preset frequent mutual visit conditions in the cross-institutional medical treatment path, and obtain the medical insurance settlement data of each of the medical institution nodes; calculate the corresponding collusion risk score in combination with the medical insurance settlement data of each of the medical institution nodes; take the medical institution node whose collusion risk score is greater than a preset threshold as the target node, and generate a medical insurance data verification report corresponding to the target node and the insured person group associated with the target node.

[0007] In a third aspect of the present application, an electronic device is provided, comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein the program can implement a medical insurance data verification method based on big data when loaded and executed by the processor.

[0008] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements a medical insurance data verification method based on big data.

[0009] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: By adopting the above technical solution, the location information, qualification information and medical information of multiple medical institutions in the target area are obtained, and based on this information, a medical institution spatial association network and a secondary association network of insured persons and medical institutions are constructed, thereby comprehensively reflecting the geographical and business associations between medical institutions. By calculating the benchmark medical radius corresponding to the medical institution spatial association network and screening out the cross-institutional medical paths that exceed the benchmark radius, potential abnormal medical behaviors can be effectively identified. The medical institution nodes that meet the preset frequent mutual visit conditions in these cross-institutional medical paths are further screened, and combined with their medical insurance settlement data, the corresponding collusion risk scores are systematically calculated. Finally, the medical institution nodes with risk scores exceeding the preset threshold are selected as target nodes, and a detailed medical insurance data verification report is generated, covering the information of the relevant insured groups. This technical solution constructs an association network by comprehensively utilizing geographical location, institution qualifications and insured behavior data, and uses dynamic screening and risk assessment mechanisms to make up for the shortcomings of existing methods in identifying inter-institutional association risks, thereby effectively preventing and combating the abuse and fraud of medical insurance funds, and achieving the effect of improving the accuracy of medical insurance data verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is a flowchart of a medical insurance data verification method based on big data provided in an embodiment of the present application; Figure 2 It is a structural diagram of a medical insurance data verification system based on big data provided in an embodiment of the present application; Figure 3 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application.

[0011] Description of reference numerals: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0012] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0013] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0014] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0015] This application embodiment provides a medical insurance data verification method based on big data. In one embodiment, please refer to Figure 1 , Figure 1 This is a flow chart of a medical insurance data verification method based on big data provided by an embodiment of the present application. The method can be implemented by a computer program, which can be integrated into an application or run as an independent tool application. The method can also be implemented by a single-chip microcomputer or run on a medical insurance data verification system based on big data based on the von Neumann system. Specifically, the method can include the following steps: Step 101: Obtain location information, qualification information, and medical information of insured persons of multiple medical institutions in the target area.

[0016] The target area specifically refers to the specific geographical scope where medical insurance data verification is required. It can be an administrative unit, such as a province, prefecture-level city, county, etc., or a functional area divided according to the distribution characteristics of medical resources, such as the coverage of a medical service consortium, etc. The division of the target area needs to take into account the spatial distribution characteristics of medical resources and the characteristics of the medical needs of the insured persons to ensure the rationality and completeness of the verification scope.

[0017] Medical institutions refer to various types of medical service institutions established and licensed in the target area, including but not limited to general hospitals, specialized hospitals, community health service centers, township health centers, etc. These medical institutions need to establish cooperative relations with the medical insurance system and be able to carry out medical insurance settlement business.

[0018] Location information includes the specific geographical location data of medical institutions, mainly including longitude coordinates, latitude coordinates, street addresses, etc. This location information is used to calculate the spatial distance between medical institutions, analyze the geographical distribution characteristics of medical institutions, and then evaluate the accessibility of medical resources and the rationality of medical treatment.

[0019] Qualification information refers to key indicators that can reflect the service capacity and scale of medical institutions, including the grade of medical institutions (such as Class III A, Class II A, etc.), the number of departments opened (reflecting the scope of diagnosis and treatment), and the number of practicing physicians (reflecting service capacity). This qualification information is used to evaluate the service level of medical institutions, determine the weight of institutions, and support the subsequent construction of association networks.

[0020] The medical information of the insured person refers to the relevant data generated during the insured person's medical treatment in a medical institution, mainly including the time of medical treatment (date and time of medical treatment), the medical institution (specific information of the medical institution visited), medical expenses (detailed details of various expenses incurred during medical treatment), etc.

[0021] Specifically, the above information is obtained through the interface of the medical insurance information system, the obtained raw data is standardized, duplicate, missing or abnormal data records are removed, and the processed data is stored in a preset database. By establishing a unified data acquisition and processing mechanism, the integrity and accuracy of the data required for subsequent analysis are ensured. For example, for the location information of medical institutions, address resolution can be performed through the geographic information system to convert text addresses into standard geographic coordinates; for qualification information, it can be standardized according to the institutional classification standards of the health administrative department; for the medical information of insured persons, data conversion can be performed according to a unified time format and cost unit. The purpose of this is to establish a complete data foundation to support subsequent network construction and risk analysis. By systematically collecting and standardizing the relevant information of medical institutions and insured persons, the medical service supply situation and the medical behavior characteristics of insured persons in the target area can be accurately reflected, providing reliable data support for identifying potential medical insurance fraud risks. At the same time, a unified data acquisition and processing mechanism can also help improve the efficiency and accuracy of medical insurance data verification, making the entire verification process more standardized and controllable.

[0022] Step 102: Based on the location information and qualification information of each medical institution, a spatial association network of medical institutions in the target area is constructed, and based on the medical information of each insured person, a secondary association network of insured persons and medical institutions in the target area is constructed.

[0023] Among them, the medical institution spatial association network is a network structure with medical institutions as nodes, which is used to describe the spatial association characteristics between medical institutions in the target area. This network structure can show the spatial distribution characteristics and service capacity differences of medical resources in the target area.

[0024] The insured person-medical institution secondary association network is a composite network structure that includes insured person nodes and medical institution nodes. It is used to describe the insured person's medical behavior characteristics and the business association relationship between medical institutions. This network structure can reflect the insured person's medical selection pattern and the business interaction relationship between medical institutions.

[0025] Specifically, a spatial association network of medical institutions is first constructed based on the location information and qualification information of the medical institutions. Specifically, by calculating the spatial distance between any two medical institutions in the target area, the medical institution pairs whose spatial distance is less than the preset distance threshold are identified as candidate association institutions. Then, based on the preset institution weight mapping table, the institution weight of each candidate association institution is determined according to the qualification information of the medical institution (including the level of the medical institution, the number of departments opened, and the number of practicing physicians). On this basis, the candidate association institution is used as the first network node, the institution weight is used as the node weight, and the spatial distance is used as the edge weight to construct the spatial association network of medical institutions in the target area.

[0026] Secondly, a secondary association network between insured persons and medical institutions is constructed based on the insured person's medical information. Specifically, data such as medical treatment time, medical institutions, and medical expenses are extracted from the insured person's medical information. Based on the temporal nature of medical treatment time, the flow path of the insured between different medical institutions is determined as the insured person's medical path. By counting the number of repetitions of each insured person's medical path within the preset time window, the medical path with a number of repetitions greater than the frequency threshold is determined as a frequent medical path. Then, the insured person nodes and medical institution nodes involved in the frequent medical path are used as the second network nodes, and the frequent medical path is used as the connecting edge between the nodes. The first weight of the insured node is determined based on the medical expenses, and the second weight of the medical institution node is determined based on the frequent medical path. Finally, a secondary association network between insured persons and medical institutions in the target area is constructed. This network can reflect the insured person's medical selection preference and the degree of business association between medical institutions.

