Constraint-driven scheduling method and system for surgical tasks

Abstract

The application provides a surgery task constraint-driven scheduling method and system, and relates to the technical field of task scheduling and process control in a medical surgery auxiliary system. On the basis of having obtained a surgery task set, the application generates a task execution strategy that can be called by a surgery execution system by comprehensively considering the task set, resource state, constraint condition, real-time state and scheduling parameter, and realizes unified organization and sequencing of multiple tasks in a surgery process by constructing a scheduling function model, so that the task execution process is changed from being driven by a fixed flow to being driven by scheduling control.

Description

Technical Field

[0001] This invention relates to the field of task scheduling and process control technology in medical surgical assistance systems, specifically to a constraint-driven scheduling method and system for surgical tasks. Background Technology

[0002] In the field of digital surgical assistance, surgical process control and intraoperative task scheduling are key underlying business functions supporting the implementation of intelligent surgery. This business mainly focuses on the arrangement of operation steps for the entire surgical process, the management of intraoperative sub-tasks, and the unified control of the surgical execution process. It is widely used in various surgical clinical scenarios and can provide core technical support for preoperative personalized surgical planning, intraoperative automated operation guidance, and standardized control of the entire surgical process. At the same time, it ensures that various surgical operations are carried out in an orderly and safe manner during the operation. It is an indispensable core business module for realizing intelligent closed-loop control during surgery.

[0003] Currently, clinical surgical procedures generally rely on digitally assisted systems to complete the entire process from preoperative planning to intraoperative execution. Existing digital surgical assistance systems use the patient's preoperative medical imaging data as a basis to plan the surgical path and generate standardized surgical execution plans. During the operation, these systems, in conjunction with navigation equipment and automated surgical execution devices, provide real-time guidance and closed-loop control for the surgeon's actions. Overall, existing digital surgical assistance technologies are designed and operated according to a linear processing model of path generation, fixed-step execution, and intraoperative status feedback. Specifically, existing technologies can be divided into three main implementation forms: The first type is an auxiliary system for generating surgical procedures and controlling local execution. It generates fixed surgical operation procedures based on preset rules or historical surgical data, completes the surgical execution according to predetermined steps, and achieves single-process status feedback and local control by relying on sensing and navigation information. The second type is a process management and resource scheduling method for the overall medical business of the hospital. It mainly focuses on macro-level offline scheduling optimization of preoperative surgical scheduling, operating room and medical resources, and improves the overall operational efficiency of the hospital's surgeries. The third type is a general task relationship modeling and scheduling algorithm. It realizes the structured expression of the process by constructing task dependencies and decomposing the surgical workflow, and optimizes the task execution sequence by relying on a general scheduling algorithm.

[0004] However, the existing surgical assistance technologies described above, which follow fixed linear processes and are designed independently for different scenarios, have clear and prominent technical shortcomings in actual complex clinical surgical applications. All existing surgical assistance solutions lack the ability to coordinate and schedule multiple tasks collaboratively during surgery. Whether it's an intraoperative local process control system, a preoperative macro-resource scheduling scheme, or a general process modeling algorithm, they can only adapt to a single, linear surgical operation process. They haven't designed dedicated scheduling mechanisms for parallel tasks or time-dependent related tasks during surgery, and cannot comprehensively manage the execution order and coordination logic of multiple surgical sub-tasks during surgery. Faced with complex scenarios commonly encountered in clinical surgery, such as multi-task parallelism, temporary switching of surgical steps, and dynamic changes in intraoperative conditions, existing technologies are prone to problems such as disordered execution sequence of surgical tasks, conflicts between different operation steps, and disjointed intraoperative operations, making it difficult to adapt to the real-world operational needs of complex surgical procedures. Summary of the Invention

[0005] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a constraint-driven scheduling method and system for surgical tasks, which solves the technical problem that existing surgical assistance solutions lack the ability to coordinate and schedule multiple tasks collaboratively during surgery.

[0006] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides a constraint-driven scheduling method for surgical tasks, comprising: Obtain the task set, resource status, constraints, real-time status, and scheduling parameters; The task priority is calculated based on the task set, constraints, real-time status, and scheduling parameters; the task set is then filtered according to the constraints to obtain a set of executable candidate tasks. Allocate available resources to each task in the set of executable candidate tasks based on resource status; After resource matching is completed, resource conflict, execution conflict and sequence conflict detection are performed on multiple tasks in the set of executable candidate tasks. When a conflict is detected, the conflict is resolved according to task priority, resource status and constraints. A task execution strategy is generated based on the conflict resolution results, resource matching results, constraints, real-time status, and scheduling parameters.

