A method, system, and storage medium for medical institution performance management based on neural network models.

Abstract

This invention discloses a method, system, and storage medium for performance management of medical institutions based on a neural network model, relating to the field of data management. Relevant information of medical work projects is divided into individual evaluation information and group evaluation information; work performance indicators are divided into individual evaluation indicators and group evaluation indicators. A first mapping relationship set is generated from individual evaluation information and individual evaluation indicators, and a second mapping relationship set is generated from group evaluation information and group evaluation indicators. Semantic feature encoding is performed on each mapping relationship set. Based on the semantic feature encoding results, a first feature vector and a second feature vector are generated, and a neural network weight matrix is ​​constructed. The neural network is trained using the neural network weight matrix to obtain a neural network model. The various indicators of the individuals or groups to be evaluated are input into the neural network model to obtain a comprehensive performance index. This invention promotes the improvement of the management level of medical institutions and enhances their sustainable development capabilities.

Description

Technical Field

[0001] This invention relates to the field of data management, and more specifically to a method, system, and storage medium for performance management of medical institutions based on a neural network model. Background Technology

[0002] With the deepening of medical reforms, hospitals' internal management levels are gradually improving. Establishing a refined performance appraisal system can help hospitals implement refined management, objectively reflect their operational efficiency and development level, and facilitate adjustments to their operational philosophies and optimization of management measures, thereby improving operational efficiency. Simultaneously, improving performance appraisal indicators provides a reasonable basis for hospital evaluation. As a management tool for promoting strategic development, performance appraisal indicators, through refined management, enable hospitals to cultivate in-depth expertise in refined management, establish a scientific and reasonable appraisal system, improve incentive mechanisms, optimize distribution methods, establish multi-dimensional performance communication methods, improve performance management effectiveness, and ultimately help hospitals improve the quality of medical services and contribute to the healthcare industry.

[0003] Many medical institutions have incomplete quality management systems, lacking systematic quality management methods and tools. Although quality management systems have been established, difficulties arise in their actual implementation, leading to quality management becoming merely a formality. Currently, many medical institutions lack sound medical quality monitoring and feedback mechanisms, lacking effective monitoring methods and timely feedback channels. While some medical institutions have conducted quality monitoring, they lack in-depth analysis of the monitoring results and the formulation of corrective measures. Therefore, how to improve performance management levels and enhance the accuracy, objectivity, efficiency, and automation of performance evaluation is a crucial area that requires research in this field. Summary of the Invention

[0004] In view of this, the present invention provides a method, system and storage medium for medical institution performance management based on a neural network model, which can significantly improve the accuracy and objectivity of performance evaluation, enhance the efficiency and automation level of performance evaluation, promote the improvement of medical institution management level and enhance the sustainable development capability of medical institutions.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A performance management method for medical institutions based on a neural network model includes the following steps:

[0007] Step 1: Obtain relevant information on medical institution work projects and text data on work performance indicators. Divide the relevant information on medical work projects into individual evaluation information and group evaluation information, and divide the work performance indicators into individual evaluation indicators and group evaluation indicators.

[0008] Step 2: Generate a first mapping relationship set from individual evaluation information and individual evaluation indicators, and a second mapping relationship set from group evaluation information and group evaluation indicators. Semantic feature encoding is then performed on the first mapping relationship set and the second mapping relationship set respectively.

[0009] Step 3: Generate the first feature vector and the second feature vector based on the semantic feature encoding results; construct the neural network weight matrix based on the first feature vector and the second feature vector; train the neural network using the neural network weight matrix to obtain the neural network model.

[0010] Step 4: Input the various indicators of the individuals or groups to be evaluated into the neural network model to obtain the comprehensive performance indicators.

[0011] Optionally, in step one, predefine individual evaluation information, group evaluation information, individual evaluation indicators, and group evaluation indicators; annotate text data, and perform semantic correlation analysis on the annotated text data with the individual evaluation information, group evaluation information, individual evaluation indicators, and group evaluation indicators respectively; compare the correlation analysis results with preset thresholds to complete information classification.

[0012] Optionally, in step one, relevant information about the medical institution's work projects includes project progress reports and the percentage of each task completed.

[0013] Optionally, in step two, semantic feature encoding is performed on the first mapping relationship set and the second mapping relationship set respectively. Specifically, the mapping relationship set is input into the local embedding model, the mapping relationship is converted into a fixed-dimensional vector, and the vector is bidirectionally cyclically encoded to obtain the semantic feature encoding.

[0014] Optionally, in step three, the neural network model includes setting up an input layer: the number of neurons in the input layer should match the number of features selected for the evaluation metric, with each neuron corresponding to one feature; setting up hidden layers: multiple hidden layers are used to capture complex patterns in the data; the number of neurons and activation function in each hidden layer are set based on the characteristics of the data; setting up an output layer: the number of neurons in the output layer is multiple rows; selecting an activation function: the activation function is at least one of ReLU, Sigmoid, and Tanh; initializing weights and biases: setting a comprehensive prediction metric, randomly initializing the weights and biases of the neural network, and adjusting them later during training.