[0027] By constructing these two network structures, the relationship between medical institutions can be characterized from the two dimensions of spatial association and business association, and the medical behavior patterns of insured persons can also be captured. The spatial association network of medical institutions helps to evaluate the rationality of medical distance, while the insured person-medical institution secondary association network helps to identify abnormal medical behavior patterns. Compared with the traditional single-point analysis method, this dual network analysis method can more comprehensively reflect the operating characteristics of the medical service system and provide a more reliable analysis basis for subsequent risk identification.

[0028] Based on the above embodiment, as an optional embodiment, in step 102: constructing a spatial association network of medical institutions in the target area based on the location information and qualification information of each medical institution, this step may also include the following steps: Step 201: Based on the location information of each medical institution, determine the spatial distance between any two medical institutions in the target area.

[0029] Specifically, first, any medical institution in the target area is taken as a reference point, and the spatial distance between the medical institution and all other medical institutions in the area is calculated in turn. In the calculation process, the curvature characteristics of the earth's surface need to be considered, so the spherical distance calculation method is used instead of the simple straight-line distance calculation method. This calculation method first converts the latitude and longitude coordinates of the medical institution into standard angular units, and then calculates the great circle distance between the locations of the two medical institutions based on the spherical geometry principle of the earth, that is, the shortest distance along the earth's surface. Such distance calculation is performed for all medical institutions in the target area, and finally a complete spatial distance matrix can be obtained, in which each element represents the actual geographical distance between two medical institutions. This calculation method based on spherical distance is more in line with the actual situation than the straight-line distance calculation, because it takes into account the curvature characteristics of the earth's surface and can more accurately reflect the actual geographical distance between medical institutions. The calculation results of spatial distance will be used to identify candidate associated institutions in the future, and are important basic data for constructing a spatial association network of medical institutions. Through the calculation of spatial distance, the geographical proximity between medical institutions can be objectively evaluated, which helps to judge the rationality of the insured person's medical choice and provide a reference basis for the spatial dimension for identifying abnormal medical behavior.

[0030] Step 202: any two medical institutions whose spatial distance is less than the distance threshold are selected as candidate associated institutions; based on a preset institution weight mapping table, the institution weight corresponding to the qualification information of each candidate associated institution is determined, and the qualification information includes the medical institution grade, the number of departments opened, and the number of practicing physicians.

[0031] Specifically, first, a reasonable distance threshold is set based on the calculated spatial distance. The setting of this distance threshold needs to take into account the distribution characteristics of medical resources in the target area and the conventional medical radius, which can be obtained by analyzing historical medical data. When the spatial distance between any two medical institutions is less than this distance threshold, the two medical institutions are identified as candidate associated institutions. This screening method based on distance threshold reflects the principle of geographical proximity, because medical institutions within a reasonable medical radius are more likely to have business dealings. Then, for the determined candidate associated institutions, their service capabilities need to be further evaluated. The qualification information of the medical institution is converted into the corresponding institutional weight through the pre-established institutional weight mapping table. The institutional weight mapping table sets the scoring criteria for the three dimensions of medical institution grade, number of departments opened, and number of practicing physicians. For example, the grade weight of a tertiary A hospital is higher than that of a secondary hospital. The medical institution with more departments opened has a higher department weight, and the medical institution with more practicing physicians has a higher personnel weight. Through the weighted calculation of these three dimensions, the comprehensive institutional weight of each candidate associated institution is finally obtained. The effect of this is to form a set of candidate associated institutions that considers both spatial proximity and service capacity differences. The threshold screening of spatial distance ensures that there is a possibility of geographical association between the selected medical institutions, while the introduction of institutional weights helps to identify the differences in service capabilities and complementary relationships between medical institutions. For example, when a low-level medical institution forms a candidate association with multiple high-level medical institutions, it may be necessary to pay attention to whether there are unnecessary referrals or decomposed hospitalizations. This analysis method combining spatial distance and institutional weights provides more targeted basic data for the subsequent construction of a medical institution spatial association network, which helps to improve the accuracy of medical insurance data verification.

[0032] Step 203: Using the candidate associated institution as the first network node, the institution weight as the node weight, and the spatial distance as the edge weight, a spatial association network of medical institutions in the target area is constructed.

[0033] Specifically, each candidate association institution is first taken as the first network node, and these nodes constitute the basic skeleton of the network. Then the calculated institution weight is assigned to the corresponding network node, so that the network node not only represents the location of the medical institution, but also reflects its service capacity level. Then, for any two candidate association institutions whose spatial distance is less than the distance threshold, a connection edge is established between their corresponding network nodes, and the calculated spatial distance is used as the edge weight. In this way, the geographical proximity relationship between medical institutions is transformed into an edge connection relationship in the network, and the edge weight reflects the actual geographical distance between medical institutions. The effect of this network construction method is to form a multi-level spatial association analysis framework. Through the node weight, the medical service capacity center in the target area can be identified; through the edge weight, the spatial accessibility between medical institutions can be evaluated; through the node connection relationship, the agglomeration of medical resources can be discovered. For example, when there is a short distance connection between a low-weight node and multiple high-weight nodes, it may be necessary to pay attention to whether there is an over-concentration of medical resources in the area; when the connection edge weight between some nodes is large but there is still frequent business dealings, it may be necessary to pay attention to whether there is unreasonable cross-regional medical behavior.

[0034] The spatial association network of medical institutions constructed in this way provides a systematic analysis tool for subsequent medical insurance data verification. By analyzing the topological structure characteristics of the network, potential risk points can be quickly identified, such as abnormal medical flows or unreasonable medical resource utilization patterns. At the same time, this networked expression method also helps medical insurance regulatory departments to intuitively understand and grasp the distribution of medical resources in the target area, providing decision-making support for the formulation of more targeted regulatory strategies.

[0035] Based on the above embodiment, as an optional embodiment, in step 102: constructing a secondary association network of insured persons and medical institutions in the target area according to the medical information of each insured person, this step may also include the following steps: Step 204: extract the medical treatment time, medical institution and medical expenses corresponding to each insured person from the medical treatment information of each insured person; for each insured person, based on the time sequence of the medical treatment time, the flow path of the insured person between different medical institutions is used as the insured person's medical treatment path.

[0036] Specifically, in order to construct path data that reflects the insured person's medical behavior pattern, it is necessary to extract and organize key elements from the insured person's medical information. First, parse each medical information record to extract the insured person's ID, medical time, medical institution ID, and medical expense information. Then, for each insured person, the flow sequence between different medical institutions is organized into a medical path according to the order of medical time. For example, if an insured person visits Hospital A, Hospital B, and Hospital C in one month, a medical path of "A→B→C" is formed. This time-series-based path extraction method can fully record the insured person's medical trajectory and provide basic data for subsequent identification of frequent medical patterns. In this way, discrete medical records can be converted into an ordered path sequence, effectively reflecting the insured person's medical selection behavior and the business association between medical institutions.

