Hospital economic risk comprehensive assessment method and system

Abstract

The invention provides a hospital economic risk comprehensive assessment method and system, and relates to the field of data analysis, and the method comprises the steps: constructing an economic operation condition evaluation index and an early warning index according to a hospital economic operation mechanism; calculating the cloud association degree of the evaluation indexes through an optimal cloud entropy method; the comprehensive weight of each evaluation index is determined through a maximum entropy criterion improved interval analytic hierarchy process, and the hospital economic operation condition is evaluated by using the cloud correlation degree and the comprehensive weight to obtain an evaluation result; analyzing a time difference correlation relationship of the early warning indexes by using a time difference correlation analysis method, and constructing a hospital economic operation early warning model in combination with an evaluation result; real-time monitoring is carried out according to the hospital economic operation early warning model, an early warning signal is generated, and the risk of hospital economic operation is analyzed in combination with the actual condition of hospital economic operation. According to the invention, a systematic and scientific economic operation data analysis framework can be provided for hospitals, and the hospitals can make more intelligent decisions in the aspect of economic management.

Description

Technical Field

[0001] The present invention relates to the field of data analysis, and in particular to a comprehensive evaluation method and system for hospital economic risks. Background Art

[0002] Hospital economic operation data refers to various economic data generated during a hospital's daily operations, including but not limited to financial data, human resources data, drug and consumables data, and medical revenue. This data comprehensively reflects a hospital's economic status, operational efficiency, and management level, and serves as a key indicator of hospital operational quality. Hospital economic operation data not only covers daily operating costs, revenue sources, and resource allocation efficiency, but also provides a basis for assessing the hospital's future development potential.

[0003] In hospital management, the analysis and utilization of economic operation data is a key means of making informed decisions. Through in-depth analysis of this data, hospital managers can identify key cost drivers, optimize resource allocation, identify potential operational risks, and formulate appropriate improvement measures. Furthermore, economic operation data can assist decision-makers in assessing whether the current business model aligns with the hospital's long-term development strategy, providing a reference for future expansion and innovation. Leveraging this data analysis, hospitals can optimize business processes, improve economic efficiency, and maintain sustainable development amidst fierce market competition.

[0004] However, traditional hospital economic performance analysis often relies on expert experience and subjective judgment, lacking scientific quantitative methods, making it difficult to accurately assess overall economic conditions. Furthermore, some early warning indicators exhibit a time lag, meaning that changes in these indicators only impact economic conditions after a certain period of time. This lag is difficult to capture using traditional analysis.

[0005] Currently, no effective solutions have been proposed for the problems in related technologies. Summary of the Invention

[0006] In view of this, the present invention provides a hospital economic risk comprehensive assessment method and system to solve the above-mentioned problems.

[0007] In order to solve the above problems, the specific technical solutions adopted by the present invention are as follows: According to one aspect of the present invention, a comprehensive evaluation method for hospital economic risks is provided, which includes the following steps: S1. Collect hospital economic operation data and establish economic operation status evaluation indicators and early warning indicators based on the hospital economic operation mechanism; S2. Based on the evaluation indicators of the hospital's economic operation status, the cloud correlation of the evaluation indicators is calculated using the optimal cloud entropy method; the comprehensive weight of each evaluation indicator is determined by improving the interval analytic hierarchy process using the maximum entropy criterion, and the cloud correlation and comprehensive weight of the evaluation indicators are used to evaluate the hospital's economic operation status and obtain the evaluation results; S3. Use the time difference correlation analysis method to analyze the time difference correlation of early warning indicators, and build a hospital economic operation early warning model based on the evaluation results; S4. Conduct real-time monitoring based on the hospital economic operation early warning model, generate early warning signals, and analyze the risks of the hospital's economic operation based on the actual situation of the hospital's economic operation; Preferably, according to the hospital economic operation status evaluation index, the cloud correlation of the evaluation index is calculated by the optimal cloud entropy method; the comprehensive weight of each evaluation index is determined by improving the interval analytic hierarchy process by the maximum entropy criterion, and the cloud correlation and comprehensive weight of the evaluation index are used to evaluate the hospital economic operation status, and obtaining the evaluation result includes the following steps: S21. Determine the expected value and super entropy of each evaluation indicator, and use the optimal cloud entropy method to calculate the optimal cloud entropy of the evaluation indicator to obtain the optimal cloud entropy set; S22. Calculate the cloud correlation of each evaluation indicator based on the expected value and the optimal cloud entropy, and construct a cloud correlation matrix; S23. Determine the weight interval of each evaluation indicator using the interval analytic hierarchy process, and determine the comprehensive weight of the evaluation indicators using the maximum entropy criterion; S24. Using the cloud correlation and comprehensive weight of the evaluation indicators, the economic operation status of the hospital is evaluated according to the principle of maximum correlation to obtain the evaluation results.

[0008] Preferably, determining the expected value and super entropy of each evaluation indicator, and using the optimal cloud entropy method to calculate the optimal cloud entropy of the evaluation indicator to obtain the optimal cloud entropy set includes the following steps: S211. Based on the evaluation indicators of the hospital's economic operation status and in combination with historical data, configure a grade standard for each evaluation indicator; S212, calculating the average value of each evaluation indicator within the range of each grade standard to obtain the expected value; S213, performing statistical analysis on the evaluation index data of each evaluation index within the range of each grade standard, and determining the super entropy by calculating the degree of dispersion of the evaluation index data; S214. Calculate the primary cloud entropy value and the secondary cloud entropy value of each evaluation indicator respectively using the triple cloud entropy rule and the half-number correlation principle. Establish a nonlinear decision optimization model based on the primary cloud entropy value and the secondary cloud entropy value and solve it to obtain the optimal cloud entropy set.

