Special management medicine intelligent shift change management system and method for hospital pharmacy

Abstract

The invention relates to the technical field of medical informatization and intelligent management, and particularly discloses a special management medicine intelligent shift change management system and method for a hospital pharmacy, and the system obtains the medicine information of the shift change shift of the pharmacy through a data collection module, and carries out the structural arrangement and standardization processing of the data. The medicine inventory and operation records of the current shift are recorded, and the abnormal operation recognition module recognizes potential operation anomalies through a single-class support vector machine model and calculates abnormal operation feature values; the drug difference identification module adopts a principal component analysis method to analyze inventory difference changes, calculates drug difference feature values, constructs comprehensive feature vectors, uses a K-means clustering algorithm to assess the safety degree of shift change management and outputs risk coefficients, and the dynamic intelligent shift change management module formulates corresponding strategies according to risk levels. The system also introduces an adaptive threshold adjustment mechanism, calculates the time sequence similarity of historical and current shift change characteristics through a dynamic time warping algorithm, and dynamically adjusts a risk threshold.

Description

Technical Field

[0001] The present invention relates to the field of medical informationization and intelligent management technology, and in particular to an intelligent handover management system and method for special-management drugs in hospital pharmacies. Background Art

[0002] In modern hospital pharmacy management, the shift handover process for specially managed medications is crucial, as these drugs often carry strict regulatory requirements and high-risk characteristics. Traditional shift handover management relies primarily on manual recording and verification of information such as drug name, quantity, status, and storage location. However, with the increasing volume of medical services and the diversification of medication types, traditional methods are no longer able to meet increasingly complex management needs. To ensure the safety and compliance of medication management and improve the transparency and accuracy of the shift handover process, an intelligent, systematic solution is urgently needed to optimize the pharmacy shift handover process. Such a solution should be able to automatically identify abnormal operations, analyze inventory discrepancies, and dynamically adjust management strategies based on risk assessment results, thereby improving overall operational efficiency and management level.

[0003] The existing technology has the following deficiencies: Although the existing pharmacy management system already has certain data collection and recording functions, there are still many deficiencies in actual application. First, the traditional data processing method lacks an effective identification mechanism for abnormal operations, which makes it difficult to detect and correct potential operational errors and human errors in a timely manner. Secondly, when analyzing drug inventory differences, the existing system usually relies only on simple statistical methods and cannot deeply explore the complex patterns and potential risks in the data. In addition, the existing risk assessment models are mostly static settings, lack adaptive capabilities, and are difficult to flexibly adjust according to actual conditions, resulting in management strategies lagging behind actual needs. Finally, the existing system lacks a closed-loop feedback mechanism and cannot incorporate the evaluation results and threshold adjustments after each shift into the new round of data analysis, thereby limiting the system's self-optimization and continuous improvement capabilities. These problems jointly restrict the efficiency and safety of pharmacy shift management. A more intelligent and flexible solution is urgently needed to meet the above challenges. Therefore, this application provides an intelligent shift management system and method for special management drugs in hospital pharmacies to meet the needs. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide an intelligent handover management system and method for special management drugs in hospital pharmacies to solve the problems in the existing technology that it is difficult to accurately monitor drug changes, cannot effectively identify abnormal operations and evaluate the safety of handover.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: The intelligent handover management method for special-management drugs in hospital pharmacies includes the following steps: S1: Obtain the special management drug information of the pharmacy handover shift, including name, quantity, status, storage location, and structure and standardize the data, screen and record the drug inventory and operation records of the current shift; S2: Extract the operation information of the handover shift, compare it with the historical data, identify abnormal operations, and calculate the abnormal operation feature value; Abnormal operations include: changes in drug quantity, non-compliant drug use; S3: Analyze the changes in drug inventory differences and calculate the drug difference characteristic value based on the changes in drug inventory; S4: Combine abnormal operation characteristics and drug difference characteristics to construct a comprehensive feature vector as the input of the risk assessment model. Based on the model output, the safety level of shift management is evaluated; S5: Implement dynamic and intelligent shift management based on the evaluation results.