[0037] Step 205: Count the number of repetitions of each insured person's medical treatment path within the preset time window, and determine the insured person's medical treatment path with a number of repetitions greater than the frequency threshold as a frequent medical treatment path; use the insured person node and medical institution node involved in the frequent medical treatment path as the second network node, and use the frequent medical treatment path as the connecting edge between the nodes.

[0038] Specifically, in order to identify regular medical treatment behavior patterns, it is necessary to conduct frequency analysis on the insured person's medical treatment path. In specific implementation, first set a reasonable time window (such as three months) and frequency threshold (such as 3 times). In this time window, count the number of repetitions of each medical treatment path. If the number of repetitions of a certain insured person's medical treatment path in the time window exceeds the frequency threshold, the path is marked as a frequent medical treatment path. For example, if the insured person repeats the medical treatment path of "A→B→C" many times within three months, and the number of repetitions exceeds the threshold, this path is determined to be a frequent medical treatment path. Then, extract all the insured persons and medical institutions involved in these frequent medical treatment paths, and use them as insured person nodes and medical institution nodes respectively to form a second network node set. At the same time, the frequent medical treatment paths are converted into connecting edges in the network to represent the medical treatment association relationship between the insured person and the medical institution. This method can effectively screen out regular medical treatment behavior patterns and provide important clues for identifying potential medical insurance risks.

[0039] Step 206: Determine the first weight corresponding to each insured person node based on each medical expense, and determine the second weight corresponding to each medical institution node based on the frequent medical treatment path; combine each second network node, the first weight and the second weight corresponding to each second network node, and each connecting edge to construct an insured person-medical institution secondary association network in the target area.

[0040] Specifically, in order to comprehensively characterize the characteristics of the insured person's medical behavior and the business association of medical institutions, it is necessary to construct a secondary association network between the insured person and medical institutions. First, the first weight of the insured person node is determined based on medical expenses. The specific method is to calculate the proportion of the total medical expenses incurred by each insured person in all frequent medical paths. This weight reflects the proportion of the insured person in the overall medical expenses, which helps to identify insured persons with high medical expenses. Then, the second weight of the medical institution node is determined based on the frequent medical paths. The specific method is to count the number of times each medical institution appears as an intermediate node in all frequent medical paths. This weight reflects the intermediary role of the medical institution in the medical flow, which helps to discover potential collusion. Finally, the insured person nodes, medical institution nodes, node weights and connection edges are integrated into a secondary association network. This network structure can simultaneously reflect the insured person's medical choice preferences and the business association patterns between medical institutions, providing a comprehensive analytical framework for medical insurance supervision. For example, by observing the connection characteristics of high-weight insured person nodes, abnormal medical behavior can be discovered; by analyzing the intermediary role of high-weight medical institution nodes, potential collusion risks can be identified.

[0041] Based on the above embodiment, as an optional embodiment, in step 206: determining the first weight corresponding to each insured person node based on each medical expense, and determining the second weight corresponding to each medical institution node based on the frequent medical treatment path, this step may also include the following steps: Step 216: For any insured person involved in the frequent medical treatment path, determine the first weight of the corresponding insured person node based on the ratio between the medical expenses corresponding to the insured person and the total medical expenses of all insured persons.

[0042] Specifically, in order to accurately reflect the relative importance of each insured person in medical expenses, it is necessary to calculate the medical expense weight of each insured person. In specific implementation, the sum of the medical expenses of all insured persons involved in the frequent medical treatment path is first counted as the benchmark value. Then, for each insured person, the medical expenses incurred in the frequent medical treatment path are divided by this benchmark value to obtain the weight value of the insured person. For example, if the medical expenses incurred by an insured person in the frequent medical treatment path are 100,000 yuan, and the total medical expenses of all insured persons are 1 million yuan, the weight of the insured person node is 0.1. This weight calculation method based on the proportion of expenses can intuitively reflect the level of medical expense expenditure of the insured person, help identify insured persons with abnormal medical expenses, and provide important reference indicators for subsequent risk monitoring.

[0043] Step 226: For any medical institution involved in the frequent medical treatment path, obtain the number of times any medical institution appears as an intermediate node in the frequent medical treatment path.

[0044] Specifically, in order to evaluate the intermediary role of medical institutions in the medical path, it is necessary to count the frequency of each medical institution appearing as an intermediate node. In specific implementation, all frequent medical paths are traversed, and for the sequence of medical institutions in each path, the medical institutions that are not the starting point and the end point are marked as intermediate nodes. Then, the number of times each medical institution appears as an intermediate node is counted. For example, in the medical path "A→B→C", medical institution B is an intermediate node; if B appears as an intermediate node in multiple frequent medical paths, its number of appearances is accumulated. This statistical method can reflect the referral or triage role of medical institutions in the process of insured persons seeking medical treatment, and provide data support for evaluating the business association model of medical institutions.

[0045] Step 236: Divide the number of times any medical institution appears as an intermediate node in the frequent medical treatment path by the total number of times all medical institutions in the target area appear as intermediate nodes to obtain the second weight of the corresponding medical institution node.

[0046] Specifically, in order to quantify the intermediary status of medical institutions in the medical network, it is necessary to convert the number of intermediate node appearances into standardized weight values. In specific implementation, the total number of appearances of all medical institutions in the target area as intermediate nodes is first counted as the denominator. Then, the number of appearances of each medical institution as an intermediate node is divided by this total number to obtain the weight value of the medical institution node. For example, if a medical institution appears 50 times as an intermediate node, and the total number of times all medical institutions in the target area act as intermediate nodes is 500 times, the weight of the medical institution node is 0.1. This standardized weight calculation method can objectively reflect the relative importance of medical institutions in the medical network and help identify medical institutions with significant intermediary roles. By observing the business characteristics of these high-weight medical institutions, potential collusion risks or unreasonable referral behaviors can be discovered in a timely manner.

[0047] Step 103: Calculate the benchmark medical treatment radius corresponding to the medical institution spatial association network, and screen out the cross-institutional medical treatment paths that exceed the benchmark medical treatment radius in the insured person-medical institution secondary association network.

[0048] Among them, the benchmark medical radius represents the reasonable medical distance range within the target area, and reflects the acceptable distance between medical institutions for insured persons during normal medical treatment.

[0049] The cross-institutional medical treatment path refers to the medical treatment transfer path of the insured between medical institutions that are far apart and have the risk of collusion in the insured person-medical institution secondary association network. When the insured seeks medical treatment in multiple different medical institutions that exceed the benchmark medical treatment radius within a certain period of time, the path formed by connecting these medical institutions in the order of medical treatment time is the cross-institutional medical treatment path.

[0050] Specifically, when obtaining medical insurance data, it is found that some insured persons frequently travel back and forth between medical institutions that are far apart for medical treatment. This situation may reflect normal medical needs, such as seeking medical treatment in other places or seeking specialist treatment; but it may also imply the risk of medical insurance fraud, such as medical institutions colluding to make false settlements. In order to accurately identify potential abnormal medical treatment behaviors, the present invention calculates the benchmark medical treatment radius as a reference standard for reasonable medical treatment distance, and combines the first weight of the insured person node and the second weight of the medical institution node to screen cross-institutional medical treatment paths that exceed the reasonable range. In specific implementation, the geographic coordinates and institutional weights of each medical institution in the medical institution spatial association network are first obtained. For each medical institution, the system obtains the number of insured persons within its preset radiation range, and calculates the proportion of these insured persons to the total number of insured persons in the institution. By taking the medical institution as the center of the circle, the search radius is gradually expanded until the insured proportion reaches a preset coverage threshold (such as 90%), thereby determining the effective service radius of the medical institution. Subsequently, the system groups the medical institutions according to different institutional weights and calculates the mean effective service radius of each group. Finally, considering the proportion of the number of nodes in the spatial association network of medical institutions according to the weight group of each institution, the mean of the effective service radius of each group was weighted and calculated to obtain the benchmark medical radius of the target area.

