A method and system for dynamic management of nuclear medicine subpackaging activity

Abstract

This invention proposes a method and system for dynamic management of radiopharmaceutical repackaging activity. The method includes: constructing a multi-dimensional safety dynamic graph; performing time-series aggregation analysis based on the multi-dimensional safety dynamic graph to predict the activity changes at each repackaging node and generate node activity prediction values; performing multi-objective analysis based on the node activity prediction values ​​and the multi-dimensional safety dynamic graph to generate a scheduling strategy for repackaging tasks, including the repackaging sequence of each task node, the transportation path of each repackaging task, and the repackaging time window; performing simulation verification on the generated scheduling strategy, correcting anomalies or potential risks, and generating a corrected scheduling strategy; executing the repackaging task using the corrected scheduling strategy as a control signal, collecting activity, transportation path risk, and time information in real time to form a feedback dataset, and performing intelligent optimization based on the feedback dataset.

Description

Technical Field

[0001] This invention belongs to the field of dynamic activity management, and particularly relates to a method and system for dynamic activity management of radiopharmaceutical packaging. Background Technology

[0002] Radiopharmaceutical dispensing systems are widely used in the production, distribution, and clinical use of radiopharmaceuticals. Their core task is to rationally, efficiently, and safely dispense high-activity radiopharmaceutical doses into various containers or usage units according to the needs of patients or departments. Radiopharmaceuticals exhibit strong time dependence, with their activity rapidly decaying over time. Therefore, dispensing management not only requires precise control of the dispensing dose but also must consider time window management during the dispensing process to maximize activity utilization and reduce resource waste. Simultaneously, radiopharmaceutical dispensing must strictly control radiation exposure for personnel, comply with radiation safety regulations, and ensure personnel safety during the dispensing process. Current radiopharmaceutical dispensing activity management generally relies on manual experience and rule bases for dispensing task scheduling. It typically employs a method of single-point activity measurement and static formula estimation, that is, using activity monitoring data from a small number of dispensing nodes combined with an exponential decay model to estimate the activity changes at each dispensing stage, and then formulating corresponding dispensing strategies.

[0003] This method has many shortcomings.

[0004] First, the complexity of the dispensing process makes it difficult to accurately reconstruct the actual distribution of activity in the dispensing chain through simple single-point measurements and static models. This is especially true when there are multiple dispensing nodes, multiple operators, dynamic path selection, and uncertainties (such as sudden operational delays or equipment failures). The activity distribution exhibits a non-linear and dynamic trend, which existing methods cannot accurately reflect in real time. Second, dispensing scheduling under existing methods usually relies on manual or pre-set heuristic rules, lacking a dynamic balancing mechanism for multiple objective factors such as activity, time, and personnel exposure. This can easily lead to low activity dispensing efficiency or increased safety risks.

[0005] Furthermore, due to the lack of dynamic global perception and intelligent decision-making capabilities, current dispensing systems struggle to effectively handle unexpected situations during the dispensing process, such as abnormal and rapid activity decay, conflicting personnel operation paths, and bottleneck node failures in scheduling. As the scale of radiopharmaceutical use in medical institutions continues to expand, the complexity of the dispensing process is constantly increasing, and the traditional "static activity prediction + manual scheduling" model is increasingly unable to meet the needs for efficient, safe, and dynamic dispensing activity management.

[0006] There is an urgent need for a new dynamic management system that can realize real-time dynamic perception of activity throughout the entire process and intelligently optimize multiple factors such as activity utilization efficiency, safety, and dispensing timeliness, so as to improve the overall efficiency and safety level of radiopharmaceutical dispensing. Summary of the Invention

[0007] The purpose of this invention is to propose a method and system for dynamic management of radiopharmaceutical activity during dispensing, which can effectively solve the prominent defects of existing methods in terms of insufficient dynamic activity perception, low intelligence in dispensing scheduling, and lack of ability to cope with complex dispensing environments.

[0008] To achieve the above objectives, a method for dynamic management of radiopharmaceutical activity during dispensing is provided in a first aspect of the present invention, the method comprising the following steps:

[0009] Construct a multidimensional security dynamic graph G = (V, E, M, F), where node V represents a physical or functional area, edge E represents a physical transit path, M represents the set of business attributes of nodes and edges, and F is a reserved physical-liveness attribute interface;

[0010] Based on the multidimensional safety dynamic graph G=(V,E,M,F), time-series aggregation analysis is performed to predict the activity changes of each packaging node and generate node activity prediction values.