[0006] As a further solution of the present invention: the process of obtaining the abnormal operation characteristic value is: Extract the operation information of the handover shift, including drug name, drug quantity, drug status, storage location, operation time and operation type, and construct the feature vector of the handover shift operation , The index of each operation record in the handover data; The feature vector is standardized, and a single-class support vector machine is used to train the model on the handover data to obtain the weight vector w and bias ρ of the support vector machine; Standardized operating information for the new shift Input the trained single-class support vector machine model and calculate the decision value , and judge whether it is an abnormal operation by comparing it with the preset threshold γ: If , it is an abnormal operation. If , then it is a normal operation, calculate the abnormal operation characteristic value, the calculation expression is: ; in, represents the decision value of the support vector machine, is the normalized input vector The L2 norm of .

[0007] As a further solution of the present invention: the process of obtaining the drug difference characteristic value is: Obtain drug inventory change data, including: drug ID, drug name, storage location, shift time, inventory quantity, and inventory change; Construct drug inventory change data into feature vectors ,in Indicates the number of features; The constructed drug inventory change feature vector is standardized to obtain the standardized feature vector ; Construct the standardized feature vectors of all drugs into a data matrix ,in Number of drug samples; Calculate drug inventory data matrix The covariance matrix of is calculated as: ; represents the covariance matrix, express , Represents a transpose operation; Perform eigenvalue decomposition on the covariance matrix to obtain the eigenvector and eigenvalues ,in Represents the number of principal components, and the calculation expression of eigenvalue decomposition is: ; Before selection The principal component vector with the largest eigenvalue Constructing the principal component matrix , Indicates the eigenvalues; Standardized drug inventory data matrix Through the principal component matrix Projection to obtain the new drug inventory difference eigenvalue matrix , the calculation expression is: ; in, is the principal component eigenvalue matrix of drug inventory changes; The drug difference characteristic values ​​are calculated by the selected principal component characteristic values. The calculation expression is: ; in, Indicates the The drug in The projection value on the principal component, Indicates the The drug difference characteristic value of each drug.

[0008] As a further solution of the present invention: the construction of the comprehensive feature vector specifically includes: Obtain abnormal operation characteristics and drug difference characteristics of drug handover shifts for standardization; Get the standardized abnormal operation characteristics and drug inventory differences ; Construct a comprehensive feature vector, which includes the standardized abnormal operation features and drug inventory difference features, and is expressed as: ,in, Indicates the Shift handover and medication records.

[0009] As a further solution of the present invention: the process of constructing a risk assessment model based on the comprehensive feature vector is: Combine abnormal operation characteristics and drug difference characteristics to construct a comprehensive feature vector as input to the risk assessment model; K-means clustering is used to randomly initialize cluster centers, each of which represents a specific type of shift and drug inventory management mode; For each comprehensive feature vector Calculate the distance to each cluster center and assign it to the category corresponding to the nearest cluster center to form a cluster label; According to the samples in each cluster, the center point of each cluster is calculated and the position of the cluster center is updated until the change in the cluster center is less than the set threshold. The clustering is completed and the clustering result is output; The clustering result is: risk coefficient of shift management.

[0010] As a further solution of the present invention, the safety level of shift handover management is evaluated based on the risk factor of shift handover management, specifically including: Determine whether the risk coefficient of shift management is greater than or equal to a first threshold. If so, record it as a high-risk shift. If not, compare the risk coefficient of shift management with a second threshold. If less than or equal to the second threshold, record it as a low-risk shift. Otherwise, record it as a medium-risk shift. The first threshold is greater than the second threshold.

[0011] As a further solution of the present invention: dynamic intelligent shift management is implemented according to the evaluation results; Based on the risk level output results, formulate dynamic adjustment strategies and adjust shift management measures, including: For low-risk shifts, no intervention will be made and shift management will be carried out according to normal procedures; For medium-risk shifts, additional inspection and confirmation steps will be implemented, requiring handover personnel to double-check the quantity and status of drugs; For high-risk shifts, high-level risk intervention measures are implemented, and real-time monitoring and emergency response mechanisms are activated; By dynamically adjusting the risk assessment threshold and adopting an adaptive threshold adjustment mechanism, the risk assessment results are optimized, and dynamic intelligent shift management is implemented based on the assessment results.