[0051] After determining the baseline medical radius, the system combines the weight information in the insured person-medical institution secondary association network to perform path screening. The first weight of the insured person node reflects the proportion of the insured person's medical expenses in the total. A higher first weight indicates that the insured person's medical expenses are significantly higher than the average level, and it is necessary to focus on their medical behavior. The second weight of the medical institution node reflects the relative frequency of the institution as an intermediate node in the frequent medical path. A higher second weight means that the institution is more likely to participate in abnormal medical activities. The system prioritizes screening medical paths with insured persons with high first weights and medical institutions with high second weights, such as insured persons and medical institutions whose first and second weights are both greater than the preset weight threshold. At the same time, the actual distance between adjacent medical institution nodes in the path is calculated, and the path that exceeds the baseline medical radius is marked as a cross-institutional medical path. This screening method based on multiple weights not only takes into account the spatial distance factor, but also takes into account the medical expense characteristics of the insured person and the transit frequency characteristics of the medical institution. This method can more accurately identify potentially risky cross-institutional medical treatment paths: when the path contains both insured persons with abnormal medical expenses (the first weight is higher than the preset weight threshold) and medical institutions that frequently participate in transfers (the second weight is higher than the preset weight threshold), even if the actual distance is only slightly beyond the benchmark medical treatment radius, it will be marked by the system as an abnormal path that requires special attention. This multi-dimensional screening mechanism not only improves the accuracy of identifying abnormal behaviors, but also reduces the probability of misjudgment of normal medical treatment behaviors, providing more reliable data support for subsequent collusion risk assessment.

[0052] Based on the above embodiment, as an optional embodiment, in step 103: calculating the reference medical radius corresponding to the medical institution spatial association network, this step may also include the following steps: Step 301: Obtain the geographic coordinates and institution weight of each medical institution in the medical institution spatial association network.

[0053] Specifically, in order to accurately calculate the benchmark medical radius and consider the impact of differences in medical institution weights on medical distance, it is necessary to obtain the key attribute information of each medical institution in the medical institution spatial association network. This information is the basic data for subsequent spatial distance analysis and differentiated weight calculation. First, the geographic coordinate information of each medical institution, including longitude and latitude values, is extracted from the medical institution basic database. These coordinate data are usually based on a unified geographic coordinate system to ensure the accuracy and consistency of spatial location. At the same time, the level information of each medical institution is obtained from the medical institution weight assessment database, including classification identifiers such as third-level, second-level, and first-level. In the acquisition process, it is necessary to associate and match the geographic coordinates and institutional weight information through the unique identifier of the medical institution to ensure the integrity of the data and the accuracy of the corresponding relationship. The effect of this is to provide a reliable data basis for subsequent spatial analysis. Through the geographic coordinates of the medical institution, the spatial distance between any two medical institutions can be accurately calculated; through the weight information of the medical institution, different degrees of influence can be given to the connection relationship between medical institutions of different levels. For example, when calculating the benchmark medical radius, the influence coefficient can be adjusted according to the difference in the institutional weight of the medical institution, so that the calculation result is more in line with the actual characteristics of the medical service scope. At the same time, these basic data also provide an important reference for identifying abnormal cross-regional medical treatment behaviors, helping to improve the accuracy and effectiveness of medical insurance supervision.

[0054] Step 302: For any medical institution, obtain the number of insured persons within the preset radiation range of the medical institution and the insured ratio of the number of insured persons to the total number of insured persons in each medical structure; with the medical institution as the center, gradually expand the search radius until the insured ratio reaches the coverage threshold, and determine the corresponding search radius as the effective service radius of the medical institution.

[0055] Specifically, in order to accurately evaluate the actual service scope of each medical institution, it is necessary to dynamically calculate the effective service radius of the medical institution based on the medical data of the insured. This service radius calculation method based on actual medical data can more objectively reflect the spatial service characteristics of the medical institution and provide a more accurate reference standard for the subsequent identification of abnormal medical behavior. In specific implementation, first set the initial preset radiation range (such as 1 km), take the geographical coordinates of the medical institution as the center of the circle, and count the number of insured persons seeking medical treatment within the range. At the same time, count the total number of all insured persons seeking medical treatment in the medical institution, and calculate the proportion of the number of insured persons within the preset radiation range to the total number of insured persons, that is, the insurance ratio. Then, adopt an iterative search method to gradually increase the search radius (such as increasing by 0.5 km each time), and recalculate the insurance ratio after each expansion of the radius. When the insurance ratio reaches the preset coverage threshold (such as 80%), the current search radius is determined as the effective service radius of the medical institution. For example, if the number of insured persons seeking medical treatment within a 5-kilometer range of a tertiary hospital accounts for 80% of its total number of insured persons seeking medical treatment, the effective service radius of the hospital is 5 kilometers. The effect of this dynamic calculation method is that it can determine the reasonable service scope of medical institutions based on actual medical data. Medical institutions of different levels and sizes will present different effective service radiuses, which is more in line with the actual situation. For example, the effective service radius of large tertiary hospitals is usually larger, while the effective service radius of primary medical institutions is relatively smaller. This differentiated service radius calculation result can provide a more reasonable spatial reference standard for subsequent identification of abnormal medical behavior.

[0056] Step 303: Divide each medical institution into a plurality of institution groups corresponding to different institution weights, and based on the effective service radius of each medical institution, calculate the average effective service radius of the institution group corresponding to each institution weight.

[0057] Specifically, in order to scientifically evaluate the service scope characteristics of medical institutions of different levels, it is necessary to perform group statistical analysis on medical institutions. Since there are significant differences in the service capabilities and radiation ranges of medical institutions of different levels, the spatial service characteristics of medical institutions at all levels can be more accurately reflected through group calculation, providing a hierarchical reference standard for the subsequent identification of abnormal medical behavior. First, the medical institutions are divided into different institutional groups according to their grade attributes, for example, the medical institutions are divided into tertiary hospital groups, secondary hospital groups, primary hospital groups, etc. For each institutional group, the effective service radius data calculated in step 302 for all medical institutions in the group are collected, and then the arithmetic mean of the effective service radius in the group is calculated to obtain the average service radius of the medical institution of that level. For example, if the effective service radius of the tertiary hospitals in a certain area is 8 kilometers, 7 kilometers, and 9 kilometers respectively, the average effective service radius of the tertiary hospital group is 8 kilometers. In the calculation process, it can be considered to remove some obviously abnormal data samples to improve the representativeness of the calculation results. The effect of this group statistical method is that it can better reflect the relationship between the weight of the medical institution and the service radius. The spatial service characteristics of medical institutions of different levels can be clearly reflected by calculating the mean of the service radius of medical institutions of different levels. For example, under normal circumstances, the average service radius of tertiary hospitals is significantly greater than that of primary hospitals, which is in line with the basic principle of hierarchical diagnosis and treatment. These average data can be used as an important reference indicator to judge whether medical behavior is reasonable. When the medical distance of the insured person significantly exceeds the average service radius corresponding to the weight of the medical institution he / she is seeking medical treatment, special attention should be paid.