[0009] Preferably, the calculation formulas for calculating the first-level cloud entropy value and the second-level cloud entropy value of each evaluation index using the triple cloud entropy rule and the half-correlation principle are respectively: E′ n =( g max - g min ) / 6; E′′ n =( g max - g min ) / 2.3548; Where, It represents the first-level cloud entropy value calculated using the triple cloud entropy rule; represents the secondary cloud entropy value calculated using the half-correlation principle; g max Indicates the upper limit of the evaluation index in the grade standard; g min Indicates the lower limit of the evaluation index in the grade standard.

[0010] Preferably, the cloud correlation degree of each evaluation index is calculated by combining the expected value of each evaluation index and the optimal cloud entropy as follows: ; Where, R ij represents the cloud correlation degree of evaluation index i on grade standard j; u i represents the actual observed value of the evaluation index; E x It indicates the expected value of the evaluation indicator under the corresponding grade standard; E ni It represents the optimal cloud entropy of the indicator under the corresponding grade standard.

[0011] Preferably, determining the weight interval of each evaluation index using the interval analytic hierarchy process and determining the comprehensive weight of the evaluation index using the maximum entropy criterion includes the following steps: S231. Compare the importance of each evaluation index pairwise and construct an interval number judgment matrix; S232. Calculate the weight interval of each evaluation index using the interval number eigenvector method according to the interval number judgment matrix; S233. The initial weight interval is used as a constraint condition of the maximum entropy criterion, and a nonlinear optimization model is constructed. By solving the nonlinear optimization model, the final weight value of each evaluation indicator is obtained.

[0012] Preferably, the time difference correlation analysis method is used to analyze the time difference correlation relationship of the early warning indicators, and the hospital economic operation early warning model is constructed in combination with the evaluation results; the following steps are included: S31. Collect historical time series indicator data of early warning indicators and calculate the benchmark indicators of early warning indicators through clustering algorithms; S32. Calculate the time difference correlation coefficient between the early warning indicator and the benchmark indicator using the time difference correlation analysis method based on the benchmark indicator of the early warning indicator; S33. Selecting early warning indicators that meet preset screening rules based on the time difference correlation coefficient between the early warning indicators and the benchmark indicators; S34. Based on the selected early warning indicators and combined with the evaluation results of the hospital's economic operation status, a hospital economic operation early warning model is constructed.

[0013] Preferably, collecting historical time series indicator data of the early warning indicator and calculating the benchmark indicator of the early warning indicator by a clustering algorithm includes the following steps: S311. Randomly select a number of data points from the historical indicator data of the early warning indicator as initial cluster centers; S312, respectively calculating the distance between each warning indicator data point and each initial cluster center, and assigning these data points to the cluster with the closest cluster center; S313. For each cluster, calculate the average value of all data points within it, and use this average value as the new cluster center; S314: Repeat steps S312 and S313 until a preset number of iterations is reached. For each early warning indicator, select the cluster center in the corresponding cluster as a benchmark indicator.

[0014] Preferably, based on the benchmark indicator of the early warning indicator, the calculation formula for the time difference correlation coefficient between the early warning indicator and the benchmark indicator using the time difference correlation analysis method is: ; Where, It represents the time difference correlation coefficient between the early warning indicator and the benchmark indicator; N Indicates the time length of the data in the early warning indicator; represents the mean of the benchmark indicator; Indicates the b Benchmark indicators at time t The value of represents the mean value of the early warning indicator; Indicates the b Early warning indicators at the moment tThe value of r Indicates the maximum delay number.

[0015] Preferably, based on the selected early warning indicators and in combination with the evaluation results of the hospital's economic operation status, constructing a hospital economic operation early warning model includes the following steps: S341. Combining the selected early warning indicators and the evaluation results of the hospital's economic operation status into a data set, and dividing the data set into a training set and a test set; S342. Using the selected early warning indicators as input and the evaluation results of the hospital's economic operation status as output, a linear regression method is used to construct and train an early warning model based on the training set; S343. Use the test set data to verify the trained early warning model, verify the prediction performance of the early warning model, and adjust the early warning model parameters according to the verification results to obtain the final early warning model.

[0016] According to another aspect of the present invention, a hospital economic risk comprehensive assessment system is provided, the hospital economic risk comprehensive assessment system comprising: an indicator construction module, an economic status assessment module, an early warning model construction module, and an economic status monitoring module, wherein the indicator construction module, the economic status assessment module, the early warning model construction module, and the economic status monitoring module are sequentially connected; The indicator construction module is used to collect hospital economic operation data and construct economic operation status evaluation indicators and early warning indicators based on the hospital economic operation mechanism; The economic status evaluation module is used to calculate the cloud correlation of the evaluation indicators based on the hospital's economic operation status evaluation indicators through the optimal cloud entropy method; determine the comprehensive weight of each evaluation indicator through the interval analytic hierarchy process improved by the maximum entropy criterion, and use the cloud correlation and comprehensive weight of the evaluation indicators to evaluate the hospital's economic operation status and obtain the evaluation results; The early warning model construction module is used to analyze the time difference correlation of early warning indicators using the time difference correlation analysis method, and to build a hospital economic operation early warning model based on the evaluation results; The economic status monitoring module is used to conduct real-time monitoring based on the hospital economic operation early warning model, generate early warning signals, and analyze the risks of the hospital's economic operation in combination with the actual situation of the hospital's economic operation.

