Hospital nursing quality evaluation system based on hospital information terminal data

Abstract

The invention discloses a hospital nursing quality evaluation system based on hospital information terminal data, and relates to the technical field of medical information, and the system comprises a data acquisition module which extracts index data based on a hospital terminal, and the index data comprises patient satisfaction subjective index data, nursing personnel working life data, performance data and adverse event occurrence probability; after preprocessing the data, constructing a data set; and the analysis module performs correlation analysis on the data of the data set based on the Pearson's correlation coefficient, constructs a nursing quality evaluation model through a neural network, and performs training optimization on the quality evaluation model. The comprehensive evaluation level is efficiently and objectively output through the evaluation module, the result is presented through a visual chart, managers can conveniently master the comprehensive evaluation level, hospitals accurately position nursing service short boards according to the evaluation result, targeted optimization is implemented, the nursing system is continuously improved, the requirements of patients are met, and it is ensured that the nursing quality is kept at a high level; and data in a hospital can be fully utilized.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and specifically to a hospital nursing quality assessment system based on hospital information terminal data. Background Art

[0002] In today's medical system, the assessment of hospital nursing quality is crucial for ensuring patient safety and improving the level of medical services. Traditional hospital nursing quality assessments mainly rely on manual inspections and subjective evaluations, which have many drawbacks. During the manual inspection process, due to differences in the professional levels and personal experiences of assessors, the assessment results may lack objectivity and accuracy. Subjective evaluations are often affected by the personal emotions and biases of evaluators and are difficult to comprehensively and truly reflect the actual quality of nursing work; the assessment process is cumbersome, consuming a large amount of manpower and time; it is difficult to conduct real-time and dynamic monitoring and assessment of nursing work. In addition, a large amount of terminal data has accumulated in the hospital information system, but these data have not been fully and effectively utilized for nursing quality assessment, resulting in a waste of resources. For this reason, we propose a hospital nursing quality assessment system based on hospital information terminal data. Summary of the Invention

[0003] To solve the above technical problems and provide a hospital nursing quality assessment system based on hospital information terminal data, this technical solution solves the problems of cumbersome assessment, low efficiency, and lack of data support.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is: a hospital nursing quality assessment system based on hospital information terminal data, including: A data acquisition module that extracts index data based on hospital terminals, including subjective index data of patient satisfaction, nursing staff working years data, performance data, and the probability of adverse events. After preprocessing the data, a data set is constructed; An analysis module that performs correlation analysis on the data in the data set based on the Pearson correlation coefficient, constructs a nursing quality assessment model through a neural network, trains and optimizes the quality assessment model, and determines the weight values of each data index; An evaluation module that inputs the acquired data into the model for quality assessment, calculates the comprehensive evaluation level, and presents and analyzes the evaluation results in the form of a chart; the hospital optimizes and improves the nursing service system based on the evaluation results.

[0005] Preferably, the data acquisition module extracts the index data from the hospital's terminal system, where the patient satisfaction is collected based on the patient feedback survey system; the working years data of the nursing staff is extracted based on the hospital's employment management system, and the performance data is obtained based on the hospital's performance management system; among them, the quantitative index of patient satisfaction is obtained through the evaluation system. In the hospital's feedback system, patients score based on the quality evaluation form to obtain the patient satisfaction value; the working years data of the nursing staff is the working years data of the current nursing staff in the hospital; the quantitative processing of the performance data is carried out by the nursing level evaluation form set by the hospital to score the nursing levels of different nursing staff to obtain the quantified performance data; the incidence of adverse events data is extracted from the hospital's adverse events reporting system, and the number of adverse events occurring in the department within a unit time and the corresponding total number of patients are counted to calculate the incidence of adverse events.

[0006] Preferably, the data preprocessing includes data cleaning, data transformation, and data integration; data cleaning includes missing value processing, outlier processing, and duplicate value processing, and the data transformation is based on Z-score standardization processing; the preprocessed data is organized into a data set.