[0058] Step 304: Based on the proportion of the number of nodes in the medical institution spatial association network of the institution group corresponding to the weight of each institution, a weighted calculation is performed on the mean of each effective service radius to obtain a benchmark medical radius.

[0059] Specifically, in order to obtain a benchmark medical radius that can comprehensively reflect the characteristics of regional medical resource distribution, it is necessary to consider the number distribution of medical institutions of different levels. Since the proportion of medical institutions of different levels in the medical institution spatial association network is different, direct average calculation may not accurately reflect the actual medical space characteristics, so it is necessary to introduce a weight calculation mechanism. In specific implementation, first count the number of nodes of the institution group corresponding to the weight of each institution in the medical institution spatial association network, and calculate the proportion of the number of nodes in each group to the total number of nodes as the weight coefficient. For example, if the total number of medical institutions in a certain area is 100, including 10 tertiary hospitals, 30 secondary hospitals, and 60 primary hospitals, the weight of the tertiary hospital is 0.1, the weight of the secondary hospital is 0.3, and the weight of the primary hospital is 0.6. Then, the mean effective service radius of medical institutions of each level calculated in step 303 is multiplied by the corresponding weight coefficient and summed to obtain the final benchmark medical radius. The specific calculation formula is: benchmark medical radius = Σ (mean effective service radius of medical institutions of each level × corresponding node number proportion). The effect of this weighted calculation method is that it can more accurately reflect the overall medical space characteristics in the region. By considering the number distribution of medical institutions, the calculated benchmark medical radius is more targeted.

[0060] Step 104: Determine multiple medical institution nodes that meet the preset frequent mutual visit conditions in the cross-institutional medical treatment path, and obtain the medical insurance settlement data of each medical institution node.

[0061] Among them, the preset frequent mutual visit conditions refer to a set of conditional parameters used to measure the degree of overlap of the medical treatment trajectories of insured persons between medical institutions. These conditional parameters include: the mutual visit time interval threshold, that is, the shortest time interval between insured persons going back and forth between two medical institutions for medical treatment; the mutual visit number threshold, that is, the lower limit of the number of insured persons going back and forth between two medical institutions for medical treatment within a specific time window; and the mutual visit frequency threshold, that is, the minimum number of times a single insured person goes back and forth between two medical institutions for medical treatment within a specific time window. When the mutual visits between medical institutions meet the above threshold requirements, it is considered that there is a frequent mutual visit relationship between these medical institutions.

[0062] A medical institution node refers to an entity unit that represents a single medical institution in the medical institution spatial association network. Each node contains the key attribute information of the medical institution, such as institution code, geographic coordinates, institution level, effective service radius, etc. These nodes form a spatial association relationship through the medical treatment path of the insured person, forming a complete medical institution spatial association network.

[0063] Medical insurance settlement data refers to the detailed transaction records generated when medical institutions settle fees with medical insurance agencies after providing medical services to insured persons. These data include: total medical expenses, medical insurance reimbursement amount, personal payment amount, cost category composition (such as drug fees, examination fees, treatment fees, etc.), settlement time, medical treatment category (outpatient / inpatient), etc. These data can reflect the cost control level of medical institutions and the use of medical insurance funds, and are an important basis for analyzing abnormal medical behavior.

[0064] Specifically, in order to identify potential abnormal medical treatment behaviors from a large number of cross-institutional medical treatment paths, it is necessary to focus on those medical institution nodes that frequently visit each other. There may be unreasonable interest associations between these medical institutions that frequently visit each other, and their medical insurance settlement data may contain clues of abnormal medical treatment or medical insurance fraud. In specific implementation, it is necessary to first define the preset frequent mutual visit conditions, including two key parameters: the mutual visit frequency threshold and the time window. For example, "the number of insured persons who refer each other for medical treatment within 3 months exceeds 50" can be set as the frequent mutual visit condition. Then, traverse all cross-institutional medical treatment paths and count the mutual visit frequency between each pair of medical institution nodes. When the mutual visit between certain medical institution nodes meets the preset frequent mutual visit conditions, these nodes are marked as key focus objects. For these key focus medical institution nodes, their detailed medical insurance settlement data, including the number of medical visits, cost composition, reimbursement amount and other information, are extracted from the medical insurance settlement database. If it is found that there are frequent two-way referrals between two medical institutions and the medical insurance settlement amount is abnormal, this may mean that there is a risk of collusion fraud. The effect of doing this is that it can effectively screen out medical institution combinations that deserve special attention and improve the efficiency of identifying abnormal behavior.

[0065] Based on the above embodiment, as an optional embodiment, in step 104: determining multiple medical institution nodes that meet the preset frequent mutual visit conditions in the cross-institution medical treatment path, this step may also include the following steps: Step 401: Obtain the insured person flow data between medical institutions in the cross-institutional medical treatment path, and calculate the insured person overlap between any two medical institutions based on the insured person flow data.

[0066] Specifically, in order to quantitatively evaluate the degree of medical association between medical institutions, it is necessary to analyze the flow of insured persons between medical institutions. In specific implementation, first extract the medical records of insured persons between each pair of medical institutions from the cross-institutional medical path, and count the insured set A who have medical records in both medical institutions within a specific time window (such as 3 months), and the insured set B who have medical records in any medical institution. Then use the Jaccard coefficient to calculate the overlap of insured persons, that is, the overlap of insured persons = |A| / |B|, where |A| represents the number of insured persons in set A, and |B| represents the number of insured persons in set B. This calculation method can effectively reflect the degree of overlap of insured persons between two medical institutions and provide a quantitative basis for identifying frequent mutual visits.

[0067] Step 402: any two medical institutions whose insured person overlap degree is greater than the overlap degree threshold are determined as frequently visited node pairs, and a medical institution association subgraph is constructed based on the frequently visited node pairs.

[0068] Specifically, in order to construct a network structure that reflects the close relationship between medical institutions, it is necessary to screen frequently visited node pairs based on the overlap of insured persons. In specific implementation, a reasonable overlap threshold (such as 0.1) is set. When the overlap of insured persons between any two medical institutions exceeds the threshold, the two medical institutions are identified as a frequently visited node pair. All frequently visited node pairs are taken as edges, and the associated medical institutions are taken as vertices to construct a directed graph as a medical institution association subgraph. This construction method can transform the loose association relationship between medical institutions into a structured network representation, which is convenient for subsequent graph algorithm analysis.

[0069] Step 403: Identify strongly connected components in the medical institution associated subgraph through a preset graph traversal algorithm, and determine the medical institution nodes in the strongly connected components as medical institution nodes that meet the preset frequent mutual visit conditions.

