An automatic extraction algorithm for abnormal appointment registration records based on data analysis

Abstract

The present invention discloses an automatic extraction algorithm for abnormal appointment registration records based on data analysis, which relates to the technical field of record extraction. After pre-processing patient flow and cost-effectiveness, an impact weight is generated for each supervision area, and abnormal behavior data is substituted into a linear regression model. The linear regression model outputs an abnormal factor for the registration account. The registration account is divided into a normal account, a management account, and an abnormal account based on the abnormal factor. After regularly obtaining the number of normal accounts, management accounts, and abnormal accounts recorded in each supervision area, an attention value is generated for each supervision area. After combining the impact weights of all supervision areas with the attention value for analysis, it is evaluated whether it is necessary to conduct overall management of the hospital in advance. The weight-raising algorithm improves the efficiency and accuracy of abnormal registration behavior management through the comprehensive application of data analysis and dynamic feedback mechanism, and optimizes the allocation of hospital resources and the formulation of management strategies.

Description

Technical Field

[0001] The present invention relates to the technical field of record extraction, and in particular to an algorithm for automatically extracting abnormal appointment registration records based on data analysis. Background Art

[0002] Abnormal appointment registration refers to the phenomenon that patients or agents obtain medical resources through various means without following normal procedures or regulations in the medical appointment registration system. The abnormal appointment registration phenomenon not only aggravates the unfairness of access to medical resources among patients, but may also lead to a decline in the actual efficiency of medical services and a waste of some resources. In addition, this phenomenon undermines the public's trust in the medical system and brings additional pressure to hospital management.

[0003] The prior art has the following deficiencies:

[0004] 1. Traditional registration systems are generally managed on a global basis, lacking fine-grained analysis and management of specific supervision areas. This method is difficult to cope with the differences in resource allocation, patient needs, abnormal behaviors, etc. between different regions, and it is impossible to accurately formulate personalized management strategies based on the characteristics of each region;

[0005] 2. Traditional systems often manage different areas in a static manner and lack dynamic feedback and adjustment mechanisms. When patient traffic and abnormal registration behaviors occur frequently, it is difficult for the system to adjust strategies in time to cope with sudden changes;

[0006] 3. The registration system lacks effective indicators to measure the contribution and importance of different areas to the overall operation of the hospital or the management of abnormal registration. In this case, it is difficult for the system to give priority to areas that have a greater impact on the hospital, resulting in unreasonable resource allocation.

[0007] Based on this, the present invention proposes an automatic extraction algorithm for abnormal appointment registration records based on data analysis. Through the comprehensive application of data analysis and dynamic feedback mechanism, the scheme improves the efficiency and accuracy of abnormal registration behavior management, and optimizes the allocation of hospital resources and the formulation of management strategies. Summary of the invention

[0008] The purpose of the present invention is to provide an automatic extraction algorithm for abnormal appointment registration records based on data analysis to solve the shortcomings of the background technology.

[0009] In order to achieve the above object, the present invention provides the following technical solution: an automatic extraction algorithm for abnormal appointment registration records based on data analysis, the extraction algorithm comprising the following steps:

[0010] The processing end generates several supervision areas based on the hospital department information, obtains the historical patient flow and cost-effectiveness of the supervision area, and generates an impact weight for each supervision area after pre-processing the patient flow and cost-effectiveness;

[0011] In the real-time registration process, the collection end obtains the abnormal behavior data of each registered account, substitutes the abnormal behavior data into the linear regression model, and the linear regression model outputs the abnormal factor for the registered account. Based on the abnormal factor, the registered account is divided into normal account, management account and abnormal account, and the corresponding management strategy is generated for the registered account according to the division result;

[0012] After regularly obtaining records of the number of normal accounts, management accounts, and abnormal accounts in each supervision area, an attention value is generated for each supervision area. After combining the impact weights of all supervision areas with the attention value for analysis, it is evaluated whether overall management of the hospital is needed in advance.

[0013] In a preferred embodiment, after regularly obtaining the number of normal accounts, management accounts, and abnormal accounts recorded in each supervision area, generating a concern value for each supervision area includes the following steps:

[0014] After regularly obtaining the number of normal accounts, management accounts, and abnormal accounts recorded in each supervision area, an attention value is generated for each supervision area. The expression is:

[0015] , where is the attention value, is the normal account number, is the number of abnormal accounts, To manage the number of accounts;

[0016] Sort all supervision areas according to the attention value from large to small, generate an area list, and send the area list to the hospital management platform.