[0012] As a further solution of the present invention: the adaptive threshold adjustment mechanism specifically includes: Using a dynamic time warping algorithm, calculate the time series similarity between abnormal operation characteristics and drug difference characteristics of historical shifts and abnormal operation characteristics and drug difference characteristics of the current shift. The similarity calculation process uses Euclidean distance for calculation; According to the calculation results of similarity, the risk threshold is dynamically adjusted. The calculation expression is: ; Where, represents the new risk threshold, represents the balance factor, Indicates the similarity calculation result between the historical handover and the current handover. Indicates the abnormal operation characteristics and drug difference characteristics of historical handovers, Indicates abnormal operation characteristics and drug difference characteristics of the current shift, Indicates the original risk threshold.

[0013] As a further solution of the present invention: based on the new risk threshold, after each shift handover is completed, the risk assessment results and threshold adjustment results are recorded through a feedback mechanism, and the risk assessment results and threshold adjustment results are used as reference inputs for a new round of shift handover data for cyclic optimization.

[0014] The intelligent shift management system for special drug management in hospital pharmacies includes: A data collection module is used to obtain information on special management drugs during the pharmacy shift, including name, quantity, status, and storage location, and to structure and standardize the data, screening and recording drug inventory and operation records for the current shift; An abnormal operation identification module is used to extract operation information of the handover shift, compare it with historical data, identify abnormal operations, and calculate abnormal operation feature values; A drug difference identification module is used to analyze changes in drug inventory differences and calculate drug difference feature values ​​based on drug inventory changes; A risk assessment model building module, which combines abnormal operation characteristics and drug difference characteristics to construct a comprehensive feature vector as the input of the risk assessment model and evaluates the safety level of shift management based on the model output; A dynamic intelligent shift handover management module implements dynamic intelligent shift handover management according to the evaluation results.

[0015] Compared with the prior art, the present invention has at least the following beneficial effects: (1) The present invention can accurately identify abnormal operations and changes in drug inventory differences by systematically collecting and analyzing special management drug information for pharmacy handover shifts. Specifically, a single-class support vector machine model is used to train and evaluate the operation information to identify potential abnormal behaviors, and the principal component analysis method is used to reduce the dimension of drug inventory differences and capture the main variation directions, thereby achieving a quantitative assessment of abnormal operations and drug differences. On this basis, a comprehensive feature vector is constructed as the input of the risk assessment model, and the K-means clustering algorithm is used to comprehensively evaluate the safety level of handover management. According to the risk level output results, the system can dynamically adjust the management strategy and take different intervention measures for low, medium and high risk shifts, such as increasing inspection links and enabling real-time monitoring and emergency response mechanisms. This method not only significantly improves the ability to identify potential risks and the speed of response, but also reduces human errors and potential operational errors through intelligent and adaptive management models, ensuring the high safety and accuracy of drug management, and further optimizing the overall operational efficiency and management level. In addition, the system also introduces an adaptive threshold adjustment mechanism, which calculates the time series similarity of historical and current handover characteristics through a dynamic time warping algorithm, dynamically adjusts the risk threshold, realizes self-optimization and continuous improvement of the system, and provides a scientific and efficient solution for pharmacy management.

[0016] (2) The present invention innovatively introduces an adaptive threshold adjustment mechanism, which calculates the time series similarity of historical and current handover characteristics through a dynamic time warping algorithm, accurately quantifies the difference between the two, and dynamically adjusts the risk assessment threshold based on this, ensuring that the system can flexibly respond to the ever-changing actual operating conditions. After each handover is completed, the system not only records the risk assessment results and the specific data of the threshold adjustment, but also uses it as an important reference input for the new round of handover data, forming a closed-loop feedback mechanism, thereby achieving self-optimization and continuous improvement of the system. This intelligent and adaptive management model significantly improves the flexibility and response speed of handover management, so that the management strategy is always in the optimal state. In addition, the system can make real-time adjustments based on new data and changing trends in actual operation, ensuring that it has a high degree of adaptability and foresight, greatly improving the overall operational efficiency and management level. In this way, the present invention not only enhances the ability to identify and control potential risks, but also provides a scientific, efficient and sustainable optimization solution for pharmacy management, ensuring that the safety and compliance of drug management meet the highest standards. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated herein and constitute a part of the specification, illustrate embodiments of the invention and, together with the description, further serve to explain the principles of the invention and to enable one skilled in the art to make and use the invention.