[0017] The beneficial effects of the present invention are: 1. The present invention collects various types of data on the economic operation of the hospital and constructs comprehensive evaluation indicators and early warning indicators, which can more comprehensively reflect the economic health of the hospital, improve the hospital's economic management level, and help the hospital better understand and control the economic operation status. Through data analysis, the hospital can allocate resources more reasonably and improve resource utilization efficiency. The early warning model can help the hospital to timely discover and respond to economic risks, enhance the hospital's risk control capabilities, and provide a systematic and scientific economic operation data analysis framework for the hospital, which helps the hospital make more informed decisions in economic management.

[0018] 2. The present invention uses the optimal cloud entropy calculation method to more precisely determine the expected value and super entropy of the evaluation index, thereby more accurately calculating the cloud correlation of the evaluation index. It combines historical data to configure the grade standard for the evaluation index, fully utilizes historical experience, and improves the scientific nature of the evaluation. It determines the super entropy by statistically analyzing the discrete degree of the evaluation index data, which helps to identify the volatility and risk of the indicator. It determines the optimal weight by combining the maximum entropy criterion, which improves the objectivity and rationality of the weight determination. Through the accurate assessment of the economic operation status, it helps hospitals to rationally allocate resources and improve operational efficiency.

[0019] 3. The present invention processes historical time series data through a clustering algorithm, which can effectively extract meaningful benchmark indicators from a large amount of data. The time-difference correlation analysis method can identify the dynamic relationship between warning indicators and benchmark indicators, which helps to capture the lag effect in time series data. By quantifying the time-difference correlation coefficient, the correlation strength between warning indicators and benchmark indicators can be clearly understood. Through effective warning, hospitals can optimize resource allocation and reduce unnecessary cost expenditures. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 is a flow chart of a comprehensive evaluation method for hospital economic risks according to an embodiment of the present invention; Figure 2 The present invention is a block diagram of a hospital economic risk comprehensive evaluation system according to an embodiment of the present invention.

[0021] In the picture: 1. Indicator construction module; 2. Economic status assessment module; 3. Early warning model construction module; 4. Economic status monitoring module. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the technical solutions in this application, the following will clearly and completely describe the technical solutions in the embodiments of this application in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0023] According to an embodiment of the present invention, a method and system for comprehensively evaluating hospital economic risks are provided.

[0024] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 According to one embodiment of the present invention, a comprehensive evaluation method for hospital economic risks is provided, which includes the following steps: S1. Collect hospital economic operation data and establish economic operation status evaluation indicators and early warning indicators based on the hospital economic operation mechanism; Specifically, hospital economic operation data can be divided into multiple categories. Common data include but are not limited to the following: Financial data: such as revenue, expenditure, profit, cash flow, assets and liabilities, etc. These data directly reflect the economic health of the hospital.

[0025] Operational data: including the number of patients, length of hospital stay, outpatient visits, number of surgeries, etc. These indicators directly affect the hospital's income and expenditure.

[0026] Resource usage: such as medical equipment utilization rate, bed utilization rate, medicine and equipment inventory, etc., which affect operational efficiency.

[0027] Cost data: such as labor costs, equipment maintenance costs, drug procurement costs, etc. These costs will affect the profitability of the hospital.

[0028] According to the economic operation mechanism of the hospital, construct evaluation indicators that can reflect the overall economic status of the hospital. The selection of evaluation indicators mainly includes: Financial status evaluation indicators: such as revenue growth rate, cost control rate, profit margin, cash flow health index, etc.

[0029] Operational efficiency evaluation indicators: including resource utilization, equipment utilization efficiency, patient turnover rate, bed turnover rate, etc.

[0030] Cost control evaluation indicators: such as per capita cost, average cost per operation, cost per hospital stay, etc.

[0031] Quality and service level indicators: patient satisfaction, service quality scores, etc., can also serve as a supplement to economic indicators and indirectly reflect the operating status of the hospital.

[0032] Early warning indicators are used to identify potential risks in the hospital's economic situation in advance, and these indicators should have a certain degree of advance or synchronization. The selection of early warning indicators mainly includes: Leading indicators: include indicators in economic data that can reflect changing trends in advance, such as changes in the number of hospital admissions, fluctuations in the income of major departments, and the impact of external policies.

[0033] Synchronous indicators: indicators that change in sync with the hospital's economic performance, such as drug costs, ward utilization rates, etc.

[0034] Risk indicators: such as the rate of change of large expenses, failure rate of key equipment, debt ratio, capital liquidity, etc. These indicators can directly reflect potential economic risks.

[0035] S2. Based on the evaluation indicators of the hospital's economic operation status, the cloud correlation of the evaluation indicators is calculated using the optimal cloud entropy method; the comprehensive weight of each evaluation indicator is determined by improving the interval analytic hierarchy process using the maximum entropy criterion, and the cloud correlation and comprehensive weight of the evaluation indicators are used to evaluate the hospital's economic operation status and obtain the evaluation results; As a preferred implementation method, based on the hospital economic operation status evaluation indicators, the cloud correlation of the evaluation indicators is calculated by the optimal cloud entropy method; the comprehensive weight of each evaluation indicator is determined by improving the interval analytic hierarchy process by the maximum entropy criterion, and the cloud correlation and comprehensive weight of the evaluation indicators are used to evaluate the hospital economic operation status. Obtaining the evaluation results includes the following steps: S21. Determine the expected value and super entropy of each evaluation indicator, and use the optimal cloud entropy method to calculate the optimal cloud entropy of the evaluation indicator to obtain the optimal cloud entropy set; As a preferred embodiment, the expected value and super entropy of each evaluation indicator are determined, and the optimal cloud entropy of the evaluation indicator is calculated using the optimal cloud entropy method. Obtaining the optimal cloud entropy set includes the following steps: S211. Based on the evaluation indicators of the hospital's economic operation status and in combination with historical data, configure a grade standard for each evaluation indicator; Specifically, each indicator is divided into multiple grades according to the different states of hospital economic operation, such as "excellent", "good", "average" and "poor". Each grade should have a clear definition and a specific numerical range, reflecting the typical value of the indicator in that state. For example: Income: You can define "excellent" as being over a certain amount, "good" as being slightly below that amount, and so on.