[0017] In a preferred embodiment, after analyzing the impact weights of all supervision areas in combination with the attention values, evaluating whether it is necessary to conduct overall management of the hospital in advance includes the following steps:

[0018] The influence weights and attention values ​​of all supervision areas are calculated comprehensively to obtain the overall registration index, which is expressed as: , where is the overall registration index, For the The influence weight of each supervised region, For the The attention value of the supervised area, To monitor the number of areas;

[0019] If the overall registration index is less than the preset index threshold, the evaluation does not require overall management of the hospital in advance. If the overall registration index is greater than or equal to the preset index threshold, the evaluation requires overall management of the hospital in advance.

[0020] In a preferred embodiment, obtaining the historical patient flow and cost-effectiveness of the supervision area, pre-processing the patient flow and cost-effectiveness, and generating an impact weight for each supervision area includes the following steps:

[0021] Obtain the historical patient flow and cost-effectiveness of the supervised area, normalize the patient flow and cost-effectiveness, map the value range of patient flow and cost-effectiveness to [0,1], obtain the normalized patient flow value and the normalized cost-effectiveness value, sum the normalized patient flow value and the normalized cost-effectiveness value to obtain the influence value of the supervised area, sum the influence values ​​of all supervised areas to obtain the total influence value, and obtain the influence weight by dividing the influence value by the total influence value.

[0022] In a preferred embodiment, the abnormal behavior data is substituted into a linear regression model, and the linear regression model outputs an abnormal factor for the registered account, including the following steps:

[0023] Abnormal behavior data include abnormal appointment frequency, number of appointments with the same device, no-consultation rate, and multi-department registration appointment rate;

[0024] Substitute the abnormal appointment frequency, the number of appointments with the same device, the no-consultation rate, and the multi-department registration appointment rate into the linear regression model. The linear regression model outputs the abnormal factor for the registration account. The model expression is: , where is the abnormal factor, is a constant term with a value of 0.142, which indicates the influence of the secondary influencing data on the abnormal factor. , , , They are abnormal appointment frequency, number of appointments with the same device, no-consultation rate, and multi-department registration rate. , , , is the regression coefficient, and the regression coefficient is greater than 0.

[0025] In a preferred embodiment, the registered accounts are divided into normal accounts, management accounts and abnormal accounts based on abnormal factors, including the following steps:

[0026] Compare the acquired abnormal factor with a preset first abnormal threshold and a preset second abnormal threshold, wherein the first abnormal threshold is used to analyze whether there is an abnormality in the registered account, and the second abnormal threshold is used to analyze the severity of the abnormality in the registered account;

[0027] If the abnormal factor is less than or equal to the first abnormal threshold, the registered account is analyzed to be normal and the registered account is classified as a normal account;

[0028] If the abnormal factor is greater than the first abnormal threshold and the abnormal factor is less than or equal to the second abnormal threshold, the registered account is analyzed to have a slight abnormality and the registered account is classified as a management account;

[0029] If the abnormal factor is greater than the second abnormal threshold, the registered account is analyzed to have severe abnormalities and the registered account is classified as an abnormal account.

[0030] In a preferred embodiment, generating a corresponding management strategy for the registered account number according to the division result includes the following steps:

[0031] Registration accounts classified as normal accounts do not need to be managed. The management strategy generated for registration accounts classified as abnormal accounts is: block abnormal accounts, and the blocking duration is set by the hospital. The management strategy generated for registration accounts classified as management accounts is: re-correct the monitoring frequency of management accounts through abnormal factors. The expression is:

[0032] , where is the corrected monitoring frequency, is the monitoring frequency before correction, The abnormal factor is the larger the abnormal factor of the management account, the higher the monitoring frequency needs to be.

[0033] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0034] After preprocessing the patient flow and cost-effectiveness, the present invention generates an impact weight for each supervision area, substitutes the abnormal behavior data into the linear regression model, and the linear regression model outputs an abnormal factor for the registration account. The registration account is divided into normal account, management account and abnormal account based on the abnormal factor. After regularly obtaining the number of normal accounts, management accounts and abnormal accounts recorded in each supervision area, an attention value is generated for each supervision area. After combining the impact weights of all supervision areas with the attention value for analysis, it is evaluated whether it is necessary to manage the hospital in advance as a whole. The weight-raising algorithm improves the efficiency and accuracy of abnormal registration behavior management through the comprehensive application of data analysis and dynamic feedback mechanism, and optimizes the allocation of hospital resources and the formulation of management strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0036] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Example 1: Please refer to Figure 1 As shown, the automatic extraction algorithm of abnormal appointment registration records based on data analysis described in this embodiment includes the following steps:

[0039] The processing end generates several supervision areas based on the hospital department information, obtains the historical patient flow and cost-effectiveness of the supervision area, and generates an impact weight for each supervision area after pre-processing the patient flow and cost-effectiveness. In the real-time registration process, the collection end obtains the abnormal behavior data of each registration account, substitutes the abnormal behavior data into the linear regression model, and the linear regression model outputs an abnormal factor for the registration account. The registration account is divided into normal accounts, management accounts and abnormal accounts based on the abnormal factor, and corresponding management strategies are generated for the registration account based on the division results. After regularly obtaining the number of normal accounts, management accounts and abnormal accounts recorded in each supervision area, an attention value is generated for each supervision area. After combining the impact weights of all supervision areas with the attention values ​​for analysis, it is evaluated whether it is necessary to conduct overall management of the hospital in advance.