[0011] Based on the predicted node activity values ​​and the multidimensional safety dynamic graph G=(V,E,M,F), a multi-objective analysis is performed to generate a scheduling strategy for the sub-packaging task, including the sub-packaging sequence of each task node, the transportation path of each sub-packaging task, and the sub-packaging time window.

[0012] The generated scheduling strategy is simulated and verified, anomalies or potential risks are corrected, and a corrected scheduling strategy is generated.

[0013] The repackaging task is executed using the revised scheduling strategy as the control signal. Real-time data collection of activity, transportation route risk, and time information is used to form a feedback dataset, which is then used for intelligent optimization.

[0014] Preferably, the multidimensional security dynamic graph G = (V, E, M, F) specifically includes:

[0015] Node attributes include spatial location, controlled area identifier, industrial or automated operation area, and node task capacity;

[0016] The edge attributes include the transportation path between two nodes in the packaging process, including transportation time, physical distance of the path, transportation method and whether the path crosses a controlled area, and security risk penalty items;

[0017] The physical-activity coupling interface, embedded in the edge attributes, is used to quantify the relationship between activity decay and time loss between nodes.

[0018] Preferably, the safety risk penalty item is used to impose avoidance or scheduling penalties on high radiation risk paths to improve the radiation safety of the packaging process.

[0019] Preferably, the step of performing time-series aggregation analysis based on the multidimensional safety dynamic graph G=(V,E,M,F) to predict the activity changes of each packaging node and generate node activity prediction values ​​specifically involves:

[0020] Based on the topological relationships of the multidimensional security dynamic graph G=(V,E,M,F), and utilizing the historical activity state, physical transport time, and path decay factor of adjacent nodes, the system completes the analysis of node v. i The activity prediction at future time t is expressed as:

[0021]

[0022] in, For node activity, Indicates adjacent node v j At time tt ji Historical activity; t ji For path e ji Transportation time; Θ ji It serves as a physical-activity coupling interface, taking into account physical distance, transportation time, and environmental impact; w ij The activity propagation weights obtained from graph structure learning are used to model adjacent nodes v j to node v i Activity affects strength;

[0023] The multidimensional security dynamic graph G = (V, E, M, F) also includes an uncertainty regularization mechanism for estimating the activity confidence interval; the uncertainty regularization mechanism is the confidence interval of the corresponding node, which is obtained by summing the squares of the physical-activity coupling interface corresponding to each node in the neighborhood.

[0024] The node activity and the corresponding confidence interval are used to form the node activity prediction value.

[0025] Preferably, the step of performing multi-objective analysis based on the predicted node activity value and the multi-dimensional security dynamic graph G=(V,E,M,F) to generate a scheduling strategy for the packaging task specifically includes:

[0026] Dynamically adjust node attributes based on the node activity and corresponding confidence intervals to generate scheduling priority scores;

[0027] The cumulative safety risks and transportation time of the path from the starting point to the target node are collected from the multidimensional safety dynamic graph G=(V,E,M,F). Through multi-factor comprehensive evaluation, a comprehensive safety-timeliness score is obtained.

[0028] The scheduling task is executed based on the scheduling priority score and the comprehensive safety-timeliness score.

[0029] Preferably, the step of executing the scheduling task based on the scheduling priority score and the safety-timeliness comprehensive score specifically includes:

[0030] The execution order of generating the packaging task {v i1 →v i2 →v in} Nodes with higher priority scores are prioritized for scheduling;

[0031] For each task node v i Select the transportation route with the best overall safety and timeliness score;

[0032] Assign dynamic packaging time windows based on node activity to packaging tasks, and combine path transportation time and node activity decay characteristics to ensure that tasks are executed within the effective activity range.

[0033] Preferably, the step of simulating and verifying the generated scheduling strategy, correcting anomalies or potential risks, and generating a corrected scheduling strategy specifically includes:

[0034] For each task node v i Perform the following verification:

[0035]

[0036] in, For node v i Simulated activity; A min This is the lower limit of activity safety. This represents the actual transportation time in the simulation. The transportation time set in S; R pi Risk value for the selected path in S; R max The maximum permissible path risk threshold set for the packaging system; As an indicator function, verify whether three types of exceptions exist;

[0037] When Ψ(v) i When the value is greater than 0, the system determines that there is a simulation anomaly at that node, enters a local correction phase, and adjusts the sub-assembly task v based on simulation feedback. i The specific scheduling time or path p i Make minor adjustments.