[0007] Preferably, the steps of correlation analysis are as follows: data preparation, retrieving the organized data in the data set, where the subjective index data of patient satisfaction is the variable , including observation values ; the working years of the nursing staff is the variable , with observation values ; the performance data is the variable , and there are also observation values ; The correlation coefficient between the subjective index data of patient satisfaction and the working years of the nursing staff is calculated as follows: where is the mean value of the variable , is the mean value of the variable ; The correlation coefficient between the subjective index data of patient satisfaction and the performance data is calculated as follows: where is the mean value of the variable ; Working years of nursing staff and performance data correlation coefficient ; Based on the formula, calculate the correlation of each data index and conduct an analysis.

[0008] Preferably, the analysis and judgment steps are as follows: The correlation coefficient has a value range of , and the specific judgment rule is: Positive correlation. When , the two variables show a positive correlation. The closer it is to 1, the stronger the positive correlation. That is, the subjective index data of patient satisfaction and the working years of nursing staff have a strong positive correlation. When the working years of nursing staff increase, the patient satisfaction also tends to increase; The subjective index data of patient satisfaction and the performance data are also positively correlated. That is, the better the performance data of nursing staff, the higher the subjective index data of patient satisfaction.

[0009] Preferably, the method for constructing the evaluation model is: Obtain the historical data set as the model construction sample set, select the recurrent neural network model, build the network structure, set the activation function, train the built model, divide the data set, set the learning rate and optimizer parameters, and perform forward and backward propagation iterative training, and evaluate whether the model meets the standard using the mean square error index.

[0010] Preferably, the method for determining the weight values of each data index is: Divide the problem of weight determination into the target layer, the criterion layer, and the scheme layer. The target layer is used to determine the weight values of the data indexes. The criterion layer includes the subjective index data of patient satisfaction, the working years of nursing staff, and the performance data. The scheme layer is used to determine the weight allocation scheme; Through the expert scoring method, compare the importance of each factor in the criterion layer relative to the target layer pairwise, construct the judgment matrix, calculate the judgment matrix, obtain the weight vector of each factor, and conduct the consistency test. After the test passes, the weight vector is reasonable.

[0011] Preferably, the steps for constructing the judgment matrix are as follows: There are m factors in the criterion layer, namely the subjective index data of patient satisfaction, the working years of nursing staff, and the performance data , and the judgment matrix is a square matrix, and its element represents the importance degree of the th factor relative to the th factor, and the 1-9 scale method is used to determine The value, specifically expressed as: , indicating factor is equally important as factor; , indicating factor is slightly more important than factor; , indicating factor is significantly more important than factor; , indicating factor is strongly more important than factor; , indicating factor is extremely more important than factor; is the intermediate value of the above adjacent judgments; If factor is not as important as factor, then ; For the subjective index data of patient satisfaction, the working years and performance data of nursing staff, the judgment matrix is: Next, calculate the maximum eigenvalue of the judgment matrix , solve the corresponding eigenvector, and normalize the eigenvector to obtain the weight vector w.

[0012] Preferably, the comprehensive evaluation level calculation formula is calculated by the weighted average method, and the expression is: Where X, Y, and Z are the quantitative variables of patient satisfaction, the variable of the working years of nursing staff, and the performance data variable respectively; w1, w2, and w3 are the corresponding weight vector values obtained by solving, and S is the comprehensive score value; Based on the calculated comprehensive score value, different degrees are divided, and the evaluated nursing quality is classified into the corresponding grade list for intuitive feedback.