[0070] Specifically, in order to discover closely related groups between medical institutions, it is necessary to identify strongly connected components in the medical institution association subgraph. In the specific implementation, a graph traversal algorithm such as the Kosaraju algorithm or the Tarjan algorithm is used. First, a depth-first search is performed on the graph to obtain the completion time order of the nodes. Then, a second depth-first search is performed on the transposition of the graph in descending order of completion time to identify the strongly connected components in the graph. There are bidirectionally reachable paths between the medical institution nodes in the strongly connected components, indicating that there is a close mutual visit relationship between these medical institutions. Therefore, these nodes are identified as medical institution nodes that meet the preset frequent mutual visit conditions. This graph theory-based analysis method can effectively identify groups of medical institutions with complex mutual visit relationships and provide important clues for subsequent abnormal behavior analysis.

[0071] Step 105: Combine the medical insurance settlement data of each medical institution node and calculate the corresponding collusion risk score.

[0072] Among them, the collusion risk score refers to a comprehensive numerical indicator used to quantitatively assess the risk level of potential collusion between medical institutions.

[0073] Specifically, in order to conduct risk assessment on medical institutions with frequent mutual visits, it is necessary to build a collusion risk assessment model based on medical insurance settlement data. Since collusion between medical institutions often leaves abnormal features in medical insurance settlement data, potential collusion risks can be effectively identified by multi-dimensional analysis and quantitative scoring of these data. In specific implementation, characteristic indicators of medical insurance settlement data are first extracted from multiple dimensions. These indicators include: number of medical visits, settlement amount, and reimbursement amount. The system will comprehensively consider these multi-dimensional data and calculate the collusion risk score to quantitatively assess the risk situation between multiple medical institution nodes.

[0074] Based on the above embodiment, as an optional embodiment, in step 105: combining the medical insurance settlement data of each medical institution node to calculate the corresponding collusion risk score, this step may also include the following steps: Step 501: extract the corresponding number of medical visits, settlement amount and reimbursement amount from each medical insurance settlement data.

[0075] Specifically, in order to accurately evaluate the service behavior characteristics of medical institutions, it is necessary to extract key quantitative indicators from the medical insurance settlement data. In specific implementation, for each medical institution node, the medical visit data (i.e. the total number of times the insured person has visited the medical institution) within a preset period (such as the past 3 months), the settlement amount data (i.e. the total cost of medical services), and the reimbursement amount data (i.e. the actual cost paid by the medical insurance fund) are counted from its medical insurance settlement database. The extraction of these basic data provides reliable data support for the subsequent calculation of risk assessment indicators, and can truly reflect the business scale and capital flow of medical institutions.

[0076] Step 502: Based on the number of visits to each medical institution, calculate the frequency of visits to each medical institution node within a preset time period; based on the settlement amount and the number of visits to each medical institution, calculate the corresponding per capita medical expenses; based on the settlement amount and reimbursement amount of each medical institution, calculate the corresponding reimbursement ratio.

[0077] Specifically, in order to build a multi-dimensional risk assessment system, it is necessary to calculate multiple standardized indicators based on the extracted basic data. In specific implementation, the frequency of medical treatment is first calculated, that is, the total number of medical visits in the medical institution within the preset period is divided by the length of the period (such as 90 days) to obtain the average number of medical visits per day; secondly, the per capita medical expenses are calculated, that is, the total settlement amount of the medical institution is divided by the total number of medical visits to obtain the average cost level of each medical visit; finally, the reimbursement ratio is calculated, that is, the total reimbursement amount of the medical institution is divided by the total settlement amount to obtain the proportion of expenses paid by the medical insurance fund. This standardized processing can eliminate the impact of differences in the size of medical institutions and make the indicators between different institutions comparable.

[0078] Step 503: Perform weighted summation on the frequency of medical visits, per capita medical expenses, and reimbursement ratio of each medical institution to obtain the corresponding collusion risk score.

[0079] Specifically, in order to comprehensively evaluate the collusion risk level of medical institutions, it is necessary to integrate multiple standardized indicators into a single risk score. In specific implementation, each indicator is first compared with the industry benchmark value to calculate the degree of abnormality. For example, the degree of abnormality in the frequency of medical visits = (actual frequency of medical visits - industry average frequency) / industry average frequency, and the degree of abnormality in per capita medical expenses and the degree of abnormality in reimbursement ratio are also calculated in a similar way. Then, a weight coefficient (such as w1, w2, w3) is set for each abnormality indicator, and the final collusion risk score is calculated by weighted summation: Collusion risk score = w1 × degree of abnormality in the frequency of medical visits + w2 × degree of abnormality in per capita medical expenses + w3 × degree of abnormality in reimbursement ratio. This weighted calculation method can be flexibly adjusted according to the importance of different indicators to improve the accuracy and adaptability of risk assessment. At the same time, by setting reasonable weight parameters, the influence of each dimension can be effectively balanced, making the risk assessment results more objective and comprehensive.

[0080] Step 106: The medical institution nodes whose collusion risk scores are greater than a preset threshold are taken as target nodes, and a medical insurance data verification report corresponding to the target nodes and the insured person groups associated with the target nodes is generated.

[0081] Among them, the target node in this embodiment of the application refers to the nodes of medical institutions whose collusion risk scores exceed the preset threshold. These nodes are medical institutions with high collusion risks identified by weighted calculation of indicators such as the frequency of medical treatment, per capita medical expenses, and reimbursement ratio of the medical institutions. These medical institutions may have conspired with other medical institutions to commit medical insurance fraud and need to be monitored and verified.

[0082] In the embodiment of this application, the medical insurance data verification report refers to the medical insurance data analysis report for the target node and its associated insured group. The report mainly includes basic information of medical institutions, abnormal indicator analysis, related party characteristics, insured group portraits, medical insurance settlement details, etc., which is used to display the abnormal behavior characteristics of the target node, related party information and the medical treatment mode of the relevant insured persons, and provide data support and direction guidance for the medical insurance regulatory department to carry out on-site verification.

[0083] In the embodiment of the present application, the insured group refers to the set of insured persons who seek medical treatment at the target node and medical institutions that have frequent visits to the target node. The medical behavior characteristics of these insured persons, such as the frequency of medical treatment, cross-institutional medical treatment paths, medical expenses and other information, are important bases for judging whether there is collusion among medical institutions. By analyzing the medical insurance data of the insured group, abnormal medical treatment patterns and fee settlement patterns can be found, which helps to identify potential medical insurance fraud.