[0038] Preferably, the feedback dataset includes: the difference between the actual activity and the node activity in the scheduling strategy, and the deviation index between the actual path risk and the scheduling plan risk.

[0039] A second aspect of the present invention provides a radiopharmaceutical dispensing activity dynamic management system, the system comprising:

[0040] Data acquisition and modeling module: used to construct a multi-dimensional security dynamic graph G = (V, E, M, F), where node V represents a physical or functional area, edge E represents a physical transit path, M represents the set of business attributes of nodes and edges, and F is a reserved physical-activity attribute interface;

[0041] Activity prediction module: used to perform time-series aggregation analysis based on the multidimensional safety dynamic graph G=(V,E,M,F) to predict the activity changes of each packaging node and generate node activity prediction values;

[0042] Scheduling strategy generation module: Based on the node activity prediction value and the multi-dimensional security dynamic graph G=(V,E,M,F), it performs multi-objective analysis to generate the scheduling strategy for the packaging task, including the packaging order of each task node, the transportation path of each packaging task and the packaging time window.

[0043] Simulation verification module: used to simulate and verify the generated scheduling strategy, correct anomalies or potential risks, and generate a corrected scheduling strategy;

[0044] The packaging execution module is used to execute packaging tasks with the modified scheduling strategy as the control signal, collect activity, transportation route risk and time information in real time, form a feedback dataset, and perform intelligent optimization based on the feedback dataset.

[0045] The beneficial technical effects of the present invention are at least as follows:

[0046] To address the aforementioned problems, this invention proposes an intelligent system for dynamic activity management in radiopharmaceutical dispensing, effectively solving the prominent shortcomings of existing methods in areas such as insufficient dynamic activity perception, low intelligence in dispensing scheduling, and inadequate ability to cope with complex dispensing environments. Starting from the system level, this invention constructs a dynamic activity perception mechanism that runs through the entire dispensing process, enabling real-time monitoring of the activity distribution status at each node in the dispensing chain, overcoming the limitations of traditional single-point measurement and static estimation in modeling dynamic changes. Based on this, this invention designs an intelligent optimization mechanism for multi-objective dispensing management, achieving a dynamic balance between maximizing activity utilization, minimizing personnel radiation exposure, and optimizing dispensing scheduling efficiency, comprehensively improving the intelligence level and safety reliability of the dispensing process. Simultaneously, this invention constructs a perception-decision integrated dynamic scheduling system, achieving bidirectional closed-loop control of activity distribution prediction and dispensing decision-making through real-time linkage between the activity perception module and the scheduling optimization module. This system can flexibly respond to complex situations such as environmental fluctuations, personnel scheduling changes, and activity anomalies during the dispensing process, possessing strong system robustness and adaptability. Through the above innovative design, this invention can significantly improve the accuracy of activity management, the intelligence of dispensing scheduling, and the operational safety in the radiopharmaceutical dispensing process, and fully meet the needs of dynamic activity management of radiopharmaceuticals in complex dispensing scenarios. Attached Figure Description

[0047] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0048] Figure 1 This is a flowchart of a method for dynamic management of radiopharmaceutical repackaging activity according to the present invention. Detailed Implementation

[0049] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0050] like Figure 1 As shown in the embodiment of the present invention, a method for dynamic management of radiopharmaceutical repackaging activity is provided, the method comprising:

[0051] S1. Construct a multi-dimensional security dynamic graph G = (V, E, M, F), where node V represents a physical or functional area, edge E represents a physical transit path, M represents the set of business attributes of nodes and edges, and F is a reserved physical-liveness attribute interface.

[0052] Specifically, in the task of dynamic management of radiopharmaceutical activity during repackaging, to achieve comprehensive system awareness of factors such as activity transfer, operational risks, and physical characteristics of the repackaging process, it is first necessary to construct a system topology graph with spatial, physical, safety, and activity coupling characteristics. Therefore, this step designs a multidimensional safety dynamic graph (NDSDG, Nuclear Drug Safety Dynamic Graph) as the basic modeling scheme. The system input is the actual layout data of the radiopharmaceutical repackaging center, including the physical coordinates of repackaging cabinets, buffer zones, and transfer channels, functional zoning (controlled / uncontrolled areas), and historical task flow data. By extracting this information, the repackaging center is modeled as a graph structure G = (V, E, M, F), where node V represents a physical or functional area, edge E represents a physical transfer path, M represents the set of business attributes of nodes and edges, and F is a reserved physical-activity attribute interface.