[0007] Preferably, the step of calculating task priority based on task set, constraints, real-time status, and scheduling parameters includes: For each task in the task set Calculate priority value The formula for calculating the priority value is as follows: in, Indicates task dependency; Indicates the urgency of the task; Indicates the degree of influence of the constraint. Indicates the degree of impact of real-time status; , , and Indicates the scheduling weight parameter; Task dependency The determination is based on a comprehensive assessment of the number of preceding tasks, the number of succeeding tasks, whether the task is on a critical task path, and the impact of the task's completion on subsequent tasks within the task set. Mission urgency It can be determined based on the current surgical stage, task waiting time, task preset execution window, whether the task affects subsequent key steps, and any abnormalities or deviations reflected in the real-time status. Degree of influence of constraints The degree of restriction imposed on the task by the safety constraints, accuracy constraints, and resource constraints in the constraints is determined; Real-time status impact Determined based on the current deviation, abnormal state, or degree of resource change in real-time status X.

[0008] Preferably, the step of allocating available resources to each task in the set of executable candidate tasks according to the resource status includes: Based on the resource status, allocate available resources to the set of executable candidate tasks and determine the corresponding equipment, instruments, execution terminals or computing resources for each task.

[0009] Preferably, the conflict resolution methods include: adjusting the task execution order, delaying conflicting tasks, reallocating resources, suspending low-priority tasks, or triggering rescheduling.

[0010] Preferably, the execution strategy includes task execution order, resource allocation results, execution conditions, control requirements, and exception handling requirements.

[0011] Preferably, the constraint-driven scheduling method further includes: When the real-time state changes and the triggering condition is met, the scheduling function F is recalculated to generate an updated task execution strategy A′; its expression is as follows: A′=F(T,R,C,X;θ) Where F represents the constraint-driven scheduling function; T represents the current task set; R represents the current resource state; C represents the constraint condition; X represents the real-time state; and θ represents the scheduling parameters.

[0012] Preferably, the triggering conditions include: execution deviation exceeding a preset threshold, resource status change, new task addition, task cancellation, abnormal device status, abnormal navigation status, or critical task execution failure.

[0013] Secondly, the present invention provides a constraint-driven scheduling system for surgical tasks, comprising: The scheduling input acquisition module is used to acquire task sets, resource status, constraints, real-time status, and scheduling parameters; The priority calculation and constraint filtering module is used to calculate task priorities based on task sets, constraints, real-time status, and scheduling parameters; and to filter the task set according to constraints to obtain a set of executable candidate tasks. The resource matching module is used to allocate available resources to each task in the set of executable candidate tasks based on the resource status; The conflict detection and resolution module is used to detect resource conflicts, execution conflicts, and sequence conflicts among multiple tasks in the set of executable candidate tasks after resource matching is completed, and to resolve conflicts based on task priority, resource status, and constraints when a conflict is detected. The strategy generation module is used to generate task execution strategies based on conflict resolution results, resource matching results, constraints, real-time status, and scheduling parameters.

[0014] Thirdly, the present invention provides a computer-readable storage medium storing a computer program for constraint-driven scheduling for surgical tasks, wherein the computer program causes a computer to execute the constraint-driven scheduling method for surgical tasks as described above.

[0015] Fourthly, the present invention provides an electronic device, comprising: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including a constraint-driven scheduling method for performing surgical tasks as described above.

[0016] (III) Beneficial Effects This invention provides a constraint-driven scheduling method and system for surgical tasks. Compared with existing technologies, it has the following advantages: Based on the already obtained set of surgical tasks, this invention integrates the task set, resource status, constraints, real-time status, and scheduling parameters to generate a task execution strategy that can be called by the surgical execution system. By constructing a scheduling function model, it enables the unified organization and sorting of multiple tasks in the surgical process, transforming the task execution process from being driven by a fixed process to being driven by scheduling control. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a block diagram of a constraint-driven scheduling method for surgical tasks according to an embodiment of the present invention; Figure 2 This is a flowchart for dynamic rescheduling. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are described clearly and completely. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] This application provides a constraint-driven scheduling method and system for surgical tasks, which solves the technical problem that existing surgical assistance solutions lack the ability to coordinate and schedule multiple tasks collaboratively during surgery. By constructing a scheduling function model, multiple tasks in the surgical process are uniformly organized and sorted, so that the task execution process is transformed from fixed process-driven to scheduling control-driven.