[0040] This application generates an impact weight for each supervision area after preprocessing patient flow and cost-effectiveness, substitutes abnormal behavior data into a linear regression model, and the linear regression model outputs an abnormal factor for the registration account. The registration account is divided into normal accounts, management accounts, and abnormal accounts based on the abnormal factor. After regularly obtaining the number of normal accounts, management accounts, and abnormal accounts in each supervision area, an attention value is generated for each supervision area. After combining the impact weights of all supervision areas with the attention value for analysis, it is evaluated whether it is necessary to manage the hospital as a whole in advance. The weight-raising algorithm improves the efficiency and accuracy of abnormal registration behavior management through the comprehensive application of data analysis and dynamic feedback mechanism, and optimizes the allocation of hospital resources and the formulation of management strategies.

[0041] Embodiment 2: The processing end generates a plurality of supervision areas according to the department information of the hospital, including the following steps:

[0042] Determine the data source of department information, which usually includes the hospital information management system (HIS), electronic medical record system (EMR), department management system, etc. These systems can provide comprehensive department-related data to ensure the authority and accuracy of the data.

[0043] For each department, the processing end needs to collect and organize the following key information:

[0044] Department Name: Specify the name of the department to ensure that each department can be uniquely identified.

[0045] Department number: Collect the department number corresponding to the hospital's internal system to facilitate subsequent data processing and association.

[0046] Department type: Distinguish department types, such as internal medicine, surgery, emergency department, pediatrics, etc., to provide a basis for subsequent analysis.

[0047] Department size: information such as the number of beds and medical equipment in the department, used to measure the resource utilization of the department.

[0048] Medical staff configuration: including the number and qualifications of doctors, nurses, technicians, etc., to help evaluate the department's service capabilities.

[0049] Patient flow: The department’s historical patient flow data, including outpatient volume, hospitalization volume, and surgical volume, is used to measure the department’s workload.

[0050] Department benefit data: Collect economic data such as department revenue, cost and cost-effectiveness ratio to help evaluate the department's economic contribution.

[0051] Equipment and resource usage: the number and utilization rate of key medical equipment and diagnosis and treatment resources used by the department.

[0052] Since hospital data may come from different systems, the format and content of the data may be inconsistent, so the collected data needs to be cleaned and standardized:

[0053] Data integrity check: Check the collected data to ensure that the key information of all departments is complete. If there is any missing or incomplete data, it is necessary to check and supplement with the corresponding data source or department.

[0054] Duplicate data processing: Ensure that there are no duplicate entries in department data. If a department's data is repeatedly collected from multiple sources, data merging is required to ensure the uniqueness of the final records.

[0055] Data format standardization: unify the data formats of different fields, such as date format, numerical units, naming rules, etc., to ensure the consistency of data in subsequent processing.

[0056] Store the processed department information data in the database and establish a data update mechanism. Since the configuration of hospital departments may change dynamically (such as department mergers, separations, expansions, etc.), the processing end needs to set up a regular data update mechanism to ensure the timeliness and accuracy of department data.

[0057] Through the step of collecting and collating department information, the processing end can generate independent supervision areas for each department and provide reliable basic data. The focus of this link is to ensure the comprehensiveness, accuracy and standardization of department information, laying a data foundation for the dynamic monitoring and management of subsequent supervision areas.

[0058] Obtain the historical patient flow and cost-effectiveness of the supervision area, pre-process the patient flow and cost-effectiveness, and generate an impact weight for each supervision area, including the following steps:

[0059] Obtain the historical patient flow and cost-effectiveness of the supervised area, normalize the patient flow and cost-effectiveness, map the value range of patient flow and cost-effectiveness to [0,1], obtain the normalized patient flow value and the normalized cost-effectiveness value, sum the normalized patient flow value and the normalized cost-effectiveness value to obtain the influence value of the supervised area, sum the influence values ​​of all supervised areas to obtain the total influence value, and obtain the influence weight by dividing the influence value by the total influence value. The larger the influence weight, the greater the importance of the department relative to other departments.

[0060] Get the patient flow history of the department

[0061] Patient flow data usually comes from the hospital's information management system (HIS) and electronic medical record system (EMR). Based on the patient registration, medical consultation, hospitalization, surgery and other information recorded in the system, the historical patient flow of the department can be queried and summarized by time dimension. The specific method is as follows:

[0062] Outpatient flow:

[0063] Registration system data: Through the hospital's registration system or HIS, query the historical registration records of each department and count the number of registrations to obtain the flow of outpatient patients.