[0036] Cost control rate: “Poor” can be defined as a control rate above a certain threshold, reflecting the risk of cost overruns.

[0037] Among them, the numerical range of each level is determined based on historical data analysis, and can be set by calculating the mean and standard deviation of each indicator under different economic operating conditions. For example: Excellent: Data that exceeds the mean by more than one standard deviation.

[0038] Good: Data within one standard deviation above and below the mean.

[0039] Fair: Data that is less than one standard deviation below the mean but still within a reasonable range.

[0040] Poor: Data that is significantly below one standard deviation from the mean.

[0041] S212, calculating the average value of each evaluation indicator within the range of each grade standard to obtain the expected value; It should be noted that the average value of the data of each evaluation indicator within each level range is calculated, that is, the expected value of the level is calculated.

[0042] S213, performing statistical analysis on the evaluation index data of each evaluation index within the range of each grade standard, and determining the super entropy by calculating the degree of dispersion of the evaluation index data; Specifically, based on the previously defined ranges, historical data for each evaluation indicator within each range is extracted. For example, for observations of the "income" indicator at the "good" level (30 million yuan < income ≤ 50 million yuan), this may include several years of historical annual income data.

[0043] The degree of dispersion of the evaluation index can be measured by variance or standard deviation. Variance and standard deviation can quantify the volatility and dispersion of data, representing the degree of dispersion of data within the level.

[0044] Hyperentropy can be expressed as the result of standard deviation or variance. Usually, hyperentropy is defined as the degree of dispersion of the indicator within different level standards, which is used to quantify the randomness and uncertainty of the data. Specifically: For levels with larger variance, it means that the observed data within this level fluctuates more, that is, the uncertainty is higher and the excess entropy value should also be higher.

[0045] For levels with smaller variance, it means that the data fluctuation is smaller, the uncertainty is lower, and the excess entropy value is relatively low.

[0046] S214. Calculate the primary cloud entropy value and the secondary cloud entropy value of each evaluation indicator respectively using the triple cloud entropy rule and the half-number correlation principle. Establish a nonlinear decision optimization model based on the primary cloud entropy value and the secondary cloud entropy value and solve it to obtain the optimal cloud entropy set.

[0047] As a preferred embodiment, the calculation formulas for calculating the first-level cloud entropy value and the second-level cloud entropy value of each evaluation index using the triple cloud entropy rule and the half-correlation principle are respectively: E′ n =( g max - g min ) / 6; E′′ n =( g max - g min ) / 2.3548; Where, It represents the first-level cloud entropy value calculated using the triple cloud entropy rule; represents the secondary cloud entropy value calculated using the half-correlation principle; g max Indicates the upper limit of the evaluation index in the grade standard; g min Indicates the lower limit of the evaluation index in the grade standard.

[0048] It should be noted that the significance of calculating cloud entropy using the triple cloud entropy rule lies in dividing the numerical range of the evaluation indicator into six equal parts, using triple cloud entropy to represent uncertainty. The denominator of 2.3548 in the half-correlation principle is derived from a statistical constant and can better represent the correlation of indicators. The secondary cloud entropy value reflects the concentration or degree of correlation of data within different ranges.

[0049] Specifically, the steps for establishing a nonlinear decision optimization model based on the first-level cloud entropy value and the second-level cloud entropy value to solve the optimal cloud entropy set are as follows: The first-level cloud entropy and secondary cloud entropy As two measures of uncertainty, a multi-objective optimization function is constructed. The general goal is to minimize the uncertainty of cloud entropy to obtain the best state of each evaluation index. By adjusting the combination of primary cloud entropy and secondary cloud entropy, an optimal cloud entropy set is obtained, and the objective function is set as: ; Where, Z represents the value of the objective function, n represents the number of evaluation indicators, λ Represents the weight coefficient for balancing the primary cloud entropy and the secondary cloud entropy, which is used to adjust the contribution of the primary and secondary cloud entropy in the objective function. i Indicates the number of the current evaluation indicator.

[0050] When building the optimization model, constraints are set to ensure the rationality of the solution. The constraints are: the cloud entropy value of each evaluation indicator should be within a reasonable range, for example: ; in, and They are the upper limits of the first-level cloud entropy and the second-level cloud entropy, respectively, and are determined by experience or actual conditions.

[0051] According to the above objective function and constraints, the gradient descent method is used to solve the problem. The optimal combination of the first-level cloud entropy and the second-level cloud entropy for each evaluation index is obtained, that is, the optimal cloud entropy set.

[0052] S22. Calculate the cloud correlation of each evaluation indicator based on the expected value and the optimal cloud entropy, and construct a cloud correlation matrix; As a preferred implementation, the expected value of each evaluation index and the optimal cloud entropy set are used to calculate the cloud correlation of each evaluation index. The calculation formula is: ; Where, R ij Indicates evaluation indicators i In grade standards j Cloud affinity on u i represents the actual observed value of the evaluation index; E x It indicates the expected value of the evaluation indicator under the corresponding grade standard; E ni It represents the optimal cloud entropy of the indicator under the corresponding grade standard.