[0018] Figure 1 A flowchart of the specific steps of the intelligent handover management method for special management drugs in hospital pharmacies; Figure 2 This is a flowchart of the intelligent handover management system for special management drugs used in hospital pharmacies. DETAILED DESCRIPTION

[0019] The following describes in detail the intelligent handover management method for special-purpose medicines in hospital pharmacies provided by the present invention, in conjunction with the accompanying drawings and specific embodiments. It is also noted that, in order to make the embodiments more detailed, the following embodiments are best and preferred embodiments, and those skilled in the art may also adopt other alternatives for some known technologies. Furthermore, the accompanying drawings are only for the purpose of describing the embodiments in more detail, and are not intended to limit the present invention in any specific way.

[0020] It should be noted that references in the specification to "one embodiment," "an embodiment," "exemplary embodiments," "some embodiments," etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not necessarily every embodiment will include such specific features, structures, or characteristics. Furthermore, when specific features, structures, or characteristics are described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement such features, structures, or characteristics in conjunction with other embodiments (whether or not explicitly described).

[0021] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0022] like Figure 1 As shown, an embodiment of the present invention provides an intelligent shift management method for special management drugs in a hospital pharmacy, comprising the following steps: S1: Obtain the special management drug information of the pharmacy handover shift, including name, quantity, status, storage location, and structure and standardize the data, screen and record the drug inventory and operation records of the current shift; S2: Extract the operation information of the handover shift, compare it with the historical data, identify abnormal operations, and calculate the abnormal operation feature value; Abnormal operations include: changes in drug quantity, non-compliant drug use; S3: Analyze the changes in drug inventory differences and calculate the drug difference characteristic value based on the changes in drug inventory; S4: Combine abnormal operation characteristics and drug difference characteristics to construct a comprehensive feature vector as the input of the risk assessment model. Based on the model output, the safety level of shift management is evaluated; S5: Implement dynamic and intelligent shift management based on the evaluation results.

[0023] In S1, obtain the special management drug information of the pharmacy handover shift, including name, quantity, status, storage location, and structure and standardize the data. Filter and record the drug inventory and operation records of the current shift, including: First, log in to the system and navigate to the Shift Record section in the Medication Management module. Select the current shift, and the system will automatically list all special medications belonging to that shift, including details such as drug name, quantity, status, and storage location. Using the system's built-in filtering tools, you can further refine the data list based on specific criteria such as drug category and status to ensure accurate information. Next, use the system's export function to download this data in a structured format (such as Excel) for subsequent standardization and organization.

[0024] Next, the exported data undergoes detailed structured organization and standardization. This involves standardizing the presentation of drug names, verifying the accuracy of quantities, and confirming that each drug's status meets expectations (e.g., "Normal," "Needs Refill," or "Expired"). Furthermore, storage location information must be clear and easy to understand, facilitating quick location. After data standardization is complete, the current shift's drug inventory status and any operational records, such as drug additions, removals, or status changes, are further filtered. Finally, the organized data is compiled into a report that not only documents the current shift's drug inventory status but also details all operational records, ensuring transparency and traceability in drug management.

[0025] In S2, the operation information of the handover shift is extracted, compared with the historical data, abnormal operations are identified, and the characteristic values ​​of abnormal operations are calculated, including: The process of obtaining the abnormal operation characteristic value is as follows: Extract the operation information of the handover shift, including drug name, drug quantity, drug status, storage location, operation time and operation type, and construct the feature vector of the handover shift operation , The index of each operation record in the handover data; The feature vector is standardized, and a single-class support vector machine is used to train the model on the handover data to obtain the weight vector w and bias ρ of the support vector machine; Standardized operating information for the new shift Input the trained single-class support vector machine model and calculate the decision value , and judge whether it is an abnormal operation by comparing it with the preset threshold γ: If , it is an abnormal operation. If , then it is a normal operation, calculate the abnormal operation characteristic value, the calculation expression is: ; in, represents the decision value of the support vector machine, is the normalized input vector The L2 norm of .