[0064] Treatment records: Extract the historical outpatient treatment records of each department from the electronic medical record system and summarize the number of patients actually seen by time period. Treatment records can more accurately reflect the outpatient workload of the department.

[0065] Inpatient flow:

[0066] Inpatient system data: Inpatient patient traffic can be obtained through the inpatient management system. Specific operations include statistics on the historical admission, discharge, and transfer records of each department, and calculation of the number of inpatients per day, month, or year.

[0067] Bed utilization rate: Querying the utilization of department beds can further understand the flow of inpatients and their demand for department resources.

[0068] Surgical patient flow:

[0069] Surgery management system data: If the department has a surgery scheduled, the surgery management system can be used to obtain the surgery records performed by the department and count the number and type of patients undergoing surgery.

[0070] Operating room usage: Querying the operating room reservation and usage records can further reflect the flow pressure of the surgical department.

[0071] Emergency patient flow:

[0072] Emergency management data: The flow of emergency patients can be obtained through the emergency management module. According to the registration and treatment records of emergency patients, the flow of emergency patients in each department (such as the emergency department) is counted.

[0073] The patient flow is obtained by adding the outpatient flow, inpatient flow, surgical patient flow and emergency patient flow.

[0074] Cost-effectiveness data mainly comes from the hospital's financial system (ERP) and department operation data. By analyzing the income and expenditure of the department, the cost-effectiveness ratio can be calculated. The specific methods for obtaining cost-effectiveness are as follows:

[0075] Billing system data: Obtain departmental billing records through the hospital billing system. Specifically, it includes outpatient registration fees, hospitalization fees, surgery fees, drug fees, examination and testing fees, etc. By summarizing these income sources by time period, the total income of the department can be obtained.

[0076] Medical insurance and self-payment settlement data: Combine the medical insurance and self-payment settlement systems to obtain the income structure of different payment methods to fully reflect the income situation of the department.

[0077] Human resource cost: Obtain the human resource expenditure data of the department from the hospital's human resource system, including the salary, benefits, bonuses, etc. of doctors, nurses and other staff. This is an important part of department expenditure.

[0078] Equipment depreciation and maintenance: Obtain equipment purchase, depreciation and daily maintenance costs related to departments in the hospital fixed asset management system. High-tech medical equipment has high maintenance costs, and equipment depreciation and maintenance costs are also part of department expenses.

[0079] Consumables and drug costs: The costs of medical consumables and drugs used by departments during diagnosis and treatment can be obtained from the material management system or drug supply chain management system. Consumption is summarized by department to calculate the relevant drug and consumable costs.

[0080] Operation and management costs: including the daily operating costs of the department (such as water, electricity, office supplies, etc.) and the management fees allocated by the hospital. They can be queried through the financial system and the total income of the supervised area divided by the total cost can be used to obtain cost-effectiveness.

[0081] The greater the patient flow, the greater the cost-effectiveness, which usually indicates that the supervision area (department) is more important relative to the hospital. The reasons can be analyzed from multiple dimensions:

[0082] High demand means high value: Departments with large patient flow usually mean high demand for medical services. Such departments may undertake more outpatient, inpatient, emergency or surgical tasks, which directly reflects the importance of the department in hospital operations. The greater the patient flow, the wider the patient group served by the department, and the greater the contribution of the department's medical services to the overall operation of the hospital.

[0083] Patient satisfaction affects hospital reputation: departments with high traffic volume are often the core medical departments of the hospital, such as the emergency department and surgery department. The service efficiency and quality of these departments directly affect patient satisfaction. If high-traffic departments are well operated, they can enhance the reputation of the hospital and attract more patients, so they are crucial to the development of the hospital.

[0084] Revenue is proportional to traffic: Generally speaking, departments with large patient traffic will generate more revenue, including outpatient expenses, hospitalization expenses, surgical expenses, examination and treatment expenses, etc. The more patients there are, the greater the direct economic benefits generated by the hospital through these departments.

[0085] Cost dilution effect: Under high-volume conditions, departments can make better use of their fixed resources, such as medical equipment, medical staff, beds, etc. Since the cost of these resources is fixed, as the number of patients increases, the average cost per patient will decrease, improving the overall cost-effectiveness ratio.

[0086] Profit feedback: Cost-effective departments can not only bring economic benefits to the hospital, but also provide additional financial support for the hospital to subsidize other non-profit departments (such as scientific research departments, social welfare departments, etc.) or for hospital equipment updates, infrastructure renovation, etc. Such departments can improve the overall financial status of the hospital and ensure the normal operation of other departments.