[0053] Specifically, the cloud correlation of each evaluation index at different levels is R ij Combined into a matrix, called the cloud correlation matrix. The structure of the matrix is ​​as follows: ; In the formula, each element Indicates the i The evaluation index is j Cloud relevance on a tiered basis, p Indicates the number of grade standards.

[0054] S23. Determine the weight interval of each evaluation indicator using the interval analytic hierarchy process, and determine the comprehensive weight of the evaluation indicators using the maximum entropy criterion; As a preferred embodiment, the weight interval of each evaluation index is determined by using the interval analytic hierarchy process, and the comprehensive weight of the evaluation index is determined by using the maximum entropy criterion, including the following steps: S231. Compare the importance of each evaluation index pairwise and construct an interval number judgment matrix; Specifically, by collecting opinions from different experts or referring to historical data, each evaluation indicator is compared pairwise.

[0055] Based on the comparison results, construct an interval number judgment matrix. Each element in this matrix is ​​an interval number, representing the relative importance of one evaluation indicator relative to another. For example, if experts believe that indicator V is slightly more important than indicator B, you can fill in the corresponding position in the matrix with an interval number indicating slightly more important, such as [1.2, 1.5].

[0056] S232. Calculate the weight interval of each evaluation index using the interval number eigenvector method according to the interval number judgment matrix; Specifically, the interval number eigenvector method is a method for calculating the weight intervals of each evaluation indicator. It is based on the interval number judgment matrix. The interval number judgment matrix is ​​normalized so that each element becomes an interval number [a, b], where a and b represent the minimum and maximum values ​​of the element, respectively.

[0057] S233. The initial weight interval is used as a constraint condition of the maximum entropy criterion, and a nonlinear optimization model is constructed. By solving the nonlinear optimization model, the final weight value of each evaluation indicator is obtained.

[0058] It should be noted that the initial weight interval of each evaluation indicator is calculated from the interval number judgment matrix. This interval reflects the evaluation range of the relative importance of each indicator by experts or historical data.

[0059] The maximum entropy criterion aims to maximize the sum of the weighted entropy values ​​of all indicators, given partial known weight information. A nonlinear optimization model is constructed to maximize entropy while satisfying the aforementioned constraints. Finally, the model is solved using the Lagrange multiplier method or other nonlinear optimization algorithms (such as the Newton method or the trust region method) to obtain the final weight values.

[0060] S24. Using the cloud correlation and comprehensive weight of the evaluation indicators, the economic operation status of the hospital is evaluated according to the principle of maximum correlation to obtain the evaluation results.

[0061] It should be noted that according to the comprehensive weight Cloud Correlation R ij , calculate each grade standard j The total correlation : ; According to the principle of maximum correlation, the grade standard with the largest total correlation is selected as the evaluation result of the hospital's economic operation status. That is: ; in, Indicates the maximum total correlation Level j This level is the evaluation grade of the hospital’s economic operation status.

[0062] For example, suppose there are three grade standards ("excellent", "good", "average"), and three evaluation indicators A 1. A 2 and A 3, the weights are w 1=0.4, w 2=0.3, w 3=0.3. The cloud correlation matrix of each indicator is as follows: ; Calculate the total relevance of each level: Excellent level: S 优秀 =0.4×0.8+0.3×0.7+0.3×0.6=0.32+0.21+0.18=0.71; Good grade: S 良好 =0.4×0.6+0.3×0.5+0.3×0.4=0.24+0.15+0.12=0.51; General Grade: S 一般 =0.4×0.4+0.3×0.3+0.3×0.2=0.16+0.09+0.06=0.31; According to the maximum correlation principle, S 优秀 =0.71 is the maximum, so the evaluation result of the hospital's economic operation status is "excellent".

[0063] S3. Use the time difference correlation analysis method to analyze the time difference correlation of early warning indicators, and build a hospital economic operation early warning model based on the evaluation results; As a preferred implementation method, the time difference correlation analysis method is used to analyze the time difference correlation relationship of the early warning indicators, and the hospital economic operation early warning model is constructed based on the evaluation results; the following steps are included: S31. Collect historical time series indicator data of early warning indicators and calculate the benchmark indicators of early warning indicators through clustering algorithms; It should be noted that the historical time series data of early warning indicators related to hospital economic operations obtained through mobile phones usually include the values ​​of each early warning indicator at multiple time points, such as daily, weekly or monthly values.

[0064] As a preferred embodiment, collecting historical time series indicator data of early warning indicators and calculating the benchmark indicators of early warning indicators through clustering algorithms include the following steps: S311. Randomly select a number of data points from the historical indicator data of the early warning indicator as initial cluster centers; S312, respectively calculating the distance between each warning indicator data point and each initial cluster center, and assigning these data points to the cluster with the closest cluster center; It should be noted that for each data point of the early warning indicator, the Euclidean distance formula is used to calculate its distance to all initial cluster centers.

[0065] S313. For each cluster, calculate the average value of all data points within it, and use this average value as the new cluster center; S314: Repeat steps S312 and S313 until a preset number of iterations is reached. For each early warning indicator, select the cluster center in the corresponding cluster as a benchmark indicator.

[0066] For example, suppose there is a time series data of an early warning indicator: 2.5, 3.0, 3.5, 4.0, 5.0, 5.5, 6.0, 6.5, select k =2 cluster centers.

[0067] Randomly select two initial centers, say 3.0 and 5.5.