[0026] It should be noted that: extracting operational information of shift handovers, constructing feature vectors and performing standardization, and using a single-class support vector machine to train a model to identify abnormal operations. This method can not only efficiently identify operational behaviors that do not conform to historical patterns, but also provides specific indicators of the degree of abnormality, thereby enhancing operational transparency and security in pharmacy management, effectively reducing human errors and improving overall operational efficiency.

[0027] In S3, the changes in drug inventory differences are analyzed, and the drug difference characteristic values ​​are calculated based on the changes in drug inventory, including: The process of obtaining the drug difference characteristic value is as follows: Obtain drug inventory change data, including: drug ID, drug name, storage location, shift time, inventory quantity, and inventory change; Construct drug inventory change data into feature vectors ,in Indicates the number of features; The constructed drug inventory change feature vector is standardized to obtain the standardized feature vector ; Construct the standardized feature vectors of all drugs into a data matrix ,in Number of drug samples; Calculate drug inventory data matrix The covariance matrix of is calculated as: ; represents the covariance matrix, express , Represents a transpose operation; Perform eigenvalue decomposition on the covariance matrix to obtain the eigenvector and eigenvalues ,in Represents the number of principal components, and the calculation expression of eigenvalue decomposition is: ; Before selection The principal component vector with the largest eigenvalue Constructing the principal component matrix , Indicates the eigenvalues; Standardized drug inventory data matrix Through the principal component matrix Projection to obtain the new drug inventory difference eigenvalue matrix , the calculation expression is: ; in, is the principal component eigenvalue matrix of drug inventory changes; The drug difference characteristic values ​​are calculated by the selected principal component characteristic values. The calculation expression is: ; in, Indicates the The drug in The projection value on the principal component, Indicates the The drug difference characteristic value of each drug.

[0028] It should be noted that by eigenvectorizing and standardizing drug inventory change data, constructing a data matrix, calculating its covariance matrix and eigenvalue decomposition, selecting the principal component vectors corresponding to the first k largest eigenvalues, and projecting the standardized data matrix onto these principal components, a new drug inventory variance eigenvalue matrix is ​​obtained. Finally, the drug variance eigenvalue is obtained by calculating the sum of the squares of each drug's projection values ​​onto the selected principal components. This method not only effectively identifies abnormal patterns in inventory changes but also quantifies the extent of these differences, helping managers more accurately monitor inventory dynamics, optimize inventory management strategies, reduce potential losses, and improve operational efficiency.

[0029] In S4, the abnormal operation characteristics and drug difference characteristics are combined to construct a comprehensive feature vector as the input of the risk assessment model. Based on the model output, the safety level of shift management is evaluated, including: The constructing of the comprehensive feature vector specifically includes: Obtain abnormal operation characteristics and drug difference characteristics of drug handover shifts for standardization; Get the standardized abnormal operation characteristics and drug inventory differences ; Construct a comprehensive feature vector, which includes the standardized abnormal operation features and drug inventory difference features, and is expressed as: ,in, Indicates the Shift handover and medication records; The process of constructing a risk assessment model based on the comprehensive feature vector is as follows: Combine abnormal operation characteristics and drug difference characteristics to construct a comprehensive feature vector as input to the risk assessment model; K-means clustering is used to randomly initialize cluster centers, each of which represents a specific type of shift and drug inventory management mode; For each comprehensive feature vector Calculate the distance to each cluster center and assign it to the category corresponding to the nearest cluster center to form a cluster label; According to the samples in each cluster, the center point of each cluster is calculated and the position of the cluster center is updated until the change in the cluster center is less than the set threshold. The clustering is completed and the clustering result is output; The clustering results are: risk coefficient of shift management; Evaluate the safety level of shift handover management based on its risk factor, specifically including: Determine whether the risk coefficient of shift management is greater than or equal to the first threshold. If so, record it as a high-risk shift. If not, compare the risk coefficient of shift management with the second threshold. If it is less than or equal to the second threshold, record it as a low-risk shift. Otherwise, record it as a medium-risk shift. The first threshold is greater than the second threshold.