[0087] Resource investment and returns: Hospitals may prioritize investing more resources in cost-effective departments, such as better equipment and better medical personnel, because these departments can generate greater returns through high patient flow and high efficiency, thereby promoting the improvement of the overall operational efficiency and competitiveness of the hospital.

[0088] Management efficiency: Cost-effective departments usually have high operational efficiency in resource management, patient process management, etc. Effective department operation can reduce operating costs and improve efficiency while ensuring the quality of medical services. Such efficient departments are often the benchmarks in hospital management and can drive the improvement of the overall operation level of the hospital.

[0089] Technological advantages: Departments with high cost-effectiveness usually have strong technical strength and can attract more patients through high-level diagnosis and treatment technologies, especially departments involving complex treatments and high technical requirements, such as cardiac surgery and oncology. Technological advantages can not only improve cost-effectiveness, but also help hospitals establish technological brands and further enhance the competitiveness of hospitals.

[0090] Strategic significance: Some departments, such as the emergency department and ICU, have a strategic position due to their large patient flow and heavy treatment tasks. Even if these departments are not effective in the short term, they are crucial from the perspective of the overall development of the hospital. High-flow departments are of great strategic significance to the hospital's sense of social responsibility and medical service guarantee, and their healthy operation also contributes to the hospital's reputation in public medical services.

[0091] Innovation and development potential: Departments with high patient flow and high efficiency tend to have greater potential for scientific research, teaching, and innovation, and can promote the advancement of hospital medical technology. These departments have advantages in attracting excellent medical staff, obtaining scientific research funds, and developing new technologies, which promotes the long-term development of the hospital.

[0092] Basis for resource allocation: Hospitals usually allocate resources based on patient flow and cost-effectiveness. Departments with large flow mean that they have a higher demand for overall hospital resources. To ensure the normal operation of these high-flow, high-efficiency departments, hospitals may give priority to allocating medical equipment, funds, medical staff and other resources to them, which reflects their relative importance to the hospital.

[0093] Strategic decision-making impact: In the strategic decision-making of the hospital, departments with large traffic and high efficiency will often become the key support objects for the hospital's development. The hospital's long-term planning, discipline construction and resource allocation will give priority to the importance of these departments to ensure that they can continuously and stably contribute to the hospital's economic benefits and medical service quality.

[0094] In summary: supervision areas (departments) with large patient flow and high cost-effectiveness play a key role in hospital operations. They not only provide economic support for hospitals, but also make vital contributions in multiple dimensions such as medical services, brand building, and technological innovation. Therefore, flow and benefit indicators can effectively measure the importance of departments to hospitals and help hospitals make optimized decisions in resource allocation, strategic planning, etc.

[0095] In the real-time registration process, the collection end obtains the abnormal behavior data of each registered account, substitutes the abnormal behavior data into the linear regression model, and the linear regression model outputs the abnormal factor for the registered account, including the following steps:

[0096] Abnormal behavior data include abnormal appointment frequency, number of appointments with the same device, no-consultation rate, and multi-department registration appointment rate;

[0097] Substitute the abnormal appointment frequency, the number of appointments with the same device, the no-consultation rate, and the multi-department registration appointment rate into the linear regression model. The linear regression model outputs the abnormal factor for the registration account. The model expression is: , where is the abnormal factor, is a constant term with a value of 0.142, which indicates the influence of secondary influencing data on abnormal factors (secondary influencing data include continuous login time in the registration system and frequent account logouts and logins to the registration system, etc.). , , , They are abnormal appointment frequency, number of appointments with the same device, no-consultation rate, and multi-department registration rate. , , , is the regression coefficient, and the regression coefficient is greater than 0;

[0098] 1) Abnormal appointment frequency

[0099] Abnormal appointment frequency refers to the frequency of abnormal appointment behavior in a registered account within a specific time period, usually manifested as excessive number of registrations, abnormal time period, etc.

[0100] Registration system log: Extract the appointment time and number of appointments for each account through the registration system log. Combined with the set abnormal standards (such as repeated appointments in a short period of time, frequent appointments late at night, etc.), mark these behaviors as abnormal appointments.

[0101] Time window analysis: Set a reasonable time window (such as one day or one week) and count the appointment frequency of each account within the time window. If it exceeds the set threshold (for example, more than 5 registrations within 24 hours), the account can be marked as abnormal and its abnormal appointment frequency can be counted.

[0102] Abnormal pattern recognition: Using data mining technology, we analyze the time, department and appointment method of registration behavior to identify possible abnormal patterns. For example, if an account frequently uses the same device to make appointments in different time periods, it can be considered abnormal.

[0103] 2) Number of reservations for the same device

[0104] The number of reservations made on the same device refers to the number of reservations made by a certain account on the same device (such as a computer or mobile phone). This data is used to identify whether multiple accounts have made reservations through the same device, which may be an abnormal reservation behavior.