[0084] Specifically, in order to achieve accurate delivery and efficient use of medical insurance regulatory resources, it is necessary to focus on monitoring and analyzing medical institutions whose collusion risk scores exceed the preset threshold. In specific implementation, first set a reasonable risk score threshold (such as the 75% quantile of the risk score) based on historical regulatory experience and risk tolerance, and mark the nodes of medical institutions with collusion risk scores higher than the threshold as target nodes. Then, build a medical insurance data verification report based on the target node, which mainly includes the following core contents: basic information of medical institutions (including institution name, level, geographical location, etc.), abnormal indicator analysis (detailed listing of specific indicators that lead to high risk scores and their degree of abnormality), related party characteristics (analysis of the characteristics of other medical institutions that have frequent visits with the target node), insured person group portrait (statistical age distribution, disease spectrum distribution, medical frequency distribution, etc. of insured persons who seek medical treatment at the target node), medical insurance settlement details (displaying suspicious fee structure and settlement mode). In the process of generating the verification report, focus on the medical behavior characteristics of the insured population. For example, by analyzing the medical records of insured persons at the target node, high-risk insured persons with abnormal medical treatment patterns can be identified, such as frequent medical treatment in a short period of time, repeated medical treatment across regions, etc. At the same time, the medical insurance settlement data of the target node is deeply mined, focusing on analyzing the rationality of the cost structure, such as whether the proportion of drug costs and the proportion of examination costs significantly deviate from the normal level of medical institutions of the same level. In addition, it is necessary to evaluate the rationality of its service scope in combination with the distribution of medical resources in the area where the target node is located. This method based on risk threshold screening and multi-dimensional analysis can help medical insurance regulatory departments accurately locate high-risk medical institutions and understand the specific manifestations and possible causes of their abnormal behaviors. The generated medical insurance data verification report provides a clear direction and focus for subsequent on-site verification, and improves the pertinence and efficiency of regulatory work. At the same time, this standardized reporting format also facilitates horizontal and vertical comparisons of regulatory results in different regions and different periods, which helps to discover and summarize the typical characteristics and evolution laws of collusion, and provide decision-making support for improving the medical insurance regulatory mechanism.

[0085] Reference Figure 2 , a medical insurance data verification system based on big data is provided in an embodiment of the present application, the system comprises: an information acquisition module, an associated network determination module, a path determination module, and a data verification module, wherein: The information acquisition module is used to obtain the location information, qualification information and medical information of multiple medical institutions in the target area as well as the medical information of the insured persons; The association network determination module is used to construct a spatial association network of medical institutions in the target area based on the location information and qualification information of each medical institution, and to construct a secondary association network of insured persons and medical institutions in the target area based on the medical information of each insured person; The path determination module is used to calculate the benchmark medical treatment radius corresponding to the medical institution spatial association network, and screen out the cross-institutional medical treatment paths that exceed the benchmark medical treatment radius in the insured person-medical institution secondary association network; The data verification module is used to determine multiple medical institution nodes that meet the preset frequent mutual visit conditions in the cross-institutional medical treatment path, and obtain the medical insurance settlement data of each medical institution node; combine the medical insurance settlement data of each medical institution node to calculate the corresponding collusion risk score; take the medical institution node with a collusion risk score greater than a preset threshold as the target node, and generate a medical insurance data verification report corresponding to the target node and the insured person group associated with the target node.

[0086] On the basis of the above embodiments, the association network determination module is also used to determine the spatial distance between any two medical institutions in the target area based on the location information of each medical institution; any two medical institutions whose spatial distance is less than the distance threshold are taken as candidate association institutions; based on a preset institution weight mapping table, the institution weight corresponding to the qualification information of each candidate association institution is determined, and the qualification information includes the level of the medical institution, the number of departments opened and the number of practicing physicians; the candidate association institution is used as the first network node, the institution weight is used as the node weight, and the spatial distance is used as the edge weight to construct a spatial association network of medical institutions in the target area.

[0087] On the basis of the above embodiments, the association network determination module is also used to extract the medical treatment time, medical institution and medical expenses corresponding to each insured person from the medical treatment information of each insured person; for each insured person, based on the time sequence of the medical treatment time, the flow path of the insured person between different medical institutions is used as the insured person's medical treatment path; the number of repeated occurrences of the medical treatment path of each insured person within the preset time window is counted, and the medical treatment path of the insured person whose number of repeated occurrences is greater than the frequency threshold is determined as a frequent medical treatment path; the insured person node and medical institution node involved in the frequent medical treatment path are used as second network nodes, and the frequent medical treatment path is used as the connecting edge between nodes; the first weight corresponding to each insured person node is determined based on each medical expense, and the second weight corresponding to each medical institution node is determined based on the frequent medical treatment path; combining each second network node, the first weight and the second weight corresponding to each second network node and each connecting edge, a secondary association network of insured persons and medical institutions in the target area is constructed.

[0088] On the basis of the above embodiment, the associated network determination module is also used to determine the first weight of the corresponding insured person node for any insured person involved in the frequent medical path based on the ratio between the corresponding medical expenses of the insured person and the total medical expenses of all insured persons; for any medical institution involved in the frequent medical path, obtain the number of times any medical institution appears as an intermediate node in the frequent medical path; divide the number of times any medical institution appears as an intermediate node in the frequent medical path by the total number of times all medical institutions in the target area appear as intermediate nodes to obtain the second weight of the corresponding medical institution node.

[0089] On the basis of the above embodiments, the path determination module is also used to obtain the geographic coordinates and institutional weights of each medical institution in the spatial association network of medical institutions; for any medical institution, the number of insured persons within the preset radiation range of the medical institution and the proportion of the number of insured persons to the total number of insured persons in each medical structure are obtained; with the medical institution as the center of the circle, the search radius is gradually expanded until the insured ratio reaches the coverage threshold, and the corresponding search radius is determined as the effective service radius of the medical institution; each medical institution is divided into institutional groups corresponding to a plurality of different institutional weights, and based on the effective service radius of each medical institution, the effective service radius mean of the institutional group corresponding to each institutional weight is calculated; based on the proportion of the number of nodes of the institutional group corresponding to each institutional weight in the spatial association network of medical institutions, the mean of each effective service radius is weightedly calculated to obtain a benchmark medical radius.

[0090] On the basis of the above-mentioned embodiment, the data verification module is also used to obtain the insured person flow data between the medical institutions in the cross-institutional medical treatment path, and based on the insured person flow data, calculate the insured person overlap between any two medical institutions; any two medical institutions whose insured person overlap is greater than the overlap threshold are determined as frequently visited node pairs, and a medical institution association subgraph is constructed based on the frequently visited node pairs; the strongly connected components in the medical institution association subgraph are identified through a preset graph traversal algorithm, and the medical institution nodes in the strongly connected components are determined as medical institution nodes that meet the preset frequent visit conditions.

[0091] On the basis of the above embodiments, the data verification module is also used to extract the corresponding number of medical visits, settlement amounts and reimbursement amounts from each medical insurance settlement data; based on the number of medical visits of each medical institution, the medical frequency of each medical institution node within a preset time period is calculated; based on the settlement amount and number of medical visits of each medical institution, the corresponding per capita medical expenses are calculated; based on the settlement amount and reimbursement amount of each medical institution, the corresponding reimbursement ratio is calculated; and the frequency of medical visits, per capita medical expenses and reimbursement ratio of each medical institution are weighted and summed up to obtain the corresponding collusion risk score.

[0092] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0093] The present application also discloses an electronic device. Figure 3 , Figure 3 The electronic device 300 may include: at least one processor 301 , at least one network interface 304 , a user interface 303 , a memory 305 , and at least one communication bus 302 .

[0094] The communication bus 302 is used to realize the connection and communication between these components.

[0095] The user interface 303 may include a display interface and a camera interface. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0096] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0097] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface diagrams and applications, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0098] Among them, the memory 305 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may optionally also be at least one storage device located away from the aforementioned processor 301. Refer to Figure 3 , the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program for a medical insurance data verification method based on big data.

[0099] exist Figure 3In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 301 can be used to call the memory 305 to store an application program for a medical insurance data verification method based on big data. When executed by one or more processors 301, the electronic device 300 executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0100] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0101] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0102] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0103] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0104] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0105] The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0106] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not recorded in the present disclosure. The description and examples are to be regarded as exemplary only.