[0030] It should be noted that the abnormal operation characteristics and drug difference characteristics of drug handover shifts are standardized and then merged to construct a comprehensive feature vector as the input of the risk assessment model. The K-means clustering algorithm randomly initializes the cluster centers and assigns categories based on the distance between each comprehensive feature vector and the cluster center. The cluster centers are iteratively updated until the change is less than the set threshold. The final clustering result represents the risk coefficient of shift handover management. This method not only accurately identifies potential high-risk shifts but also quantitatively assesses the safety level of shift handover management, thereby helping managers take timely measures to optimize management processes, reduce risks, and improve the safety and efficiency of overall operations.

[0031] In S5, dynamic and intelligent shift management is implemented based on the evaluation results, including: According to the evaluation results, dynamic and intelligent shift management is implemented; Based on the risk level output results, formulate dynamic adjustment strategies and adjust shift management measures, including: For low-risk shifts, no intervention will be made and shift management will be carried out according to normal procedures; For medium-risk shifts, additional inspection and confirmation steps will be implemented, requiring handover personnel to double-check the quantity and status of drugs; For high-risk shifts, high-level risk intervention measures are implemented, and real-time monitoring and emergency response mechanisms are activated; By dynamically adjusting risk assessment thresholds and adopting an adaptive threshold adjustment mechanism, risk assessment results can be optimized, and dynamic intelligent shift management can be implemented based on the assessment results; The adaptive threshold adjustment mechanism specifically includes: Using a dynamic time warping algorithm, calculate the time series similarity between abnormal operation characteristics and drug difference characteristics of historical shifts and abnormal operation characteristics and drug difference characteristics of the current shift. The similarity calculation process uses Euclidean distance for calculation; According to the calculation results of similarity, the risk threshold is dynamically adjusted. The calculation expression is: ; Where, represents the new risk threshold, represents the balance factor, Indicates the similarity calculation result between the historical handover and the current handover. Indicates the abnormal operation characteristics and drug difference characteristics of historical handovers, Indicates abnormal operation characteristics and drug difference characteristics of the current shift, represents the original risk threshold; Based on the new risk threshold, after each shift handover, the risk assessment results and threshold adjustment results are recorded through the feedback mechanism, and the risk assessment results and threshold adjustment results are used as reference inputs for the new round of shift handover data for cyclic optimization.

[0032] The working principle of this invention is to implement dynamic and intelligent shift management based on the evaluation results, and formulate corresponding dynamic adjustment strategies through the risk level output results. This method not only improves the flexibility and accuracy of shift management, but also enhances the ability to identify and control potential risks, ensures the safety and efficiency of drug management, and ultimately improves the safety level and management efficiency of the overall operation.

[0033] like Figure 2 As shown in the figure, the intelligent shift management system for special management drugs used in hospital pharmacies includes: A data collection module is used to obtain information on special management drugs during the pharmacy shift, including name, quantity, status, and storage location, and to structure and standardize the data, screening and recording drug inventory and operation records for the current shift; An abnormal operation identification module is used to extract operation information of the handover shift, compare it with historical data, identify abnormal operations, and calculate abnormal operation feature values; A drug difference identification module is used to analyze changes in drug inventory differences and calculate drug difference feature values ​​based on drug inventory changes; A risk assessment model building module, which combines abnormal operation characteristics and drug difference characteristics to construct a comprehensive feature vector as the input of the risk assessment model and evaluates the safety level of shift management based on the model output; A dynamic intelligent shift handover management module implements dynamic intelligent shift handover management according to the evaluation results.