[0013] Preferably, based on the historical data accumulated in the hospital, the incidence rate of adverse events in different time periods is statistically analyzed, and the standard deviation is used to analyze the distribution characteristics of the data, ensuring that the incidence rate is within the acceptable range. At the same time, the range of abnormal fluctuations can be detected in a timely manner, and values are taken within the range. The current value is the threshold of the incidence rate of adverse events. The probability value of adverse events occurring in the current hospital is compared with the threshold. When it is greater than the threshold, the feedback mechanism is triggered, and the hospital conducts an in-depth investigation of the nursing process, analyzes the causes of adverse events, and formulates improvement measures. Based on the evaluation results, the hospital improves the nursing service system; in terms of personnel management, training programs are launched, and special skill improvement courses are arranged; in view of the nursing workload and personnel load in different departments, nursing human and material resources are allocated, the optimization effect is continuously tracked, and the strategy is dynamically adjusted.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: Through the data acquisition module, the present invention comprehensively collects multi-source data, discovers data associations based on the Pearson correlation coefficient, constructs a precise evaluation model with a neural network and determines the index weights. The evaluation module efficiently and objectively outputs the comprehensive evaluation level, and presents the results in an intuitive chart for the convenience of managers to master. Based on the evaluation results, the hospital accurately locates the short board of nursing services, implements targeted optimization, continuously improves the nursing system, meets the needs of patients, ensures that the nursing quality maintains a high level, and fully utilizes and efficiently analyzes the data within the hospital. Description of the Drawings

[0015] Figure 1 It is the framework diagram of the quality evaluation system of the present invention; Figure 2 It is the flowchart of the steps for constructing the evaluation model of the present invention; Figure 3 It is the flowchart of the steps for determining the weight values of each data index of the present invention. Detailed Embodiments

[0016] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0017] Refer to Figure 1 As shown, the hospital nursing quality evaluation system based on hospital information terminal data includes: A data acquisition module extracts index data based on the hospital terminal, including subjective index data of patient satisfaction, nursing staff working years data, performance data, and the probability of adverse events occurring. After preprocessing the data, a data set is constructed. An analysis module conducts correlation analysis on the data in the dataset based on the Pearson correlation coefficient, constructs a nursing quality assessment model through a neural network, trains and optimizes the quality assessment model, and determines the weight values of each data indicator; An evaluation module inputs the acquired data into the model for quality assessment, calculates the comprehensive evaluation level, and presents and analyzes the evaluation results in the form of charts; The hospital optimizes and improves the nursing service system based on the evaluation results.

[0018] This application widely collects subjective index data of patient satisfaction, working years data of nursing staff, and performance data from the hospital terminal, covering all key links of nursing services. The subjective index data of patient satisfaction reflects the intuitive feelings of patients towards nursing services, the nursing staff data includes personnel qualifications and work experience information, and the performance data quantifies the results of nursing work. The integration of such multi-source data avoids the one-sidedness of a single data source and ensures the comprehensiveness and richness of the evaluation basis; Preprocessing the extracted data can effectively remove missing values, outliers, and duplicate values in the data and unify the data format. This process greatly improves the accuracy and usability of the data, provides a reliable data foundation for subsequent analysis and evaluation, and avoids evaluation biases caused by data quality problems; Conducting correlation analysis on in-depth data based on the Pearson correlation coefficient can uncover potential relationships between different data indicators. It can be found that there is a positive correlation between the training duration of nursing staff and patient satisfaction, or a negative correlation between the nursing operation error rate and performance data. These insights provide clear directions for optimizing nursing services, helping hospital managers understand which factors have the most significant impact on nursing quality. Constructing a nursing quality assessment model using a neural network gives full play to the powerful non-linear processing ability of the neural network. Through learning and training on a large amount of data, the model can automatically capture complex patterns and rules in the data. During the training process, continuously optimize the model parameters and determine the weight values of each data indicator to make the evaluation model more in line with the actual situation and improve the accuracy and reliability of the evaluation; Based on the evaluation results, the hospital can accurately locate the weak links in nursing services and formulate targeted improvement measures. If the evaluation results show that the patient satisfaction in a certain department is relatively low, the hospital can start from aspects such as the service attitude, communication skills, and nursing processes of nursing staff for improvement; if the performance data reflects that there are many nursing operation mistakes, the hospital can strengthen the skills training and operation norms of nursing staff. This targeted improvement can quickly improve the nursing quality and meet the needs of patients. By continuously tracking the evaluation results, the hospital can continuously adjust and optimize the nursing service system. With the development of medical technology, the changes in patient needs, and the continuous improvement of hospital management requirements, the nursing service system needs to continuously adapt to these changes. The continuous optimization based on the evaluation results can ensure that the hospital's nursing services always maintain a high level and provide more high-quality and safe nursing services for patients.