Claims

1. A medical insurance data verification method based on big data, characterized in that: include: Obtain location information, qualification information, and medical information of insured persons of multiple medical institutions in the target area; Based on the location information and qualification information of each of the medical institutions, a spatial association network of medical institutions in the target area is constructed, and based on the medical information of each of the insured persons, a secondary association network of insured persons and medical institutions in the target area is constructed; Calculate the benchmark medical radius corresponding to the medical institution spatial association network, and screen out the cross-institutional medical paths that exceed the benchmark medical radius in the insured person-medical institution secondary association network; Determine multiple medical institution nodes that meet preset frequent mutual visit conditions in the cross-institution medical treatment path, and obtain medical insurance settlement data of each of the medical institution nodes; Combine the medical insurance settlement data of each of the medical institution nodes to calculate the corresponding collusion risk score; The medical institution node whose collusion risk score is greater than a preset threshold is taken as the target node, and a medical insurance data verification report corresponding to the target node and the insured person group associated with the target node is generated.

2. The medical insurance data verification method based on big data according to claim 1 is characterized in that: The step of constructing a spatial association network of medical institutions in the target area based on the location information and qualification information of each medical institution includes: Based on the location information of each of the medical institutions, determining the spatial distance between any two of the medical institutions in the target area; Taking any two medical institutions whose spatial distance is less than a distance threshold as candidate associated institutions; Based on a preset institution weight mapping table, determine the institution weight corresponding to the qualification information of each candidate associated institution, wherein the qualification information includes the medical institution grade, the number of departments opened, and the number of practicing physicians; The candidate associated institution is used as the first network node, the institution weight is used as the node weight, and the spatial distance is used as the edge weight to construct a spatial association network of medical institutions in the target area.

3. The medical insurance data verification method based on big data according to claim 1 is characterized in that: The step of constructing a secondary association network of insured persons and medical institutions in the target area according to the medical information of each insured person includes: Extracting the medical treatment time, medical institution and medical expenses corresponding to each insured person from the medical treatment information of each insured person; For each of the insured persons, based on the time sequence of the medical treatment time, the flow path of the insured person between different medical institutions is used as the insured person's medical treatment path; Counting the number of repetitions of the medical treatment paths of each insured person within a preset time window, and determining the medical treatment paths of the insured persons whose repetitions are greater than a frequency threshold as frequent medical treatment paths; The insured person node and the medical institution node involved in the frequent medical treatment path are used as second network nodes, and the frequent medical treatment path is used as a connecting edge between nodes; Determine a first weight corresponding to each of the insured person nodes based on each of the medical expenses, and determine a second weight corresponding to each of the medical institution nodes based on the frequent medical treatment paths; By combining each of the second network nodes, the first weight and the second weight corresponding to each of the second network nodes, and each of the connecting edges, a secondary association network of insured persons and medical institutions in the target area is constructed.

4. The medical insurance data verification method based on big data according to claim 3 is characterized in that: The determining of the first weight corresponding to each of the insured person nodes based on each of the medical expenses, and the determining of the second weight corresponding to each of the medical institution nodes based on the frequent medical treatment paths, includes: For any insured person involved in the frequent medical treatment path, determining a first weight of the corresponding insured person node based on a ratio between the medical expenses corresponding to the insured person and the total medical expenses of all the insured persons; For any medical institution involved in the frequent medical treatment path, obtaining the number of times any medical institution appears as an intermediate node in the frequent medical treatment path; The number of times any medical institution appears as an intermediate node in the frequent medical treatment path is divided by the total number of times all medical institutions in the target area appear as intermediate nodes to obtain the second weight of the corresponding medical institution node.

5. The medical insurance data verification method based on big data according to claim 1 is characterized in that: The calculating of the reference medical radius corresponding to the medical institution spatial association network includes: Obtaining the geographic coordinates and institution weight of each medical institution in the medical institution spatial association network; For any of the medical institutions, the number of insured persons within the preset radiation range of the medical institution and the ratio of the number of insured persons to the total number of insured persons of each of the medical institutions are obtained; Taking the medical institution as the center of the circle, gradually expand the search radius until the insurance participation ratio reaches the coverage threshold, and determine the corresponding search radius as the effective service radius of the medical institution; Divide each of the medical institutions into a plurality of institution groups corresponding to different institution weights, and calculate the mean effective service radius of the institution group corresponding to each of the institution weights based on the effective service radius of each of the medical institutions; Based on the proportion of the number of nodes in the medical institution spatial association network of the institution group corresponding to the weight of each institution, a weighted calculation is performed on the mean of each effective service radius to obtain a benchmark medical radius.

6. The medical insurance data verification method based on big data according to claim 1 is characterized in that: The step of determining a plurality of medical institution nodes in the cross-institution medical treatment path that meet preset frequent mutual visit conditions includes: Obtaining the insured person flow data between the medical institutions in the cross-institution medical treatment path, and calculating the insured person overlap between any two of the medical institutions based on the insured person flow data; Determine any two medical institutions whose insured person overlap degree is greater than an overlap degree threshold as frequently visited node pairs, and construct a medical institution association subgraph based on the frequently visited node pairs; The strongly connected components in the medical institution associated subgraph are identified through a preset graph traversal algorithm, and the medical institution nodes in the strongly connected components are determined as medical institution nodes that meet the preset frequent mutual visit conditions.

7. The medical insurance data verification method based on big data according to claim 1 is characterized in that: The combining of the medical insurance settlement data of each of the medical institution nodes to calculate the corresponding collusion risk score includes: Extracting corresponding medical visits, settlement amounts and reimbursement amounts from each of the medical insurance settlement data; Based on the number of visits to each of the medical institutions, the frequency of visits to each of the medical institution nodes within a preset time period is calculated; Calculate the corresponding per capita medical expenses based on the settlement amount and number of medical visits of each medical institution; Calculate the corresponding reimbursement ratio based on the settlement amount and reimbursement amount of each medical institution; The frequency of medical treatment, per capita medical expenses and reimbursement ratio of each medical institution are weighted and summed to obtain the corresponding collusion risk score.

8. A medical insurance data verification system based on big data, characterized in that: The system comprises: The information acquisition module is used to obtain the location information, qualification information and medical information of multiple medical institutions in the target area as well as the medical information of the insured persons; An association network determination module is used to construct a spatial association network of medical institutions in the target area based on the location information and qualification information of each medical institution, and to construct a secondary association network of insured persons and medical institutions in the target area based on the medical information of each insured person; A path determination module is used to calculate the reference medical radius corresponding to the medical institution spatial association network, and screen out the cross-institution medical paths that exceed the reference medical radius in the insured person-medical institution secondary association network; The data verification module is used to determine multiple medical institution nodes that meet the preset frequent mutual visit conditions in the cross-institutional medical treatment path, and obtain the medical insurance settlement data of each of the medical institution nodes; calculate the corresponding collusion risk score in combination with the medical insurance settlement data of each of the medical institution nodes; take the medical institution node whose collusion risk score is greater than a preset threshold as the target node, and generate a medical insurance data verification report corresponding to the target node and the insured person group associated with the target node.

9. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the medical insurance data verification method based on big data as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the medical insurance data verification method based on big data as described in any one of claims 1 to 7 is executed.

Images