[0034] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0035] The present invention provides a multi-step process for optimizing the safety and efficiency of pharmacy shift handover management. The system uses a data acquisition module to acquire information on special medications for pharmacy handovers, structure and standardize this data, and record the current shift's medication inventory and operation records. The abnormal operation identification module extracts and compares the handover operation information with historical data, identifies abnormal operations, and calculates their characteristic values. This information is then trained and evaluated using a single-class support vector machine model. The drug difference identification module analyzes changes in drug inventory, calculates drug difference characteristic values ​​based on the inventory changes, and uses principal component analysis to reduce dimensionality and capture the main directions of variation. Subsequently, the risk assessment model building module combines the abnormal operation characteristics and drug difference characteristics to construct a comprehensive characteristic vector as input to the risk assessment model. The K-means clustering algorithm is used to assess the safety of shift handover management and output a risk factor for the shift handover management. The dynamic intelligent shift handover management module develops dynamic adjustment strategies and implements appropriate shift handover management measures based on different risk levels: low-risk shifts follow the normal process, medium-risk shifts implement additional inspections, and high-risk shifts utilize real-time monitoring and emergency response mechanisms. In addition, the system also adopts an adaptive threshold adjustment mechanism, which calculates the time series similarity of historical and current handover characteristics through a dynamic time warping algorithm, and dynamically adjusts the risk threshold based on the similarity results to achieve cyclic optimization.

[0036] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0037] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. An intelligent shift management method for special-management drugs in hospital pharmacies, characterized in that: The following steps are involved: S1: Obtain the special management drug information of the pharmacy handover shift, including name, quantity, status, storage location, and structure and standardize the data, screen and record the drug inventory and operation records of the current shift; S2: Extract the operation information of the handover shift, compare it with the historical data, identify abnormal operations, and calculate the abnormal operation feature value; Abnormal operations include: changes in drug quantity, non-compliant drug use; S3: Analyze the changes in drug inventory differences and calculate the drug difference characteristic value based on the changes in drug inventory; S4: Combine abnormal operation characteristics and drug difference characteristics to construct a comprehensive feature vector as the input of the risk assessment model. Based on the model output, the safety level of shift management is evaluated; S5: Implement dynamic and intelligent shift management based on the evaluation results.

2. The intelligent shift management method for special-management drugs in hospital pharmacies according to claim 1 is characterized in that: The process of obtaining the abnormal operation characteristic value is as follows: Extract the operation information of the handover shift, including drug name, drug quantity, drug status, storage location, operation time and operation type, and construct the feature vector of the handover shift operation , The index of each operation record in the handover data; The feature vector is standardized, and a single-class support vector machine is used to train the model on the handover data to obtain the weight vector w and bias ρ of the support vector machine; Standardized operating information for the new shift Input the trained single-class support vector machine model and calculate the decision value , and judge whether it is an abnormal operation by comparing it with the preset threshold γ: If , it is an abnormal operation. If , then it is a normal operation, calculate the abnormal operation characteristic value, the calculation expression is: ; in, represents the decision value of the support vector machine, is the normalized input vector The L2 norm of .

3. The intelligent shift management method for special-management drugs in hospital pharmacies according to claim 1 is characterized in that: The process of obtaining the drug difference characteristic value is as follows: Obtain drug inventory change data, including: drug ID, drug name, storage location, shift time, inventory quantity, and inventory change; Construct drug inventory change data into feature vectors ,in Indicates the number of features; The constructed drug inventory change feature vector is standardized to obtain the standardized feature vector ; The standardized eigenvectors of all drugs are converted into data matrices ,in Number of drug samples; Calculate drug inventory data matrix The covariance matrix of is calculated as: ； represents the covariance matrix, express , Represents a transpose operation; Perform eigenvalue decomposition on the covariance matrix to obtain the eigenvector and eigenvalues ,in Represents the number of principal components, and the calculation expression of eigenvalue decomposition is: ; Before selection The principal component vector with the largest eigenvalue Constructing the principal component matrix , Indicates the eigenvalues; Standardized drug inventory data matrix Through the principal component matrix Projection to obtain the new drug inventory difference eigenvalue matrix , the calculation expression is: ； in, is the principal component eigenvalue matrix of drug inventory changes; The drug difference characteristic values ​​are calculated by the selected principal component characteristic values. The calculation expression is: ； in, Indicates the The drug in The projection value on the principal component, Indicates the The drug difference characteristic value of each drug.

4. The intelligent shift management method for special-management drugs in hospital pharmacies according to claim 1 is characterized in that: The constructing of the comprehensive feature vector specifically includes: Obtain abnormal operation characteristics and drug difference characteristics of drug handover shifts for standardization; Get the standardized abnormal operation characteristics and drug inventory differences ; Construct a comprehensive feature vector, which includes the standardized abnormal operation features and drug inventory difference features, and is expressed as: ,in, Indicates the Shift handover and medication records.