[0053] In terms of node modeling, the system treats each physical packaging unit (such as packaging cabinet A, buffer B, and transfer channel C) as a node v in the graph. i Each v i Additional attribute p i (spatial location), r i (Controlled area sign), c i (Manual or automated operation area) and li (Node task capacity). For example, node v1 of “Dispensing Cabinet A” located in the controlled radiation area has r1=1 to indicate that it is in the controlled area, c1=manual to indicate that only manual operation is allowed, and l1=3 to indicate that it supports dispensing 3 batches at the same time.

[0054] In path modeling, the edge e of the graph ij Describes the transportation path between two nodes in the packaging process, including t ij (Transportation time), d ij (path physical distance), b ij (Transportation method) and q ij (Whether the route crosses a controlled area). For example, the transport route from "Buffer Zone B" to "Container A" e 12 Because it traverses the controlled region, the value q is assigned. 12 =1.

[0055] To enhance the model's adaptability to personnel safety and the physical environment, an innovative safety risk penalty term R is introduced into the path's attributes. ij This serves as an important basis for determining path safety in subsequent activity sensing and scheduling optimization.

[0056] R ij =γ·q ij ·r i

[0057] Where γ is the penalty coefficient within the system, q ij Indicates whether the path traverses a controlled area (1 for traversal), r i Indicates the starting node v i Whether it belongs to a controlled area (1 is a controlled area). This design enables the model to dynamically apply significant strategy avoidance or scheduling penalties to high-radiation-risk paths during subsequent activity management, thereby improving the radiation safety of the packaging process.

[0058] Furthermore, considering the time sensitivity in radiopharmaceutical repackaging scenarios, the model design reserves a physical-activity coupling interface Θ. ij It is directly embedded in the attribute F of the graph edge to quantify the relationship between the activity decay and time loss between nodes:

[0059]

[0060] In the formula, τ represents the half-life (in minutes) of the processed radiopharmaceutical, and t ij For historical transportation time, d ij Θ represents the path length, and δ represents the environmental sensitivity factor, taking into account the weighted effects of temperature and humidity on activity decay along the path (e.g., δ is greater in high-temperature environments). ijAs a decay reference in the subsequent dynamic activity prediction, it solves the problem that ordinary packaging diagram models cannot reflect the multiple couplings of "space-time-activity" in the transportation path.

[0061] For example, if from "buffer B" (v2) through path e 12 To "Dispensing Container A" (v1), path e 12 The length is 20 meters, the processing time is 5 minutes, the half-life of the processed radiopharmaceutical is τ = 110 minutes, and the current ambient temperature compensation coefficient is δ = 0.02, then:

[0062] Safety penalty R 12 =γ×1×1=γ;

[0063] Activity decay interface

[0064] In this way, Figure G not only realizes the physical topology modeling of the packaging system, but also pre-integrates the activity propagation and security constraint interfaces, providing data support for the subsequent steps of "inter-node activity flow modeling" and "security-aware scheduling optimization".

[0065] S2. Based on the multidimensional safety dynamic graph G=(V,E,M,F), perform time-series aggregation analysis to predict the activity changes of each packaging node and generate node activity prediction values.

[0066] Specifically, based on the multidimensional safety dynamic graph G=(V,E,M,F) established in step 1, this step focuses on completing the dynamic perception and prediction of activity within the radiopharmaceutical dispensing system. The goal is to construct a spatiotemporal distribution prediction model of the activity of each dispensing node in the system based on the topological relationship, physical attributes, and activity propagation interface of graph G, so as to provide accurate activity status input for subsequent scheduling strategies.

[0067] Radiopharmaceutical dispensing systems exhibit typical characteristics such as "dynamic activity flow across nodes," "time-consuming physical path coupling transport," and "high time sensitivity." To address these issues, this step proposes a "Physically Guided Graph Activity Prediction Mechanism (PE-GAPM)," which uses nodes v in graph G as the basis for prediction. i ∈V represents the prediction target, based on the activity propagation interface Θ in the graph structure. ij By combining the historical activity of nodes, the dynamic inference of system activity can be completed.

[0068] The model first uses the topological relationships of graph G, utilizing the historical activity state of adjacent nodes, physical transport time, and path decay factor, to complete the analysis of node v. i Activity prediction at a future time t. The specific definition is as follows:

[0069]

[0070] in, Indicates adjacent node v j At time tt ji Historical activity; t ji For path e ji Transportation time; Θ ji The activity propagation interface defined in step 1 is combined with physical distance, transportation time, and environmental impact; w ij The activity propagation weights obtained from graph structure learning are used to model v j to v i The activity affects the strength.