[0019] The data acquisition module extracts index data from the hospital's terminal system. Among them, patient satisfaction is collected based on the patient feedback survey system; the data of the working years of nursing staff is extracted based on the hospital's employment management system, and the performance data is obtained based on the hospital's performance management system; among them, the quantification index of patient satisfaction is obtained through the evaluation system. Patients score based on the quality evaluation form in the hospital's feedback system to obtain the patient satisfaction value; the data of the working years of nursing staff is the data of the current working years of the hospital's nursing staff; the quantitative processing of performance data is carried out by the nursing level evaluation form set by the hospital to score the nursing levels of different nursing staff to obtain the quantified performance data; the data of the incidence of adverse events is extracted from the hospital's adverse event reporting system, and the number of adverse events occurring in the department within a unit time and the corresponding total number of patients are counted to calculate the incidence of adverse events.

[0020] The data acquisition module of this application accurately collects various types of index data through the hospital terminal system. Patient satisfaction comes from the patient feedback survey system, and patients score according to the quality evaluation form, truly reflecting the experience and ensuring the accuracy of the data. The data of the working years of nursing staff is extracted from the employment management system, providing a reliable basis for measuring experience. Performance data borrows from the performance management system and is quantified according to the nursing level evaluation form, with professionalism and authority. The multi-source integrated data covers dimensions such as patient feelings, personnel information, and work performance, enabling a comprehensive evaluation of nursing quality, facilitating a comprehensive analysis of the relationships between various data, mining potential problems and improvement directions. Using a standardized process, both patient satisfaction and performance data are scored according to the evaluation form, improving the consistency and comparability of the data, and relying on dedicated systems for acquisition, reducing manual errors, ensuring the accuracy and completeness of the data, laying a solid foundation for subsequent data analysis and model construction, and helping the hospital improve the nursing management level.

[0021] Data preprocessing includes data cleaning, data transformation, and data integration; data cleaning includes handling missing values, outliers, and duplicate values, and data transformation is based on Z-score standardization; the preprocessed data is organized into a dataset.

[0022] The data preprocessing of this application mainly covers three core parts: data cleaning, data transformation, and data integration. Data cleaning aims to improve the quality and usability of data. Handling missing values is one of the important tasks. When there are missing values in the dataset, it will cause bias in the analysis results. For a small number of missing values, if they are numerical data, the mean filling method can be used, that is, calculate the average value of the data in this column to fill in the missing place; if they are categorical data, the mode can be used for filling. If there are many missing values and they are concentrated in some records, and without affecting the overall data structure and analysis objectives, these records can be considered for deletion. Outlier handling cannot be ignored either. Outliers are identified through methods such as box plots and standard deviation methods. In the data of nursing staff working hours, if there are records of working hours far beyond the normal range, after being judged as outliers, boundary value replacement can be used, that is, replace the outliers with boundary values within a reasonable range to avoid their interference with the overall data trend. Duplicate value handling is relatively straightforward. By comparing all fields of data records, completely duplicate records are identified and deleted to eliminate data redundancy. Data transformation is based on Z - score standardization to unify the scales of different data metrics; data integration is to integrate the subjective index data of patient satisfaction, the working years data of nursing staff, and the performance data obtained from different information terminals in the hospital. These data may come from multiple data sources such as patient feedback survey systems, recruitment management systems, and performance management systems, and there are differences in data formats and encoding methods. During the integration process, it is necessary to unify the data format and solve data semantic conflicts. For example, the expressions of nursing staff titles in different systems are different and need to be normalized. By establishing effective data association rules, the scattered data is integrated into a complete and ordered dataset.