[0068] Calculate the distance of each data point to 3.0 and 5.5, and assign the data point to the nearest center: 2.5, 3.0, 3.5, 4.0 may be assigned to the 3.0 cluster, and 5.0, 5.5, 6.0, 6.5 may be assigned to the 5.5 cluster. Compute the new center of each cluster like: (2.5+3.0+3.5+4.0) / 4=3.25 and (5.0+5.5+6.0+6.5) / 4=5.75; Repeat the above steps and update the centers until convergence. The final cluster centers may be 3.25 and 5.75.

[0069] S32. Calculate the time difference correlation coefficient between the early warning indicator and the benchmark indicator using the time difference correlation analysis method based on the benchmark indicator of the early warning indicator; It's important to note that temporal correlation analysis is a method used to study the correlation between time series data, particularly when analyzing the degree of correlation between two time series at different delays. In early warning models for hospital economic operations, temporal correlation analysis can be used to analyze the correlation between early warning indicators and benchmark indicators, thereby helping to identify potential warning signals and trends.

[0070] As a preferred embodiment, based on the benchmark indicator of the early warning indicator, the calculation formula for the time difference correlation coefficient between the early warning indicator and the benchmark indicator is calculated using the time difference correlation analysis method: ; Where, It represents the time difference correlation coefficient between the early warning indicator and the benchmark indicator; N Indicates the time length of the data in the early warning indicator; represents the mean of the benchmark indicator; Indicates the b Benchmark indicators at time t The value of represents the mean value of the early warning indicator; Indicates the b Early warning indicators at the moment t The value of r Indicates the maximum delay number.

[0071] S33. Selecting a warning indicator that meets a preset screening rule based on the time difference correlation coefficient between the warning indicator and the benchmark indicator; S34. Based on the selected early warning indicators and combined with the evaluation results of the hospital's economic operation status, a hospital economic operation early warning model is constructed.

[0072] As a preferred implementation, based on the selected early warning indicators and combined with the evaluation results of the hospital's economic operation status, constructing a hospital economic operation early warning model includes the following steps: S341. Combining the selected early warning indicators and the evaluation results of the hospital's economic operation status into a data set, and dividing the data set into a training set and a test set; It should be noted that the selected early warning indicator data and the evaluation results of the hospital's economic operation status are combined into a complete data set; the selected early warning indicator data serve as the characteristics of the model, that is, the input variables of the model; and the evaluation results of the hospital's economic operation status serve as the target variables of the model, that is, the output of the model.

[0073] The dataset is divided into a training set and a test set. The common division ratio is 80% for the training set and 20% for the test set. The training set is used for model training, while the test set is used for model verification to evaluate the generalization ability of the model.

[0074] S342. Using the selected early warning indicators as input and the evaluation results of the hospital's economic operation status as output, a linear regression method is used to construct and train an early warning model based on the training set; It should be noted that the linear regression model assumes that the relationship between the hospital's economic operation status assessment results and the early warning indicators is linear, that is, it can be described by a linear equation.

[0075] The linear regression model is trained using the training set data. The training process finds the optimal linear relationship between the early warning indicators (input variables) and the economic performance (output variable). The core of the model is to find the regression coefficient for each early warning indicator so that the prediction error is minimized, meaning the model can predict the economic performance assessment results as accurately as possible.

[0076] S343. Use the test set data to verify the trained early warning model, verify the prediction performance of the early warning model, and adjust the early warning model parameters according to the verification results to obtain the final early warning model.

[0077] It is important to note that the trained early warning model is validated using the test set data, and predictive performance metrics such as mean squared error (MSE) and mean absolute error (MAE) are calculated to assess the model's accuracy and robustness. Based on the validation results, it is determined whether the model's predictive performance meets expectations. If the prediction results are unsatisfactory, adjustments to the model parameters or the introduction of additional features (such as adding polynomial terms to form a polynomial regression) can be attempted to optimize the model. After repeated model adjustments and validation, a final early warning model with optimal predictive performance is obtained.

[0078] S4. Conduct real-time monitoring based on the hospital economic operation early warning model, generate early warning signals, and analyze the risks of the hospital's economic operation based on the actual situation of the hospital's economic operation; Specifically, the hospital's information systems (such as financial systems, operation systems, resource management systems, etc.) are integrated with the early warning model to realize the automatic collection and transmission of real-time data.

[0079] Necessary preprocessing of real-time data, including data cleaning, format conversion, and missing value processing, is performed to ensure data quality and consistency. The preprocessed real-time data is then fed into a trained early warning model, which then predicts the input data and generates a forecast of the hospital's economic performance.

[0080] Based on the model's predictions, an early warning threshold is set. When the prediction exceeds the threshold, an early warning signal is generated. Different warning levels can be set based on the severity of the prediction, such as "minor risk," "medium risk," and "high risk."

[0081] Interpret the generated warning signals and analyze the specific reasons that may have led to the warning. For example, the decline in revenue is due to a decrease in the number of patients or policy adjustments.

[0082] The actual financial status of the hospital should be considered to assess potential risk points. For example, if the forecast results indicate tight cash flow, further analysis is needed to determine whether there are reasons for large amounts of uncollected accounts receivable or increased expenses.

[0083] Based on the early warning signals and risk assessment results, appropriate response measures can be formulated. For example, if cost control is found to be poor, resource allocation can be optimized or procurement strategies can be adjusted.