[0071] Through the above formula, the model realizes a cross-node activity dynamic aggregation mechanism based on "transport path - activity physical decay - graph structure", which can capture the activity flow relationship of radiopharmaceuticals across time and space in the packaging system, and conforms to the objective law that radiopharmaceutical activity decays with changes in transportation time and physical environment.

[0072] Due to the uncertainty of activity in the packaging system caused by environmental fluctuations and the diversity of transportation routes, this invention further designs a graph-based physical-environment-aware activity uncertainty regularization mechanism, specifically for estimating activity confidence intervals. It is defined as follows:

[0073]

[0074] Where η is the regularity coefficient, controlling the magnitude of uncertainty; Θ ji For path e ji The activity propagation interface, whose squared-term enhancement model perceives the activity uncertainty of "high decay, high time consumption" paths.

[0075] For example, when "buffer zone B" (v2) transports radiopharmaceuticals to "dispensing container A" (v1), t 12 =5 minutes, Θ 12 =0.5, then and Predicted output This activity status is used by the subsequent scheduling strategy generation module for reference.

[0076] Through the above mechanism, the model realizes the ability to dynamically perceive activity across multiple nodes, paths, and time scales in the packaging system, outputs a complete system activity distribution prediction map, and directly connects to the subsequent packaging scheduling and optimization module, serving the integrated "perception-decision-execution-feedback" dynamic activity management system in the patent.

[0077] S3. Based on the predicted node activity value and the multidimensional safety dynamic graph G=(V,E,M,F), perform multi-objective analysis to generate a scheduling strategy for the sub-packaging task, including the sub-packaging sequence of each task node, the transportation path of each sub-packaging task, and the sub-packaging time window.

[0078] Specifically, this step uses the node activity prediction output from step 2. Using the system graph G = (V, E, M, F) constructed in step 1 as input, a multi-objective scheduling strategy for the radiopharmaceutical dispensing task is generated. The dispensing process requires the system to rationally arrange the dispensing sequence, transportation path, and time schedule of each node under conditions of dynamic activity changes, restricted transportation paths, and controllable safety risks. Therefore, this step proposes an "Activity-Pathway-Safety Multi-Objective Optimization Scheduling Mechanism (APSM)," specifically designed for generating multi-objective scheduling strategies in the radiopharmaceutical dispensing scenario.

[0079] Input data includes node activity. and its confidence interval In Figure G, the system considers the security level, transportation time, and physical topology of nodes and paths, as well as the target node {v} of the sub-packaging task. target To constrain this, an innovative multi-objective scoring mechanism of "activity-risk-timeliness" was designed for scheduling order and path decision-making.

[0080] The scheduling priority scoring function is:

[0081]

[0082] in, For node v i Current predicted activity β represents the uncertainty of activity, and β is the uncertainty sensitivity coefficient. This scoring mechanism ensures that the scheduling strategy prioritizes target nodes with high activity and strong predictive stability, which meets the requirements of radiopharmaceutical repackaging tasks for controllable activity.

[0083] Based on this, the system designed a "safety-timeliness integrated score" function for the transportation route to determine the transportation route for the repackaging task:

[0084]

[0085] Among them, R pi For path p i From the starting point to the target node v i The path accumulates security risks (derived from the R of graph G). ij ), T pi The route transportation time is represented by γ, and λ is the weighting factor for safety risk and transportation timeliness, respectively. This scoring ensures that the scheduling system dynamically balances the radiation safety of the route and transportation time when planning the distribution route, and automatically selects efficient and low-risk routes under tasks that are sensitive to activity decay and have tight time constraints.

[0086] Based on the above two-level scoring mechanism, the system completes the following scheduling tasks:

[0087] Execution order of generating packaging tasks Priority scheduling U(v) i Nodes with high scores;

[0088] For each task node v i Select Score(p) i The optimal transportation route p i ;

[0089] Assigning packaging tasks based on The dynamic packaging time window, combined with the transportation time T pi And the node activity decay characteristic ensures that tasks are executed within the effective activity range.

[0090] For example, if the current activity prediction of node v1 R of path p1 p1 =0.8, T p1 = 8 minutes, coefficients β = 0.5, γ = 1, λ = 0.05, then:

[0091] Node priority U(v1) = 50 - 0.5 × 0.5;

[0092] Path score Score(p1) = e -0.8-0.4 =e -1.2 .