[0023] The steps of correlation analysis are as follows: data preparation, retrieve the organized data in the dataset, where the subjective index data of patient satisfaction is the variable , including observations ; the working years of nursing staff is the variable , with observations ; the performance data is the variable , also with observations ; The subjective index data of patient satisfaction and the working years of nursing staff correlation coefficient The calculation formula is: where is the mean value of variable , is the mean value of variable ; The subjective index data of patient satisfaction and the performance data The correlation coefficient The calculation formula is: where is the mean value of variable ; The working years of nursing staff and the performance data The correlation coefficient ; Based on the formula, calculate the correlation of each data index and conduct an analysis.

[0024] In this application, the correlation analysis is of great significance. It can reveal the internal relationship of data, assist in accurate evaluation, clarify the impact of each factor on nursing quality, optimize the evaluation model accordingly, reasonably allocate index weights. Through analysis, potential problems can be found. If the correlation between the working years of nursing staff and performance is weak, it may imply an unreasonable career development plan, and then provide a direction for improvement and formulate targeted strategies. For example, in view of the negative correlation between patient satisfaction and the nursing operation error rate, strengthen skills training. In addition, the correlation analysis also supports decision-making. Allocate resources reasonably according to the results. If the training investment is positively correlated with performance and patient satisfaction, increase the training investment; at the same time, tracking the change of correlation can predict the trend of nursing quality, prevent problems in advance, and improve the hospital management level.

[0025] Among them, the analysis and judgment steps are: The correlation coefficient The value range of is , and the specific judgment rule is: Positive correlation. When , the two variables show a positive correlation. The closer it is to 1, the stronger the positive correlation degree, that is, there is a strong positive correlation between the subjective index data of patient satisfaction and the working years of nursing staff . When the working years of nursing staff increase, the patient satisfaction also tends to increase; The subjective index data of patient satisfaction and the performance data are also positively correlated, that is, the better the performance data of nursing staff, the higher the subjective index data of patient satisfaction.

[0026] The analysis and judgment steps based on the range of correlation coefficient values in this application have many benefits in hospital nursing quality assessment. In positive correlation analysis, it can clarify the development direction of nursing staff. For example, based on the positive correlation between working years and patient satisfaction, it can strengthen the inheritance of experience, optimize the promotion and deployment of personnel, and can also optimize the allocation of training resources according to the key factors determined by positive correlation. Negative correlation analysis can help the hospital locate the key performance indicators affecting patient satisfaction, such as the number of nursing operation errors, and then improve the nursing process targeted and establish a continuous quality monitoring mechanism. At the same time, the results of positive and negative correlation analysis can provide a scientific basis for hospital management decisions, making resource allocation more reasonable, including reasonably planning resource investment and formulating strategic development plans. For example, it can layout the future development direction according to the relationship between remote nursing services and patient satisfaction, and improve the competitiveness and nursing level of the hospital.

[0027] The method for constructing the evaluation model is as follows: Obtain the historical data set as the model construction sample set, select the recurrent neural network model, build the network structure, set the activation function, train the built model, divide the data set, set the learning rate and optimizer parameters, and perform iterative training through forward and backward propagation. Use the mean square error index to evaluate whether the model meets the standard.

[0028] This application uses the historical data set as the sample set, enables the model to be trained based on real nursing scenario data, and continuously optimizes with the accumulation of data, greatly improving the evaluation accuracy. Selecting the recurrent neural network model, which fits the time series characteristics of nursing data, can effectively mine complex non-linear relationships, comprehensively and accurately evaluate nursing quality, scientifically divide the data set, reasonably set the learning rate and optimizer parameters, optimize the training process, accelerate the model convergence, and at the same time strictly measure the model with the mean square error as the index to ensure that the model performance meets the standard, providing a reliable and stable evaluation tool for the hospital to optimize nursing services.