[0015] Optionally, it also includes using a loss function to iteratively optimize and control the neural network model to achieve the optimal overall performance index.

[0016] A performance management system for medical institutions based on a neural network model, comprising:

[0017] Text data acquisition and classification module: used to acquire text data related to work projects of medical institutions and work performance indicators. The relevant information of medical work projects is divided into individual evaluation information and group evaluation information, and the work performance indicators are divided into individual evaluation indicators and group evaluation indicators.

[0018] Semantic feature encoding module: used to generate a first mapping relationship set from individual evaluation information and individual evaluation indicators, and a second mapping relationship set from group evaluation information and group evaluation indicators, and to perform semantic feature encoding on the first mapping relationship set and the second mapping relationship set respectively;

[0019] The neural network model building module is used to generate a first feature vector and a second feature vector based on the semantic feature encoding results, construct a neural network weight matrix based on the first feature vector and the second feature vector, and train the neural network using the neural network weight matrix to obtain the neural network model.

[0020] Performance indicator output module: This module is used to input various indicators of the individuals or groups to be evaluated into the neural network model to obtain comprehensive performance indicators.

[0021] A computer storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of any one of the methods for medical institution performance management based on a neural network model.

[0022] As can be seen from the above technical solution, compared with the prior art, the present invention provides a medical institution performance management method, system, and storage medium based on a neural network model, which has the following beneficial effects:

[0023] 1. Improve the accuracy and objectivity of performance evaluation: This method subdivides relevant information and performance indicators of medical work projects into two dimensions: individual and group, which can more comprehensively reflect the work situation and performance within medical institutions. By semantically encoding the first and second mapping relationship sets, key information in the text data can be captured, providing an accurate data foundation for subsequent feature vector construction and neural network training. Utilizing a neural network model for comprehensive performance evaluation can simulate the learning and decision-making process of the human brain, thereby achieving an accurate understanding and assessment of complex performance relationships.

[0024] 2. Improve the efficiency and automation of performance evaluation: Neural network models can quickly process large amounts of performance data and output performance evaluation results in real time, greatly improving the efficiency of performance evaluation. This method automates the performance evaluation process, reduces the influence of human intervention and subjective judgment, and improves the objectivity and fairness of performance evaluation.

[0025] 3. Improve the management level of medical institutions: This method can provide managers of medical institutions with accurate performance evaluation results, helping them to better understand the work situation and performance of their employees, thereby making more scientific decisions. Through performance evaluation results, medical institutions can establish more reasonable incentive mechanisms, stimulate the work enthusiasm and creativity of employees, and improve the overall service and management level of medical institutions.

[0026] 4. Enhancing the Sustainable Development Capacity of Medical Institutions: This method can continuously monitor and evaluate the performance of medical institutions, helping them to identify and resolve problems in a timely manner, and achieve continuous improvement and optimization. By improving the management and service levels of medical institutions, it can enhance their competitiveness and sustainable development capabilities, laying a solid foundation for their long-term development. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0028] Figure 1 This is a schematic diagram of the process of the present invention;

[0029] Figure 2 This is a schematic diagram of the structure of the present invention. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Performance management refers to a continuous cycle in which managers and employees at all levels participate in performance planning, performance coaching and communication, performance appraisal, application of performance results, and improvement of performance goals in order to achieve organizational goals. The purpose of performance management is to continuously improve the performance of individuals, departments, and the organization.

[0032] A sound performance management system can organically integrate the strengths of all departments and personnel within an organization, and promote the most efficient and effective use of resources in line with the organization's overall goals. Performance management has become an important management technique or tool for improving an organization's operational and management capabilities.

[0033] This invention discloses a performance management method for medical institutions based on a neural network model, such as... Figure 1 As shown, it includes the following steps:

[0034] Step 1: Obtain relevant information on medical institution work projects and text data on work performance indicators. Divide the relevant information on medical work projects into individual evaluation information and group evaluation information, and divide the work performance indicators into individual evaluation indicators and group evaluation indicators.

[0035] Step 2: Generate a first mapping relationship set from individual evaluation information and individual evaluation indicators, and a second mapping relationship set from group evaluation information and group evaluation indicators. Semantic feature encoding is then performed on the first mapping relationship set and the second mapping relationship set respectively.

[0036] Step 3: Generate the first feature vector and the second feature vector based on the semantic feature encoding results; construct the neural network weight matrix based on the first feature vector and the second feature vector; train the neural network using the neural network weight matrix to obtain the neural network model.