[0093] By combining two levels of scoring, the system prioritizes repackaging tasks that have high activity utilization, safe routes, and efficient transportation.

[0094] Finally, the scheduling policy S output by APSM includes:

[0095] The order in which each task node is packaged;

[0096] The transportation path p for each repackaging task i ;

[0097] The packaging time window is the optimal execution time interval based on activity prediction.

[0098] This step focuses on "strategy generation," providing a complete scheduling plan for subsequent scheduling execution (step 4), and ensuring that the input (activity state, graph structure) and output (scheduling scheme) are highly consistent, strictly conforming to the chain process of "perception-decision-execution."

[0099] S4. Simulate and verify the generated scheduling strategy, correct any anomalies or potential risks, and generate a revised scheduling strategy.

[0100] Specifically, this step focuses on systematically simulating and verifying the scheduling strategy S generated in step 3. For any anomalies or potential risks discovered during the verification process, limited strategy modifications are performed to ensure the physical feasibility, activity effectiveness, and transportation safety of the repackaging scheme within the radiopharmaceutical repackaging system. The simulation verification inputs include the complete scheduling strategy S generated in step 3 and the predicted node activity values ​​output in step 2. confidence interval It also relies on the physical constraints and path attributes provided by the graph G=(V,E,M,F) established in step 1.

[0101] During the simulation, the system reproduces the assembly process task by task and path by path according to the assembly sequence, assembly path, and time schedule given in S. This is achieved by simulating each assembly task node v during the assembly process. i The simulated activity is calculated based on the transportation and operational behavior. And compare with v i Predicted activity In conjunction with safety indicators, examine whether there are any unreasonable situations such as insufficient activity, mission timeout, or high-risk paths.

[0102] The system defines a "simulation consistency verification index" for each task node v. i Perform the following verification:

[0103]

[0104] in, For node v i The simulated activity, based on step 2 By path p i The transit time and Θ ij Attenuation interface dynamic calculation; A min The activity safety limit (e.g., 20 mCi) is derived from industry standards. This represents the actual transportation time in the simulation. The transportation time set in S; R pi Risk value for the selected path in S; R max The maximum permissible path risk threshold set for the packaging system; As an indicator function, verify the existence of three types of anomalies (insufficient activity, transit timeout, and excessively high route risk).

[0105] When Ψ(v) i When the value is greater than 0, the system determines that there is a simulation anomaly at that node and enters the "local correction mechanism". Based on the simulation feedback, the system adjusts the sub-assembly task v. i The specific scheduling time or path p i Make minor adjustments. For example, if in the simulation... The system will either move the task execution time forward or recommend an alternative path based on the original strategy.

[0106] During the simulation phase, the system will also output a task-level simulation verification report, including information such as whether the activity of task nodes is compliant, whether the path exceeds the risk threshold, and whether the transportation time meets the activity requirements for repackaging, for use in the subsequent repackaging execution phase.

[0107] For example, if the simulated activity of the packaging cabinet A (v1) Below A min =20mCi, while simulating transportation time Minutes beyond the plan If the time is less than 1 minute, then Ψ(v1) = 2 > 0, and the system will mark the task as abnormal in the simulation results.

[0108] The final output of this step is the scheduling strategy S, which has been verified by simulation and locally corrected. * As the input for the sub-assembly execution in step 5, it ensures that the system fully verifies and guarantees the safety, timeliness and activity management requirements of S in a physical simulation environment.

[0109] S5. The repackaging task is executed using the modified scheduling strategy as the control signal. Real-time collection of activity, transportation route risk and time information is generated to form a feedback dataset. Intelligent optimization is performed based on the feedback dataset.

[0110] Specifically, this step uses the simulation-corrected scheduling strategy S output in step 4. * As input, it guides the radiopharmaceutical dispensing system to perform dispensing tasks in a real environment. During the execution process, it collects information on activity, transportation path risks, and key time information in real time during the dispensing operation, forming a systematic feedback dataset that provides closed-loop support for the entire process of "dynamic activity management and scheduling execution" of this patent.

[0111] The system first sets S * The order of the packaging tasks, path selection, and packaging time window are translated into specific packaging control instructions, which cover task nodes v i Dispensing start time End time Actual transportation route p i and transportation time The automated control module of the packaging system is based on S... * This gradually drives the packaging equipment to complete the actual packaging, transportation, and handover tasks at each node.