[0105] Device information acquisition: Each time you register, the device information used for the reservation (such as the device unique identifier UUID, IP address, MAC address, device type, etc.) is recorded. This information can be obtained through the registration system's network log or client application.

[0106] Account association analysis: Perform association analysis on device identifiers to check whether multiple accounts use the same device to register. If a device frequently registers multiple different accounts in a short period of time, the number of appointments for the device can be counted and further analyzed to see if there is any abnormal behavior.

[0107] Device frequency statistics: Count the number of registrations for each device, arrange the reservations for the same device in chronological order, and calculate the number of registrations for a specific account. According to the set threshold, if the same device has too many reservations for a certain account, it can be marked as a suspicious account.

[0108] 3) Rate of not seeking medical treatment

[0109] The non-treatment rate refers to the situation where patients fail to see a doctor on time after registration (i.e. the proportion of patients who register but do not receive actual treatment at the scheduled time). This indicator can reflect whether the registration account has maliciously occupied registration resources.

[0110] Comparison of registration and medical treatment data: Obtain the registration records of each account through the registration system, and obtain the corresponding medical records from the electronic medical record system (EMR) or HIS, and match the two. Accounts with registration records but no corresponding medical records are considered as non-medical treatment behavior.

[0111] Count the non-visit rate: within a certain period of time (such as one month or three months), count the number of registrations and actual visits for each account, and calculate the non-visit rate by dividing the number of non-visits by the total number of registrations;

[0112] Focus on analyzing accounts with a high frequency of not visiting the hospital: For accounts with a high rate of not visiting the hospital, further analyze whether they have the behavior of continuously registering but not visiting the hospital, especially accounts that repeatedly register in popular departments but do not visit the hospital, which may be malicious account-swiping behavior.

[0113] 4) Multi-department registration appointment rate

[0114] The multi-department registration rate refers to the proportion of an account that makes registrations for multiple departments within a certain period of time (such as one day or one week). Usually, a normal patient will not frequently register for multiple departments in a short period of time. If there is a multi-department registration, it may reflect abnormal registration behavior.

[0115] Registration department records: The registration records of each account are extracted through the registration system, especially the name of the registration department and the registration time, and summarized and analyzed by time period (such as one day or one week).

[0116] Multi-department registration statistics: Count the number of departments registered by each account within a certain period of time. For example, if an account makes appointments in multiple departments in one day, such as internal medicine, surgery, and pediatrics, its multi-department registration behavior is recorded, and the multi-department registration appointment rate is obtained by dividing the number of multi-department registrations by the total number of registrations.

[0117] Analysis of abnormal department registration patterns: Combined with the patient's medical records, determine whether the account has reasonable medical needs (for example, whether there are multiple diseases that require treatment). If an account frequently registers in multiple departments but does not have corresponding medical needs, it can be considered that the account's multi-department registration behavior is abnormal.

[0118] The logical factors of the abnormal factors when used in the present invention are as follows: taking the influence of various variables on the normal registration of the department as an example, the first is the indicator, that is, the factor that causes the abnormal registration of the department (the present invention refers to the influence of various variables on the normal registration of the department); the second is the weight of these indicators, that is, the proportion of each variable when it is generated; the third is the operation equation, that is, what kind of mathematical operation process is used to obtain the result, and the abnormal factor is obtained by calculating the indicators with their respective weights through the operation equation.

[0119] The variables obtained from the sample were transformed and processed into data language recognized by computer software; secondly, these evaluation factors were analyzed by logistic regression using SPSS software to screen out factors and their weights that were significantly correlated with the results; thirdly, the evaluation factors and weights were brought into the logistic regression equation for calculation to obtain the results, which are as follows:

[0120] First, ensure the integrity of each variable, handle missing values ​​and outliers, and convert the data into a format that SPSS software can recognize. Usually, the data is stored in .csv, .xlsx and other formats, and then imported into SPSS. Open SPSS software, import the processed data file, and transform the variables as needed. For example, for continuous variables, standardize or normalize them. Select the "Analyze" menu, then select the "Binary Logistic" option under "Regression". In the dialog box, add the dependent variable (outcome) and independent variable (each variable) to the corresponding box. SPSS will fit the Logistic regression model based on the selected variables. In the output results, you will see the coefficients, standard errors, p-values ​​and other information of the model. Check the coefficients and p-values ​​in the output results to determine which variables are significantly correlated with the results. Usually, a p-value less than 0.05 is considered significant. While fitting the model, use variable selection methods, such as stepwise regression, to help screen the most relevant factors. According to the coefficients of the Logistic regression model, the size of the coefficients reflects the degree of influence of each factor on the result, and the positive and negative signs of the coefficients indicate the direction of the influence. After obtaining the significant factors and their coefficients, the Logistic regression equation is obtained, which is used to calculate the probability of each sample and then predict the results.