[0021] The technical solution in this application is to solve the above-mentioned technical problems, and the general idea is as follows: In actual clinical applications, existing technical solutions still have many shortcomings in terms of surgical task scheduling and execution control, specifically in the following aspects: (1) Lack of a unified scheduling mechanism for the surgical execution process Existing technical solutions (such as Chinese Patent CN117500451A, "System, Method and Apparatus for Dental Implant Surgery Using Kinematic Data Augmentation," published on February 2, 2024; and Chinese Patent CN121075558A, "A Method and Apparatus for Automatic Management of Surgical Procedures Using Multi-Sensor Fusion," published on December 5, 2025) mainly achieve surgical operation control through process generation or path planning. Their typical execution flow can be summarized as: process generation → path execution → status feedback. These systems focus on the execution of a single operation path or independent step, and the control logic is mostly based on preset processes or fixed rules.

[0022] In this implementation, the surgical procedure can only be executed in a fixed sequence of steps, lacking a unified and global scheduling mechanism among the various operational units. Therefore, when the surgery involves multiple parallel or related operational tasks and the process structure is complex, the system struggles to coordinate and control the task execution order in a unified manner, significantly limiting the organization and operational efficiency of the overall surgical process.

[0023] (2) Existing scheduling methods are mainly geared towards macro-level scheduling and are difficult to apply directly to surgical execution control. Existing related research (such as Chinese Patent CN121171520A, "Intelligent Scheduling Management Method and its Intelligent Agent for Multiple Business Processes in Hospitals," published on December 19, 2025) mainly focuses on modeling surgical scheduling or medical resource allocation problems. Their scheduling objectives typically concentrate on optimizing surgical time arrangements and improving resource utilization. The scheduling objects of these methods are mainly macro-level tasks at the surgical or departmental level. The scheduling process is usually completed offline before surgery, and the scheduling results are only used for the planning and arrangement of the overall business process.

[0024] In this situation, the scheduling results are difficult to dynamically adjust and update in real time during the surgical procedure, and cannot be directly applied to the fine-grained control of intraoperative operations, resulting in limited applicability of this type of method in surgical procedures with high real-time requirements.

[0025] (3) Lack of comprehensive scheduling capability under multiple constraints In actual surgical procedures, task execution is often constrained by a combination of factors, including operational precision, safety, and equipment status. However, existing surgical assistance systems (such as Chinese Patent CN121075558A, "An Automatic Management Method and Device for Surgical Processes Based on Multi-Sensor Fusion," published on December 5, 2025) primarily focus on image data processing or basic process execution, while traditional scheduling methods mainly focus on resource allocation optimization. Various solutions typically optimize independently only for a single objective or local constraints.

[0026] Therefore, in complex surgical scenarios with multiple constraints, existing technologies cannot uniformly determine task execution priorities and lack coordination and handling mechanisms for conflicts between different constraints, thus limiting the rationality and clinical feasibility of scheduling results.

[0027] (4) Lack of scheduling decision-making mechanism oriented towards task conflicts During the parallel execution of multiple tasks, resource contention or execution logic conflicts can easily arise between different tasks. For example, multiple tasks may compete for the same equipment resources simultaneously, or there may be mutual exclusion relationships between different operation steps. Existing surgical workflow systems typically execute operations in a fixed order, and traditional scheduling methods often rely on preset rules or offline optimization results.

[0028] When conflicts actually occur, the system lacks a unified mechanism for conflict identification, assessment, and automatic resolution, and most cases can only be handled through manual intervention. Therefore, in complex surgical scenarios, existing technologies struggle to achieve stable and reliable automated scheduling decisions.

[0029] (5) Lack of dynamic scheduling capability based on real-time status During surgery, the system state and surgical environment may change at any time, such as changes in patient position, displacement of surgical instruments, and gradual accumulation of execution errors. Although existing methods can achieve surgical risk monitoring, their application is limited to anomaly detection and cannot be directly used for intraoperative scheduling and control.

[0030] In existing technologies, surgical procedures mainly rely on pre-set pathways, and scheduling plans are typically determined once before surgery. When the intraoperative state changes in real time, the system struggles to adjust the task execution order promptly, and the scheduling process lacks effective real-time feedback and dynamic correction mechanisms, resulting in a significant deficiency in the adaptability of existing methods to dynamically changing environments.

[0031] (6) Existing task scheduling methods lack effective coupling with surgical procedure modeling. In the existing technology system, process modeling and task scheduling are independent and separate from each other. The scheduling algorithm cannot be directly applied to the surgical task structure, making it difficult to form a unified "modeling-scheduling-execution" integrated operation mechanism.