[0112] During the dispensing process, the system simultaneously collects data on the v at each dispensing node in real time through activity detection sensors and a risk monitoring module. i Actual activity value Route risk exposure data during transportation The actual completion time of each task node forms a complete execution feedback data stream.

[0113] To provide a standardized execution feedback mechanism, the system defines the "execution-scheduling activity difference index" Γ(v i ), to measure true activity With scheduling strategy S * Mid-expected activity Differences:

[0114]

[0115] in, For node v collected when the packaging is complete i Actual activity The activity prediction value is input to the scheduling strategy. This difference serves as the basis for subsequent feedback in the system, reflecting the degree of agreement between the activity prediction and the actual activity during execution.

[0116] In addition, the system also collects the "deviation index between actual path risk and scheduling plan risk" Ω(v i This is used to reflect the risk fluctuations during the execution of transportation routes.

[0117]

[0118] in, For real-time monitoring of risks in transportation missions, R pi For S * The path risk value set in the scheduling.

[0119] For example, when the system completes the disassembly process in disassembly cabinet A (v1), it records... Scheduling settings Then Γ(v1)=-1mCi; if And planned risk R p1 =0.8, then Ω(v1) =0.05, and the system automatically archives it as standard feedback data.

[0120] Through this step, the packaging system has fully realized the processing of S * The physical execution of the strategy and high-quality data feedback complete the technical closed loop of "perception-decision-verification-execution-feedback" in the patent design, ensuring the activity utilization rate, safety and system stability of the radiopharmaceutical dispensing system in practical applications.

[0121] This invention also provides a radiopharmaceutical dispensing activity dynamic management system, the system comprising:

[0122] Data acquisition and modeling module: used to construct a multi-dimensional security dynamic graph G = (V, E, M, F), where node V represents a physical or functional area, edge E represents a physical transit path, M represents the set of business attributes of nodes and edges, and F is a reserved physical-activity attribute interface;

[0123] Activity prediction module: used to perform time-series aggregation analysis based on the multidimensional safety dynamic graph G=(V,E,M,F) to predict the activity changes of each packaging node and generate node activity prediction values;

[0124] Scheduling strategy generation module: Based on the node activity prediction value and the multi-dimensional security dynamic graph G=(V,E,M,F), it performs multi-objective analysis to generate the scheduling strategy for the packaging task, including the packaging order of each task node, the transportation path of each packaging task and the packaging time window.

[0125] Simulation verification module: used to simulate and verify the generated scheduling strategy, correct anomalies or potential risks, and generate a corrected scheduling strategy;

[0126] The packaging execution module is used to execute packaging tasks with the modified scheduling strategy as the control signal, collect activity, transportation route risk and time information in real time, form a feedback dataset, and perform intelligent optimization based on the feedback dataset.

[0127] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0128] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of apparatuses or units may be electrical, mechanical, or other forms.

[0129] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0130] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for dynamic management of radiopharmaceutical repackaging activity, characterized in that, The method includes the following steps: Constructing a multi-dimensional security dynamic graph ,node Represents a physical or functional area, edge Indicates the physical transport path, This represents the set of business attributes of nodes and edges. For the reserved physical-activity attribute interface; Based on the multidimensional security dynamic graph Perform time-series aggregation analysis to predict the activity changes of each packaging node and generate node activity prediction values; Based on node activity prediction values ​​and multidimensional security dynamics graph Multi-objective analysis is performed to generate a scheduling strategy for the packaging tasks, including the packaging order of each task node, the transportation path of each packaging task, and the packaging time window. The generated scheduling strategy is simulated and verified, anomalies or potential risks are corrected, and a corrected scheduling strategy is generated. The repackaging task is executed using the revised scheduling strategy as the control signal. Real-time data collection of activity, transportation route risk, and time information is used to form a feedback dataset, which is then used for intelligent optimization.

2. The method for dynamic management of radiopharmaceutical repackaging activity according to claim 1, characterized in that, The multidimensional security dynamic diagram Specifically, it includes: Node attributes include spatial location, controlled area identifier, manual or automated operation area, and node task capacity; The edge attributes include the transportation path between two nodes in the packaging process, including transportation time, physical distance of the path, transportation method and whether the path crosses a controlled area, and security risk penalty items; The physical-activity coupling interface, embedded in the edge attributes, is used to quantify the relationship between activity decay and time loss between nodes.

3. The method for dynamic management of radiopharmaceutical repackaging activity according to claim 2, characterized in that, The aforementioned safety risk penalty item is used to impose avoidance or scheduling penalties on high radiation risk paths, thereby improving the radiation safety of the packaging process.