[0121] Based on the abnormal factors, the registered accounts are divided into normal accounts, management accounts and abnormal accounts, and corresponding management strategies are generated for the registered accounts according to the division results, including the following steps:

[0122] Compare the acquired abnormal factor with a preset first abnormal threshold and a preset second abnormal threshold, wherein the first abnormal threshold is used to analyze whether there is an abnormality in the registered account, and the second abnormal threshold is used to analyze the severity of the abnormality in the registered account;

[0123] If the abnormal factor is less than or equal to the first abnormal threshold, the registered account is analyzed to be normal and the registered account is classified as a normal account;

[0124] If the abnormal factor is greater than the first abnormal threshold and the abnormal factor is less than or equal to the second abnormal threshold, the registered account is analyzed to have a slight abnormality and the registered account is classified as a management account;

[0125] If the abnormal factor is greater than the second abnormal threshold, the registered account is analyzed to have severe abnormalities and the registered account is classified as an abnormal account.

[0126] Registration accounts classified as normal accounts do not need to be managed. The management strategy generated for registration accounts classified as abnormal accounts is: block the abnormal accounts, and the blocking period is set by the hospital (usually 7d-30d). The management strategy generated for registration accounts classified as management accounts is: re-correct the monitoring frequency of management accounts through abnormal factors. The expression is: , where is the corrected monitoring frequency, is the monitoring frequency before correction, The abnormal factor is the larger the abnormal factor of the management account, the higher the monitoring frequency needs to be.

[0127] After regularly obtaining the number of normal accounts, management accounts, and abnormal accounts recorded in each supervision area, an attention value is generated for each supervision area, including the following steps:

[0128] After regularly obtaining the number of normal accounts, management accounts, and abnormal accounts recorded in each supervision area, an attention value is generated for each supervision area. The expression is:

[0129] , where is the attention value, is the normal account number, is the number of abnormal accounts, To manage the number of accounts;

[0130] The larger the attention value, the more attention the supervision area needs. All supervision areas are sorted from large to small according to the attention value, and an area list is generated and sent to the hospital management platform.

[0131] After analyzing the impact weights and attention values ​​of all supervision areas, assess whether it is necessary to conduct overall management of the hospital in advance, including the following steps:

[0132] The influence weights and attention values ​​of all supervision areas are calculated comprehensively to obtain the overall registration index, which is expressed as: , where is the overall registration index, For the The influence weight of each supervised region, For the The attention value of the supervised area, To monitor the number of areas;

[0133] If the overall registration index is less than the preset index threshold, the evaluation does not need to conduct overall management of the hospital in advance. If the overall registration index is greater than or equal to the preset index threshold, the evaluation needs to conduct overall management of the hospital in advance, and the management includes:

[0134] Dynamically adjust registration quotas: Dynamically adjust daily or weekly registration quotas based on the patient flow and registration needs of each supervision area (department) to reduce congestion in popular departments.

[0135] Priority registration: Set up priority registration channels for emergency patients, patients with long-term chronic diseases or special groups (such as the elderly, children, etc.) to prevent these patients from being affected by abnormal registration behaviors.

[0136] Increase the shifts of doctors and nurses: In departments with an unusually high number of registrations, the hospital can deploy more doctors or medical staff, increase outpatient hours or extend working hours to cope with the abnormally increased registration needs.

[0137] Cross-department support: For some doctors who have cross-department diagnosis and treatment qualifications, cross-department support will be arranged to help alleviate the pressure on other departments.

[0138] Appointment restrictions: Set a limit on the number of appointments based on patient identity, account history, and other data. For users who frequently make appointments in multiple departments or the same department in a short period of time, their appointment permissions can be temporarily restricted.

[0139] IP address / device binding: limit the registration frequency of each device or IP address, and intercept abnormal registration behaviors of multiple accounts of the same device or IP address.

[0140] Multi-factor authentication: For areas where abnormal registrations occur, the complexity of identity authentication is increased, such as requiring registrants to provide SMS verification codes, health card numbers or identity documents for secondary verification.

[0141] Real-time anomaly detection: Through the real-time monitoring system, abnormal registration behavior can be identified, such as a large number of appointments for a specific department in a short period of time, and alarms can be issued in time and emergency measures can be taken.

[0142] Automatic blocking: The registration system should be equipped with an automatic blocking function. When abnormal behavior is detected, it will automatically suspend or limit the registration rights of the account to avoid further resource occupation.