[0032] In summary, while existing technologies have made some progress in process execution, task scheduling, and optimization methods, they still have significant shortcomings: a unified task scheduling model has not been established to standardize the description of the entire surgical process; a global scheduling mechanism that comprehensively considers task relationships, resource status, and multiple constraints has not been formed; and an intelligent scheduling method that can dynamically adjust based on real-time status is lacking.

[0033] In this context, existing technologies have not yet developed a unified scheduling function model for the entire surgical procedure, making it difficult to effectively organize and control multi-task, complex surgical processes.

[0034] To address the aforementioned issues, this invention provides a constraint-driven scheduling method and system for surgical tasks. By uniformly modeling the task set, resource status, constraints, and real-time status during the surgical process, a scheduling function model for generating task execution strategies is constructed, thereby achieving unified organization and control of multi-task surgical procedures.

[0035] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0036] This invention provides a constraint-driven scheduling method for surgical tasks, such as... Figure 1 As shown, it includes: S100, obtain the task set T, resource status R, constraint C, real-time status X and scheduling parameters θ; S200. Calculate task priorities based on task set T, constraint C, real-time status X, and scheduling parameter θ; filter task set T according to constraint C to obtain executable candidate task set T_e; S300. Allocate available resources to each task in the executable candidate task set T_e according to the resource status R; S400. After completing resource matching, resource conflict, execution conflict and sequence conflict detection are performed on multiple tasks in the executable candidate task set T_e. When a conflict is detected, the conflict is resolved according to the task priority, resource status R and constraint C. S500: Generate task execution strategy A based on conflict resolution results, resource matching results, constraint condition C, real-time status X, and scheduling parameter θ.

[0037] Based on the already obtained set of surgical tasks, this embodiment of the invention integrates the task set, resource status, constraints, real-time status, and scheduling parameters to generate a task execution strategy that can be called by the surgical execution system. By constructing a scheduling function model, multiple tasks in the surgical process are uniformly organized and sorted, so that the task execution process is transformed from fixed process-driven to scheduling control-driven.

[0038] In step S100, the task set T, resource status R, constraint condition C, real-time status X, and scheduling parameter θ are obtained. The specific implementation process is as follows: Obtain the set of tasks to be scheduled T, the current available resource status R, the constraints C that need to be satisfied during scheduling, the real-time status X, and the scheduling parameters θ. It should be noted that step S100 belongs to the scheduling input acquisition stage. The task set T, resource status R, constraint C, real-time status X, and scheduling parameters θ can be obtained in parallel from different data interfaces, different system modules, or different data tables, or they can be obtained sequentially according to system implementation needs; there is no strict order in the acquisition of the above-mentioned input data. After step S100 is completed, subsequent steps S200 to S500 sequentially use the above-mentioned input data to complete priority calculation, constraint filtering, resource matching, conflict resolution, and strategy generation.

[0039] The task set T includes the task nodes to be executed, the prerequisite relationships between tasks, the current status of the tasks, and the task attributes. T can be derived from the results of structured task modeling or from surgical procedure task data that meets a preset format.

[0040] The current available resource status R includes the status of surgical equipment, execution tools, navigation equipment, doctor's terminal, computing resources, and other resource statuses related to task execution.

[0041] Constraint C includes safety constraints, accuracy constraints, resource constraints, execution order constraints, mutual exclusion constraints, and time constraints. Safety constraints limit the task execution scope, safety boundaries, neural tube avoidance, sinus floor avoidance, and other risk control conditions. Accuracy constraints limit path deviation, angle deviation, positioning error, and allowable operational error ranges. Resource constraints describe the occupancy and availability of equipment, instruments, computing resources, navigation equipment, and execution terminals. Order constraints describe the sequential execution relationships between tasks. Mutual exclusion constraints describe tasks that cannot be executed simultaneously or cannot share the same resources. Time constraints describe task execution windows, waiting times, or response time requirements. Through this set of constraints C, embodiments of the present invention can simultaneously consider factors such as safety, accuracy, resources, and execution order during scheduling, avoiding task arrangement based solely on a fixed order or a single objective.

[0042] Real-time status X includes current task execution progress, completed tasks, incomplete tasks, path deviation, equipment status changes, abnormal information, and changes in patient or device position.

[0043] The scheduling parameters θ include priority weights, constraint weights, conflict resolution parameters, and rescheduling trigger thresholds.