4. The method for dynamic management of radiopharmaceutical repackaging activity according to claim 2, characterized in that, The multi-dimensional security dynamic graph Perform time-series aggregation analysis to predict the activity changes at each packaging node and generate predicted node activity values, specifically: Based on multidimensional security dynamic graph The topological relationships are used to complete the node analysis by utilizing the historical activity state, physical transport time, and path decay factor of adjacent nodes. In the future The activity prediction is expressed as: ； in, For node activity, Indicates adjacent nodes At any moment Historical activity; For path Transportation time; It serves as a physical-activity coupling interface, taking into account physical distance, transportation time, and environmental impact. The activity propagation weights obtained from graph structure learning are used to model adjacent nodes. To the node Activity affects strength; The multidimensional security dynamic diagram It also includes an uncertainty regularization mechanism for estimating the activity confidence interval; the uncertainty regularization mechanism is the confidence interval of the corresponding node, which is obtained by summing the squares of the physical-activity coupling interfaces of each node in the adjacent nodes. The node activity and the corresponding confidence interval are used to form the node activity prediction value.

5. The method for dynamic management of radiopharmaceutical repackaging activity according to claim 4, characterized in that, The method is based on node activity prediction values ​​and multidimensional security dynamic graphs. Multi-objective analysis is performed to generate a scheduling strategy for the packaging task, specifically including: Dynamically adjust node attributes based on the node activity and corresponding confidence intervals to generate scheduling priority scores; From the multidimensional security dynamic diagram The data collection process accumulates safety risks and transportation time along the route from the origin to the target node. Through a comprehensive evaluation of multiple factors, a safety-timeliness integrated score is obtained. The scheduling task is executed based on the scheduling priority score and the comprehensive safety-timeliness score.

6. The method for dynamic management of radiopharmaceutical repackaging activity according to claim 5, characterized in that, The step of executing the scheduling task based on the scheduling priority score and the safety-timeliness comprehensive score specifically includes: Execution order of generating packaging tasks Nodes with higher priority scores will be scheduled first. For each task node Select the transportation route with the best overall safety and timeliness score; Assign dynamic packaging time windows based on node activity to packaging tasks, and combine path transportation time and node activity decay characteristics to ensure that tasks are executed within the effective activity range.

7. The method for dynamic management of radiopharmaceutical repackaging activity according to claim 1, characterized in that, The process of simulating and verifying the generated scheduling strategy, correcting anomalies or potential risks, and generating a corrected scheduling strategy specifically includes: For each task node Perform the following verification: ；； in, For nodes Simulated activity; This is the lower limit of activity safety. This represents the actual transportation time in the simulation. For scheduling strategy The transportation time set in the system; For scheduling strategy Risk value of the selected path; The maximum permissible path risk threshold set for the packaging system; As an indicator function, verify whether three types of exceptions exist; when When this occurs, the system determines that a simulation anomaly exists at that node, initiates local correction, and adjusts the specific scheduling time or path for the assembly task based on simulation feedback. Make minor adjustments.

8. The method for dynamic management of radiopharmaceutical repackaging activity according to claim 1, characterized in that, The feedback dataset includes: the difference between the actual activity level and the node activity level in the scheduling strategy, and the deviation index between the actual path risk and the scheduling plan risk.

9. A dynamic management system for radiopharmaceutical dispensing activity, characterized in that, The system includes: Data acquisition and modeling module: used to construct multidimensional security dynamic graphs. ,node Represents a physical or functional area, edge Indicates the physical transport path, This represents the set of business attributes of nodes and edges. For the reserved physical-activity attribute interface, Activity prediction module: used to predict activity based on the multidimensional safety dynamic graph. Perform time-series aggregation analysis to predict the activity changes of each packaging node and generate node activity prediction values; Scheduling strategy generation module: used to generate scheduling strategies based on node activity predictions and multi-dimensional security dynamic graphs. Multi-objective analysis is performed to generate a scheduling strategy for the packaging tasks, including the packaging order of each task node, the transportation path of each packaging task, and the packaging time window. Simulation verification module: used to simulate and verify the generated scheduling strategy, correct anomalies or potential risks, and generate a corrected scheduling strategy; The packaging execution module is used to execute packaging tasks with the modified scheduling strategy as the control signal, collect activity, transportation route risk and time information in real time, form a feedback dataset, and perform intelligent optimization based on the feedback dataset.

Images