[0037] Step 4: Input the various indicators of the individuals or groups to be evaluated into the neural network model to obtain the comprehensive performance indicators.

[0038] Furthermore, in step one, individual evaluation information, group evaluation information, individual evaluation indicators, and group evaluation indicators are predefined; text data is labeled, and semantic correlation analysis is performed between the labeled text data and the individual evaluation information, group evaluation information, individual evaluation indicators, and group evaluation indicators. The correlation analysis results are compared with preset thresholds to complete information classification.

[0039] The principles for information classification are as follows:

[0040] Clearly distinguish between individual and group evaluation information, and between individual and group evaluation indicators, to avoid confusion.

[0041] Relevance: The selected information or indicators should be directly related to the medical work project and be able to accurately reflect the work performance.

[0042] Objectivity: Use objective and quantitative standards for evaluation as much as possible to reduce the influence of subjective factors.

[0043] Operability: The selected information or indicators should be easy to obtain, calculate and understand, and easy to implement in practice.

[0044] Furthermore, in step one, the relevant information for the medical institution's work projects includes project progress reports and the percentage of each task completed.

[0045] Furthermore, in step two, semantic feature encoding is performed on the first mapping relationship set and the second mapping relationship set respectively. Specifically, the mapping relationship set is input into the local embedding model, the mapping relationship is converted into a fixed-dimensional vector, and the vector is bidirectionally cyclically encoded to obtain the semantic feature encoding.

[0046] The bidirectional cyclic encoding calculation process is as follows:

[0047]

[0048] Among them, T a Let W represent the first encoded vector, f represent the probability coefficients, and W represent the first encoded vector. a Let T represent the forward transition matrix. se(a) W|gd| represents the features corresponding to the vector text during forward encoding, W|gd| represents the transition matrix, t(j-1) represents the text information during forward looping, and T represents the features corresponding to the vector text during forward encoding. b W represents the second encoded vector. b Let T represent the subsequent transition matrix. se(b) The first encoding vector and the second encoding vector are obtained, and the correlation between the first encoding vector and the second encoding vector is calculated. Based on the correlation, the concatenated encoding vectors corresponding to the first encoding vector and the second encoding vector are calculated using the following formula.

[0049] T(a, b) = relu[T a T b ];

[0050] Where T(a, b) represents the concatenated coding vector corresponding to the first and second coding vectors, and ReLU is the activation function. a T represents the first encoded vector. b This represents the second encoded vector.

[0051] Furthermore, in step three, the neural network model includes setting up an input layer: the number of neurons in the input layer should match the number of features selected for the evaluation metric, with each neuron corresponding to one feature; setting up hidden layers: multiple hidden layers are used to capture complex patterns in the data; the number of neurons and activation function in each hidden layer are set based on the characteristics of the data; setting up an output layer: the number of neurons in the output layer is multiple rows; selecting an activation function: the activation function is at least one of ReLU, Sigmoid, and Tanh; initializing weights and biases: setting a comprehensive prediction metric, randomly initializing the weights and biases of the neural network, and adjusting them later during training.

[0052] Furthermore, it also includes using a loss function to iteratively optimize and control the neural network model to achieve the optimal comprehensive performance index, specifically including the following steps:

[0053] Define the loss function: A loss function measures the difference between the predicted values ​​and the actual values ​​of a neural network model. In the context of performance management in medical institutions, commonly used loss functions such as mean squared error (MSE) and cross-entropy loss can be selected. These functions can reflect the degree of deviation between the model's predicted values ​​and the actual performance indicators, thus providing a clear direction for model optimization.

[0054] Iterative optimization: Iterative optimization methods such as gradient descent, momentum method, and adaptive learning rate are used to continuously update the parameters of the neural network model based on the gradient information of the loss function. During this process, appropriate hyperparameters such as the number of iterations and the learning rate need to be set to ensure that the model converges to or near the optimal solution.

[0055] Monitoring the loss function value: During the iteration process, it is necessary to continuously monitor the changes in the loss function value. If the loss function value continues to decrease, it indicates that the model is optimizing in the right direction; if the loss function value fluctuates or increases, it may be necessary to adjust the hyperparameters or redesign the model structure.

[0056] Achieving the optimal comprehensive performance index: Through continuous iterative optimization, when the loss function value reaches a preset threshold or converges to a stable state, the neural network model can be considered to have achieved the optimal comprehensive performance index. At this point, the model can be applied to actual medical institution performance management to accurately evaluate individual work performance and overall group performance.

[0057] and Figure 1 Corresponding to the method shown, this invention also discloses a medical institution performance management system based on a neural network model for... Figure 1 The implementation of the method, specifically the structure is as follows: Figure 2 As shown, it includes:

[0058] Text data acquisition and classification module: used to acquire text data related to work projects of medical institutions and work performance indicators. The relevant information of medical work projects is divided into individual evaluation information and group evaluation information, and the work performance indicators are divided into individual evaluation indicators and group evaluation indicators.