[0084] like Figure 2 As shown, according to another embodiment of the present invention, a hospital economic risk comprehensive assessment system is provided, the hospital economic risk comprehensive assessment system comprising: an indicator construction module 1, an economic status assessment module 2, an early warning model construction module 3 and an economic status monitoring module 4, wherein the indicator construction module 1, the economic status assessment module 2, the early warning model construction module 3 and the economic status monitoring module 4 are sequentially connected; Indicator construction module 1 is used to collect hospital economic operation data and construct economic operation status evaluation indicators and early warning indicators based on the hospital economic operation mechanism; The economic status evaluation module 2 is used to calculate the cloud correlation of the evaluation indicators based on the hospital economic operation status evaluation indicators through the optimal cloud entropy method; determine the comprehensive weight of each evaluation indicator through the maximum entropy criterion improved interval analytic hierarchy process, and use the cloud correlation and comprehensive weight of the evaluation indicators to evaluate the hospital's economic operation status and obtain the evaluation results; Early warning model construction module 3 is used to analyze the time difference correlation relationship of early warning indicators using the time difference correlation analysis method, and to construct a hospital economic operation early warning model based on the evaluation results; The economic status monitoring module 4 is used to perform real-time monitoring based on the hospital economic operation early warning model, generate early warning signals, and analyze the risks of the hospital economic operation in combination with the actual situation of the hospital economic operation.

[0085] In summary, with the help of the above technical solutions of the present invention, the present invention collects various types of data on the economic operation of the hospital and constructs comprehensive evaluation indicators and early warning indicators, which can more comprehensively reflect the economic health of the hospital, improve the economic management level of the hospital, and help the hospital better understand and control the economic operation status. Through data analysis, the hospital can more reasonably allocate resources and improve resource utilization efficiency. The early warning model can help the hospital to timely discover and respond to economic risks and enhance the hospital's risk control capabilities. It can provide a systematic and scientific economic operation data analysis framework for the hospital, which helps the hospital make more informed decisions in economic management. The present invention uses the optimal cloud entropy calculation method to more accurately determine the expected value and super entropy of the evaluation index, thereby more accurately calculating the cloud correlation of the evaluation index. It combines historical data as the evaluation index configuration grade standard, fully utilizes historical experience, and improves the scientific nature of the evaluation. It determines the super entropy by statistically analyzing the discrete degree of the evaluation index data, which helps to identify the volatility and risk of the indicator. It determines the optimal weight by combining the maximum entropy criterion, which improves the objectivity and rationality of the weight determination. Through the accurate evaluation of the economic operation status, it helps the hospital to reasonably allocate resources and improve operational efficiency. The present invention processes historical time series data through a clustering algorithm, and can effectively extract meaningful benchmark indicators from a large amount of data. The time-difference correlation analysis method can identify the dynamic relationship between early warning indicators and benchmark indicators, which helps to capture the lag effect in time series data. By quantifying the time-difference correlation coefficient, the strength of the correlation between early warning indicators and benchmark indicators can be clearly understood. Through effective early warning, hospitals can optimize resource allocation and reduce unnecessary cost expenditures.

[0086] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, optical storage, etc.) containing computer-usable program code.

[0087] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A comprehensive evaluation method for hospital economic risks, characterized by: include: S1. Construct economic operation status evaluation indicators and early warning indicators based on the hospital's economic operation mechanism; S2, calculate the cloud correlation of the evaluation index by the optimal cloud entropy method; The comprehensive weight of each evaluation index was determined by improving the interval analytic hierarchy process with the maximum entropy criterion, and the economic operation status of the hospital was evaluated using cloud correlation and comprehensive weight to obtain the evaluation results. S3. Use the time difference correlation analysis method to analyze the time difference correlation of early warning indicators, and build a hospital economic operation early warning model based on the evaluation results; S4. Conduct real-time monitoring based on the hospital economic operation early warning model, generate early warning signals, and analyze the risks of the hospital's economic operation based on the actual situation of the hospital's economic operation; The S3 includes: Collect historical time series indicator data of early warning indicators and calculate the benchmark indicators of early warning indicators through clustering algorithms; According to the benchmark indicators of the early warning indicators, the time difference correlation coefficient between the early warning indicators and the benchmark indicators is calculated using the time difference correlation analysis method; The calculation formula of the time difference correlation coefficient is: ； Where, It represents the time difference correlation coefficient between the early warning indicator and the benchmark indicator; N Indicates the time length of the data in the early warning indicator; and represent the mean values ​​of the benchmark indicator and the early warning indicator respectively; and Respectively represent b Benchmark indicators and b Early warning indicators at the moment t The numerical value of r Indicates the maximum delay number.

2. A comprehensive evaluation method for hospital economic risks according to claim 1, characterized in that: The cloud correlation of the evaluation indicators is calculated by the optimal cloud entropy method; the comprehensive weight of each evaluation indicator is determined by improving the interval analytic hierarchy process through the maximum entropy criterion, and the cloud correlation and comprehensive weight are used to evaluate the economic operation status of the hospital. The evaluation results include: Determine the expected value and super entropy of each evaluation index, and use the optimal cloud entropy method to calculate the optimal cloud entropy of the evaluation index; The cloud correlation of each evaluation index is calculated based on the expected value and the optimal cloud entropy, and a cloud correlation matrix is ​​constructed; The interval analytic hierarchy process is used to determine the weight interval of each evaluation index, and the maximum entropy criterion is used to determine the comprehensive weight of the evaluation index; By using the cloud correlation and comprehensive weight of the evaluation indicators, the economic operation status of the hospital is evaluated through the principle of maximum correlation to obtain the evaluation results.