[0143] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0144] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0145] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0146] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0147] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An automatic extraction algorithm for abnormal appointment registration records based on data analysis, characterized by: The extraction algorithm comprises the following steps: The processing end generates several supervision areas based on the hospital department information, obtains the historical patient flow and cost-effectiveness of the supervision area, and generates an impact weight for each supervision area after pre-processing the patient flow and cost-effectiveness; In the real-time registration process, the collection end obtains the abnormal behavior data of each registered account, substitutes the abnormal behavior data into the linear regression model, and the linear regression model outputs the abnormal factor for the registered account. Based on the abnormal factor, the registered account is divided into normal account, management account and abnormal account, and the corresponding management strategy is generated for the registered account according to the division result; After regularly obtaining the number of normal accounts, management accounts, and abnormal accounts in each supervision area, generate a focus value for each supervision area, and combine the impact weights of all supervision areas with the focus value for analysis to evaluate whether the hospital needs to be managed in advance; After regularly obtaining the number of normal accounts, management accounts, and abnormal accounts recorded in each supervision area, an attention value is generated for each supervision area, including the following steps: After regularly obtaining the number of normal accounts, management accounts, and abnormal accounts recorded in each supervision area, an attention value is generated for each supervision area. The expression is: , where is the attention value, is the normal account number, is the number of abnormal accounts, To manage the number of accounts; Sort all the supervision areas according to the attention value from large to small, generate an area list, and send the area list to the hospital management platform; After analyzing the impact weights and attention values ​​of all supervision areas, assess whether it is necessary to conduct overall management of the hospital in advance, including the following steps: The influence weights and attention values ​​of all supervision areas are calculated comprehensively to obtain the overall registration index, which is expressed as: , where is the overall registration index, For the The influence weight of each supervised region, For the The attention value of the supervised area, To monitor the number of areas; If the overall registration index is less than the preset index threshold, the evaluation does not require overall management of the hospital in advance; if the overall registration index is greater than or equal to the preset index threshold, the evaluation requires overall management of the hospital in advance; Obtain the historical patient flow and cost-effectiveness of the supervision area, pre-process the patient flow and cost-effectiveness, and generate an impact weight for each supervision area, including the following steps: Obtain the historical patient flow and cost-effectiveness of the supervised area, normalize the patient flow and cost-effectiveness, map the value range of the patient flow and cost-effectiveness to [0,1], obtain the normalized value of the patient flow and the normalized value of the cost-effectiveness, sum the normalized value of the patient flow and the normalized value of the cost-effectiveness to obtain the influence value of the supervised area, sum the influence values ​​of all supervised areas to obtain the total influence value, and obtain the influence weight by dividing the influence value by the total influence value; Substitute the abnormal behavior data into the linear regression model, and the linear regression model outputs the abnormal factor for the registered account, including the following steps: Abnormal behavior data include abnormal appointment frequency, number of appointments with the same device, no-consultation rate, and multi-department registration appointment rate; Substitute the abnormal appointment frequency, the number of appointments with the same device, the no-consultation rate, and the multi-department registration appointment rate into the linear regression model. The linear regression model outputs the abnormal factor for the registration account. The model expression is: , where is the abnormal factor, is a constant term with a value of 0.142, which indicates the influence of the secondary influencing data on the abnormal factor. , , , They are abnormal appointment frequency, number of appointments with the same device, no-consultation rate, and multi-department registration rate. , , , is the regression coefficient, and the regression coefficient is greater than 0.

2. According to claim 1, an automatic extraction algorithm for abnormal appointment registration records based on data analysis is characterized by: The registered accounts are divided into normal accounts, management accounts and abnormal accounts based on abnormal factors, including the following steps: Compare the acquired abnormal factor with a preset first abnormal threshold and a preset second abnormal threshold, wherein the first abnormal threshold is used to analyze whether there is an abnormality in the registered account, and the second abnormal threshold is used to analyze the severity of the abnormality in the registered account; If the abnormal factor is less than or equal to the first abnormal threshold, the registered account is analyzed to be normal and the registered account is classified as a normal account; If the abnormal factor is greater than the first abnormal threshold and the abnormal factor is less than or equal to the second abnormal threshold, the registered account is analyzed to have a slight abnormality and the registered account is classified as a management account; If the abnormal factor is greater than the second abnormal threshold, the registered account is analyzed to have severe abnormalities and the registered account is classified as an abnormal account.

3. The automatic extraction algorithm of abnormal appointment registration records based on data analysis according to claim 2 is characterized by: Generate a corresponding management strategy for the registered account according to the classification result, including the following steps: Registration accounts classified as normal accounts do not need to be managed. The management strategy generated for registration accounts classified as abnormal accounts is: block abnormal accounts, and the blocking duration is set by the hospital. The management strategy generated for registration accounts classified as management accounts is: re-correct the monitoring frequency of management accounts through abnormal factors. The expression is: , where is the corrected monitoring frequency, is the monitoring frequency before correction, The abnormal factor is the larger the abnormal factor of the management account, the higher the monitoring frequency needs to be.

Images