[0029] The method for determining the weight values of each data index is as follows: Divide the problem of weight determination into the target layer, criterion layer, and scheme layer. The target layer is used to determine the weight values of data indexes. The criterion layer includes subjective index data of patient satisfaction, working years and performance data of nursing staff. The scheme layer is used to determine the weight allocation scheme; through the method of expert scoring, compare the importance of each factor in the criterion layer relative to the target layer pairwise, construct the judgment matrix, calculate the judgment matrix, obtain the weight vector of each factor, and perform consistency test. After the test passes, the weight vector is reasonable.

[0030] The steps for constructing the judgment matrix are as follows: There are m factors in the criterion layer, namely subjective index data of patient satisfaction, working years and performance data of nursing staff , the judgment matrix is a square matrix, and its element represents the The importance degree of a factor relative to the th factor is determined using the 1-9 scale method for the value of , indicating that the factor is equally important as the factor; , indicating that the factor is slightly more important than the factor; , indicating that the factor is significantly more important than the factor; , indicating that the factor is strongly more important than the factor; , indicating that the factor is extremely more important than the factor; is the intermediate value of the adjacent judgments above; If the factor is not as important as the factor, then ; For the subjective index data of patient satisfaction, the working years and performance data of nursing staff, the judgment matrix is: Next, calculate the maximum eigenvalue of the judgment matrix , solve the corresponding eigenvector, and normalize the eigenvector to obtain the weight vector w.

[0031] The comprehensive evaluation level calculation formula is calculated using the weighted average method, and the expression is: where X, Y, and Z are the quantitative variables of patient satisfaction, the variable of the working years of nursing staff, and the performance data variable respectively; w1, w2, and w3 are the corresponding weight vector values obtained by solving, and S is the comprehensive score value; Based on the calculated comprehensive score value, different degrees are divided, and the evaluated nursing quality is classified into the corresponding grade list for intuitive feedback.

[0032] Based on the historical data accumulated by the hospital, the incidence rate of adverse events in different time periods is statistically analyzed. The standard deviation is used to analyze the distribution characteristics of the data, ensuring that the incidence rate is within the acceptable range. At the same time, the range of abnormal fluctuations can be detected in a timely manner, and values are taken within the range. The current value is the threshold of the incidence rate of adverse events. The probability value of adverse events occurring in the current hospital is compared with the threshold. When it is greater than the threshold, the feedback mechanism is triggered, and the hospital conducts an in-depth investigation of the nursing process, analyzes the causes of adverse events, and formulates improvement measures. The hospital improves the nursing service system based on the evaluation results; in terms of personnel management, training programs are launched, and special skill improvement courses are arranged; according to the nursing task volume and personnel load in different departments, nursing human and material resources are allocated, and the optimization effect is continuously tracked, and the strategy is dynamically adjusted.

[0033] The hospital determines the threshold of the incidence rate of adverse events through in-depth mining and analysis of historical data, and constructs a complete medical quality control system. First, data collection and collation are carried out. Adverse event records in the past five years are collected from each department and nursing unit of the hospital, covering various types such as falls and infections. The time, location, and patient information content of the events are detailedly recorded. The number of adverse events occurring is statistically counted according to different time dimensions, and the incidence rate is calculated in combination with the total number of patients, forming multiple groups of data. These incidence rate data are analyzed using the standard deviation. The average value of the data is calculated to understand the overall level, and then the standard deviation is calculated to reflect the degree of data dispersion. Based on the analysis results, a reasonable range is determined by floating a certain multiple of the standard deviation above and below the average value. This range represents the normal fluctuation interval. Taking into account the actual situation of the hospital and the risk degree factors of each department, a suitable value is selected within the reasonable range as the threshold of the incidence rate of adverse events. The threshold for high-risk departments is set low, and the threshold for low-risk departments is appropriately increased. A monitoring and feedback mechanism is established. The hospital information system calculates the probability of adverse events occurring in real time and compares it with the threshold. If the probability is greater than the threshold, the feedback mechanism is triggered, and relevant personnel such as nursing experts and medical quality management personnel are quickly organized to conduct an in-depth investigation of the nursing process. Reasons are found from aspects such as operation procedures, personnel training, and resource allocation, and targeted improvement measures are formulated, such as optimizing processes, organizing training, and adjusting resource allocation. At the same time, a tracking and evaluation mechanism is established to ensure the effective implementation of improvement measures, continuously reduce the incidence rate of adverse events, and improve the quality of medical services. The comprehensive evaluation grade of this application is calculated using a weighted average formula, fully incorporating three core variables: patient satisfaction, nursing staff working years, and performance data, and performing operations according to the corresponding weight vector values. This ensures that the evaluation process comprehensively covers the key factors affecting nursing quality and avoids a single factor dominating the evaluation results; based on the evaluation results, training programs are carried out in personnel management. If the evaluation shows that the nursing staff in a certain department is weak in specific skills, special skill improvement courses are arranged. Continuously track the evaluation results to promptly grasp the effects after the implementation of the optimization strategy. After implementing a new personnel deployment plan or training program, re-evaluate the nursing quality, observe the changes in the comprehensive scores and grades, and find out whether the improvement measures are effective.