3. A comprehensive evaluation method for hospital economic risks according to claim 2, characterized in that: The method of determining the expected value and super entropy of each evaluation index and calculating the optimal cloud entropy of the evaluation index using the optimal cloud entropy method includes: Based on the evaluation indicators of the hospital's economic operation status and combined with historical data, a grade standard is configured for each evaluation indicator; By calculating the average value of each evaluation indicator within the range of each grade standard, the expected value is obtained; Conduct statistical analysis on the evaluation index data of each evaluation index within the range of each grade standard, and determine the super entropy by calculating the degree of dispersion of the evaluation index data; The triple cloud entropy rule and half-correlation principle are used to calculate the first-level cloud entropy value and the second-level cloud entropy value of each evaluation index respectively. According to the first-level cloud entropy value and the second-level cloud entropy value, a nonlinear decision optimization model is established and solved to obtain the optimal cloud entropy set.

4. A comprehensive evaluation method for hospital economic risks according to claim 3, characterized in that: The calculation formulas for calculating the first-level cloud entropy value and the second-level cloud entropy value of each evaluation index using the triple cloud entropy rule and the half-correlation principle are respectively: E′ n =( g max - g min ) / 6； E′′ n =( g max - g min ) / 2.3548； Where, It represents the first-level cloud entropy value calculated using the triple cloud entropy rule; represents the secondary cloud entropy value calculated using the half-correlation principle; g max Indicates the upper limit of the evaluation index in the grade standard; g min Indicates the lower limit of the evaluation index in the grade standard.

5. A comprehensive evaluation method for hospital economic risks according to claim 4, characterized in that: The calculation formula for calculating the cloud correlation of each evaluation indicator by combining the expected value of each evaluation indicator and the optimal cloud entropy is: ； Where, R ij Indicates evaluation indicators i In grade standards j Cloud affinity on u i represents the actual observed value of the evaluation index; E x It indicates the expected value of the evaluation indicator under the corresponding grade standard; E ni It represents the optimal cloud entropy of the indicator under the corresponding grade standard.

6. A comprehensive evaluation method for hospital economic risks according to claim 4, characterized in that: The method of determining the weight interval of each evaluation index by using the interval analytic hierarchy process and determining the comprehensive weight of the evaluation index by using the maximum entropy criterion includes: Compare the importance of each evaluation index pairwise and construct the interval number judgment matrix; According to the interval number judgment matrix, the weight interval of each evaluation index is calculated using the interval number eigenvector method; The initial weight interval is used as the constraint condition of the maximum entropy criterion, and a nonlinear optimization model is constructed. By solving the nonlinear optimization model, the final weight value of each evaluation index is obtained.

7. A comprehensive evaluation method for hospital economic risks according to claim 1, characterized in that: The method of analyzing the time difference correlation relationship of the early warning indicators by using the time difference correlation analysis method and constructing the hospital economic operation early warning model in combination with the evaluation results also includes: According to the time difference correlation coefficient between the early warning indicator and the benchmark indicator, the early warning indicator that meets the preset screening rules is selected; Based on the selected early warning indicators and combined with the evaluation results of the hospital's economic operation status, a hospital economic operation early warning model is constructed.

8. A comprehensive evaluation method for hospital economic risks according to claim 7, characterized in that: The collection of historical time series indicator data of early warning indicators and the calculation of benchmark indicators of early warning indicators through clustering algorithms include: Randomly select several data points from the historical indicator data of the early warning indicator as the initial cluster centers; Calculate the distance between each warning indicator data point and each initial cluster center, and assign these data points to the cluster with the nearest cluster center; For each cluster, calculate the average value of all data points within it and use this average value as the new cluster center; The cluster center is updated repeatedly until the preset number of iterations is reached. For each early warning indicator, the cluster center in the corresponding cluster is selected as the benchmark indicator.

9. A comprehensive evaluation method for hospital economic risks according to claim 8, characterized in that: The hospital economic operation early warning model is constructed based on the selected early warning indicators and combined with the evaluation results of the hospital economic operation status, including: The selected early warning indicators and the evaluation results of the hospital's economic operation status are combined into a data set, and the data set is divided into a training set and a test set; The selected early warning indicators are used as input and the evaluation results of the hospital's economic operation status are used as output. Based on the training set, the early warning model is constructed and trained using the linear regression method. The trained early warning model is verified using the test set data to verify the prediction performance of the early warning model, and the early warning model parameters are adjusted according to the verification results to obtain the final early warning model.

10. A hospital economic risk comprehensive assessment system, used to implement the hospital economic risk comprehensive assessment method according to any one of claims 1 to 9, characterized in that: The hospital economic risk comprehensive assessment system includes: an indicator construction module, an economic status assessment module, an early warning model construction module and an economic status monitoring module, and the indicator construction module, the economic status assessment module, the early warning model construction module and the economic status monitoring module are connected in sequence; The indicator construction module is used to construct economic operation status evaluation indicators and early warning indicators based on the hospital's economic operation mechanism; The economic status evaluation module is used to calculate the cloud correlation of the evaluation indicators through the optimal cloud entropy method; determine the comprehensive weight of each evaluation indicator through the maximum entropy criterion improved interval analytic hierarchy process, and use the cloud correlation and comprehensive weight to evaluate the economic operation status of the hospital to obtain the evaluation results; The early warning model construction module is used to analyze the time difference correlation relationship of the early warning indicators using the time difference correlation analysis method, and to construct a hospital economic operation early warning model based on the evaluation results; The economic status monitoring module is used to perform real-time monitoring based on the hospital economic operation early warning model, generate early warning signals, and analyze the risks of the hospital's economic operation in combination with the actual situation of the hospital's economic operation.

Images