[0059] Semantic feature encoding module: used to generate a first mapping relationship set from individual evaluation information and individual evaluation indicators, and a second mapping relationship set from group evaluation information and group evaluation indicators, and to perform semantic feature encoding on the first mapping relationship set and the second mapping relationship set respectively;

[0060] The neural network model building module is used to generate a first feature vector and a second feature vector based on the semantic feature encoding results, construct a neural network weight matrix based on the first feature vector and the second feature vector, and train the neural network using the neural network weight matrix to obtain the neural network model.

[0061] Performance indicator output module: This module is used to input various indicators of the individuals or groups to be evaluated into the neural network model to obtain comprehensive performance indicators.

[0062] This embodiment discloses a computer storage medium storing a computer program, which, when executed by a processor, implements the steps of any one of the methods for medical institution performance management based on a neural network model.

[0063] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0064] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A performance management method for medical institutions based on a neural network model, characterized in that, Includes the following steps: Step 1: Obtain relevant information on medical institution work projects and text data on work performance indicators. Divide the relevant information on medical work projects into individual evaluation information and group evaluation information, and divide the work performance indicators into individual evaluation indicators and group evaluation indicators. Step 2: Generate a first mapping relationship set from individual evaluation information and individual evaluation indicators, and a second mapping relationship set from group evaluation information and group evaluation indicators. Semantic feature encoding is then performed on the first mapping relationship set and the second mapping relationship set respectively. Step 3: Generate the first feature vector and the second feature vector based on the semantic feature encoding results; construct the neural network weight matrix based on the first feature vector and the second feature vector; train the neural network using the neural network weight matrix to obtain the neural network model. Step 4: Input the various indicators of the individuals or groups to be evaluated into the neural network model to obtain the comprehensive performance indicators.

2. The medical institution performance management method based on a neural network model according to claim 1, characterized in that, In step one, individual evaluation information, group evaluation information, individual evaluation indicators, and group evaluation indicators are predefined; text data is labeled, and semantic correlation analysis is performed on the labeled text data with the individual evaluation information, group evaluation information, individual evaluation indicators, and group evaluation indicators respectively. The correlation analysis results are compared with preset thresholds to complete information classification.

3. The medical institution performance management method based on a neural network model according to claim 1, characterized in that, In step one, relevant information about the medical institution's work projects includes project progress reports and the percentage of each task completed.

4. The medical institution performance management method based on a neural network model according to claim 1, characterized in that, In step two, semantic feature encoding is performed on the first mapping relationship set and the second mapping relationship set respectively. Specifically, the mapping relationship set is input into the local embedding model, the mapping relationship is converted into a fixed-dimensional vector, and the vector is bidirectionally cyclically encoded to obtain the semantic feature encoding.

5. The medical institution performance management method based on a neural network model according to claim 1, characterized in that, In step three, the neural network model includes setting up the input layer: the number of neurons in the input layer should match the number of features selected for the evaluation metric, with each neuron corresponding to one feature; Setting up hidden layers: Multiple hidden layers are used to capture complex patterns in the data; the number of neurons and activation function in each hidden layer are set based on the characteristics of the data; Setting up output layers: The number of neurons in the output layer is multiple rows; Selecting activation functions: The activation function is at least one of ReLU, Sigmoid, and Tanh; Initializing weights and biases: Setting up a comprehensive prediction index, the weights and biases of the neural network are randomly initialized and adjusted later during training.

6. The medical institution performance management method based on a neural network model according to claim 1, characterized in that, It also includes using loss functions to iteratively optimize neural network models in order to achieve the optimal comprehensive performance index.

7. A medical institution performance management system based on a neural network model, characterized in that, include: Text data acquisition and classification module: used to acquire text data related to work projects of medical institutions and work performance indicators. The relevant information of medical work projects is divided into individual evaluation information and group evaluation information, and the work performance indicators are divided into individual evaluation indicators and group evaluation indicators. Semantic feature encoding module: used to generate a first mapping relationship set from individual evaluation information and individual evaluation indicators, and a second mapping relationship set from group evaluation information and group evaluation indicators, and to perform semantic feature encoding on the first mapping relationship set and the second mapping relationship set respectively; The neural network model building module is used to generate a first feature vector and a second feature vector based on the semantic feature encoding results, construct a neural network weight matrix based on the first feature vector and the second feature vector, and train the neural network using the neural network weight matrix to obtain the neural network model. Performance indicator output module: This module is used to input various indicators of the individuals or groups to be evaluated into the neural network model to obtain comprehensive performance indicators.

8. A computer storage medium, characterized in that, The computer storage medium stores a computer program, which, when executed by a processor, implements the steps of a medical institution performance management method based on a neural network model as described in any one of claims 1-6.

Images