[0034] The above has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and all these changes and improvements fall within the scope of the present invention claimed.

Claims

1. A hospital nursing quality assessment system based on hospital information terminal data, characterized in that, Including: A data acquisition module extracts index data based on hospital terminals, including subjective index data of patient satisfaction, working years data of nursing staff, performance data, and the probability of adverse events. After preprocessing the data, a data set is constructed. An analysis module conducts correlation analysis on the data in the data set based on the Pearson correlation coefficient, constructs a nursing quality assessment model through a neural network, trains and optimizes the quality assessment model, and determines the weight values of each data index. An evaluation module inputs the acquired data into the model for quality assessment, calculates the comprehensive evaluation level, and presents and analyzes the evaluation results in the form of a chart. The hospital optimizes and improves the nursing service system based on the evaluation results.

2. The hospital nursing quality assessment system based on hospital information terminal data according to claim 1, wherein The data acquisition module extracts index data from the terminal system of the hospital. Among them, patient satisfaction is collected based on the patient feedback survey system; the working years data of nursing staff is obtained based on the hospital's employment management system, and the performance data is obtained based on the hospital's performance management system; the quantitative index of patient satisfaction is obtained through the evaluation system. Patients score based on the quality evaluation form in the hospital's feedback system to obtain the patient satisfaction value; the working years data of nursing staff is the working years data of the current nursing staff in the hospital; the quantitative processing of performance data is carried out by scoring the nursing levels of different nursing staff through the nursing level evaluation form set by the hospital to obtain the quantified performance data; the adverse event incidence data is extracted from the hospital's adverse event reporting system, and the number of adverse events occurring in the department and the corresponding total number of patients within a unit time are statistically counted to calculate the adverse event incidence.

3. The hospital nursing quality assessment system based on hospital information terminal data according to claim 1, wherein, Data preprocessing includes data cleaning, data transformation, and data integration; data cleaning includes missing value processing, outlier processing, and duplicate value processing, and data transformation is based on Z-score standardization processing. The preprocessed data is organized into a data set.

4. The hospital nursing quality assessment system based on hospital information terminal data according to claim 1, wherein, The steps of correlation analysis are as follows: data preparation, retrieving the organized data in the dataset, where the subjective index data of patient satisfaction is the variable , including observations ; The working years of nursing staff are variables , with observations ; The performance data are variables , also with observations ; Subjective index data of patient satisfaction With the working years of nursing staff Correlation coefficient The calculation formula is: wherein is the mean value of the variable , and is the mean value of the variable . Subjective index data of patient satisfaction and performance data The correlation coefficient The calculation formula is as follows: wherein is a variable is the mean value Working years of nursing staff and performance data correlation coefficient ; Calculate the correlation of each data index based on the formula and conduct analysis.