5. The intelligent shift management method for special-management drugs in hospital pharmacies according to claim 4 is characterized in that: The process of constructing a risk assessment model based on the comprehensive feature vector is as follows: Combine abnormal operation characteristics and drug difference characteristics to construct a comprehensive feature vector as input to the risk assessment model; K-means clustering is used to randomly initialize cluster centers, each of which represents a specific type of shift and drug inventory management mode; For each comprehensive feature vector Calculate the distance to each cluster center and assign it to the category corresponding to the nearest cluster center to form a cluster label; According to the samples in each cluster, the center point of each cluster is calculated and the position of the cluster center is updated until the change in the cluster center is less than the set threshold. The clustering is completed and the clustering result is output; The clustering result is: risk coefficient of shift management.

6. The intelligent shift management method for special-management drugs in hospital pharmacies according to claim 5 is characterized in that: Evaluate the safety level of shift handover management based on its risk factor, specifically including: Determine whether the risk coefficient of shift management is greater than or equal to a first threshold. If so, record it as a high-risk shift. If not, compare the risk coefficient of shift management with a second threshold. If less than or equal to the second threshold, record it as a low-risk shift. Otherwise, record it as a medium-risk shift. The first threshold is greater than the second threshold.

7. The intelligent shift management method for special-management drugs in hospital pharmacies according to claim 1 is characterized in that: According to the evaluation results, dynamic and intelligent shift management is implemented; Based on the risk level output results, formulate dynamic adjustment strategies and adjust shift management measures, including: For low-risk shifts, no intervention will be made and shift management will be carried out according to normal procedures; For medium-risk shifts, additional inspection and confirmation steps will be implemented, requiring handover personnel to double-check the quantity and status of drugs; For high-risk shifts, high-level risk intervention measures are implemented, and real-time monitoring and emergency response mechanisms are activated; By dynamically adjusting the risk assessment threshold and adopting an adaptive threshold adjustment mechanism, the risk assessment results are optimized, and dynamic intelligent shift management is implemented based on the assessment results.

8. The intelligent shift management method for special-management drugs in hospital pharmacies according to claim 7 is characterized in that: The adaptive threshold adjustment mechanism specifically includes: Using a dynamic time warping algorithm, calculate the time series similarity between abnormal operation characteristics and drug difference characteristics of historical shifts and abnormal operation characteristics and drug difference characteristics of the current shift. The similarity calculation process uses Euclidean distance for calculation; According to the calculation results of similarity, the risk threshold is dynamically adjusted. The calculation expression is: ； Where, represents the new risk threshold, represents the balance factor, Indicates the similarity calculation result between the historical handover and the current handover. Indicates the abnormal operation characteristics and drug difference characteristics of historical handovers, Indicates abnormal operation characteristics and drug difference characteristics of the current shift, Indicates the original risk threshold.

9. The intelligent shift management method for special-management drugs in hospital pharmacies according to claim 8 is characterized in that: Based on the new risk threshold, after each shift handover, the risk assessment results and threshold adjustment results are recorded through the feedback mechanism, and the risk assessment results and threshold adjustment results are used as reference inputs for the new round of shift handover data for cyclic optimization.

10. The intelligent shift management system for special drug management in hospital pharmacies is characterized by: The intelligent shift management method for special-management drugs in a hospital pharmacy according to any one of claims 1 to 9 comprises: A data collection module is used to obtain information on special management drugs during the pharmacy shift, including name, quantity, status, and storage location, and to structure and standardize the data, screening and recording drug inventory and operation records for the current shift; An abnormal operation identification module is used to extract operation information of the handover shift, compare it with historical data, identify abnormal operations, and calculate abnormal operation feature values; A drug difference identification module is used to analyze changes in drug inventory differences and calculate drug difference feature values ​​based on drug inventory changes; A risk assessment model building module, which combines abnormal operation characteristics and drug difference characteristics to construct a comprehensive feature vector as the input of the risk assessment model and evaluates the safety level of shift management based on the model output; A dynamic intelligent shift handover management module implements dynamic intelligent shift handover management according to the evaluation results.

Images