5. The hospital nursing quality assessment system based on hospital information terminal data according to claim 4, wherein Among them, the analysis and judgment steps are as follows: the correlation coefficient The value range of is as follows, and the specific judgment rule is: Positively correlated. When occurs, the two variables show a positive correlation. The closer it is to 1, the stronger the positive correlation. That is, the subjective index data of patient satisfaction and the working years of nursing staff have a strong positive correlation. When the working years of nursing staff increase, the patient satisfaction also tends to increase. The subjective index data of patient satisfaction and the performance data are also positively correlated, that is, the better the performance data of the nursing staff, the higher the subjective index data of patient satisfaction.

6. The hospital nursing quality assessment system based on hospital information terminal data according to claim 1, characterized in that The method for constructing the evaluation model is as follows: Obtain the historical data set as the model construction sample set, select the recurrent neural network model, build the network structure, set the activation function, train the built model, divide the data set, set the learning rate and optimizer parameters, and perform forward and backward propagation iterative training, and evaluate whether the model meets the standard using the mean square error index.

7. The hospital nursing quality assessment system based on hospital information terminal data according to claim 1, characterized in that, The method for determining the weight values of each data index is as follows: Divide the problem of weight determination into the target layer, criterion layer, and scheme layer. The target layer is used to determine the weight values of the data index, the criterion layer includes the subjective index data of patient satisfaction, the working years of nursing staff, and performance data, and the scheme layer is used to determine the weight allocation scheme. Through the expert scoring method, compare the importance of each factor in the criterion layer relative to the target layer pairwise, construct a judgment matrix, calculate the judgment matrix, obtain the weight vector of each factor, and conduct a consistency test. After passing the test, the weight vector is reasonable.

8. The hospital nursing quality assessment system based on hospital information terminal data according to claim 7, characterized in that, The steps for constructing the judgment matrix are as follows: There are m factors in the criterion layer, namely, the subjective index data of patient satisfaction, the working years and performance data of nursing staff , the judgment matrix is a square matrix, and its element represents the importance degree of the th factor relative to the th factor. The 1-9 scale method is used to determine the value, which is specifically expressed as: , indicates that factor and factor are equally important; , indicating Factor ratio The factor is slightly more important; , indicating Factor ratio Factor is significantly more important; , indicates Factor ratio Factor is strongly important; indicates factor ratio The factor is extremely important; is the intermediate value of the above adjacent judgment; If the factor is less than the factor is important, then ; For the subjective index data of patient satisfaction, the working years and performance data of nursing staff, the judgment matrix is as follows: Next, calculate the judgment matrix for its maximum eigenvalue , solve the corresponding eigenvector, and obtain the weight vector w after normalizing the eigenvector.

9. The hospital nursing quality assessment system based on hospital information terminal data according to claim 1, characterized in that, The comprehensive evaluation level calculation formula is calculated by the weighted average method, and the expression is: Where X, Y, and Z are the quantitative variables of patient satisfaction, the variable of the working years of nursing staff, and the performance data variable respectively; w1, w2, and w3 are the weight vector values obtained by corresponding solutions, and S is the comprehensive score value; Based on the calculated comprehensive score value, different degrees are divided, and the evaluated nursing quality is classified into the corresponding grade list for intuitive feedback.

10. The hospital nursing quality assessment system based on hospital information terminal data according to claim 1, wherein Based on the historical data accumulated in the hospital, the incidence of adverse events in different time periods is statistically analyzed, and the standard deviation is used to analyze the distribution characteristics of the data, ensuring that the incidence is within the acceptable range, and at the same time, the range of abnormal fluctuations can be detected in time. The current value within the range is the threshold of the incidence of adverse events. The current probability value of adverse events in the hospital is compared with the threshold. When it is greater than the threshold, the feedback mechanism is triggered, and the hospital conducts an in-depth investigation of the nursing link, analyzes the causes of adverse events, and formulates improvement measures; The hospital improves the nursing service system based on the evaluation results; In terms of personnel management, training programs are launched, and special skill improvement courses are arranged; according to the nursing workload and personnel load in different departments, nursing human and material resources are allocated, the optimization effect is continuously tracked, and the strategy is dynamically adjusted.

Images