[0044] It should be noted that in this embodiment of the invention, the weight parameter θ={w1,w2,w3,w4,…} is used only as a scheduling input parameter. This embodiment of the invention protects the computational mechanism for constraint-driven scheduling using this parameter, and does not limit the specific source of the weight parameter. Furthermore, the method of updating θ based on execution data can be implemented separately by a parameter optimization method.

[0045] In step S200, task priorities are calculated based on the task set T, constraint C, real-time state X, and scheduling parameter θ; the task set T is then filtered according to constraint C to obtain an executable candidate task set T_e. The specific implementation process is as follows: Step S200 may include a priority calculation sub-step S210 and a constraint filtering sub-step S220. The priority calculation sub-step S210 is used to obtain a task priority sequence, and the constraint filtering sub-step S220 is used to determine whether a single task meets the conditions for entering the executable candidate set. Both are based on the task set T, constraint conditions C, real-time state X, and scheduling parameters θ obtained in step S100. In specific system implementations, they can be calculated in parallel, or the priority can be calculated first and then the constraint filtering can be performed. For ease of explanation, this embodiment describes the process in the order of calculating the priority first and then performing the constraint filtering.

[0046] The priority of each task is calculated based on task dependency, task urgency, degree of constraint impact, and degree of real-time status impact, forming a task priority sequence; for each task t in the task set T... i Calculate the priority value P i Priority values ​​can be calculated as follows: in, Indicates task dependency; Indicates the urgency of the task; Indicates the degree of influence of the constraint. Indicates the degree of impact of real-time status; , , and This represents the scheduling weight parameter.

[0047] Task dependency The task dependency can be determined by comprehensively considering the number of preceding and succeeding tasks in the task set T, whether it is on the critical task path, and the impact of its completion on subsequent tasks. For example, tasks with many preceding and succeeding tasks, or those on the critical path, can have a higher task dependency value. Task urgency The urgency level can be determined based on the current surgical stage, task waiting time, preset task execution window, whether the task affects subsequent critical steps, and any anomalies or deviations reflected in the real-time status X. For example, when the waiting time for a task exceeds a preset time threshold, or when the task directly affects subsequent critical steps, its urgency level can be set to a higher value. (Constraint Impact Degree) The degree of restriction imposed on the task by the safety constraints, accuracy constraints, and resource constraints in constraint condition C can be determined; the degree of influence of real-time status. It can be determined based on the current deviation, abnormal status, or degree of resource change in the real-time status X.

[0048] Based on the constraint C, the task set T is filtered out, and tasks that do not currently meet the safety constraint, precision constraint, basic resource constraint, or sequence constraint are eliminated, forming an executable candidate task set T_e.

[0049] Among them, the basic resource constraints are used to determine whether a task has the minimum resource conditions to enter the candidate set, and the specific resource allocation and resource occupation conflict handling are completed by the subsequent resource matching and conflict detection and resolution steps.

[0050] It should be noted that the constraint screening step here is mainly used to determine whether a single task meets the safety constraints, accuracy constraints, basic resource constraints, and execution order constraints, and to form a set of executable candidate tasks T_e accordingly; its purpose is different from the subsequent conflict detection and resolution steps, which are used to handle the mutual conflicts between multiple executable candidate tasks.

[0051] In step S300, available resources are allocated to each task in the executable candidate task set T_e according to the resource status R. The specific implementation process is as follows: Based on the resource status R, available resources are allocated to the set of executable candidate tasks T_e, determining the corresponding equipment, instruments, execution terminals, or computing resources for each task. The resource matching step, occurring after constraint screening and before conflict detection and resolution, is used to match candidate tasks that meet the single-task admission criteria with available resources, providing a basis for subsequent determination of whether resource contention exists among multiple tasks.

[0052] In step S400, after resource matching is completed, resource conflict, execution conflict, and sequence conflict detection are performed on multiple tasks in the executable candidate task set T_e. When a conflict is detected, conflict resolution is performed based on task priority, resource status R, and constraint conditions C. The specific implementation process is as follows: The conflict detection and resolution phase is executed after the executable candidate task set T_e is formed and resource matching is completed. It processes multiple candidate tasks that have met the single-task constraints and completed initial resource matching. This phase further determines whether there are resource competition, execution mutual exclusion, or sequence conflicts among the multiple tasks, and determines the scheduling result after conflict resolution through priority adjustment, delayed execution, or resource reallocation.

[0053] Therefore, in this embodiment of the invention, "constraint filtering" is used to determine whether a single task meets the conditions for entering the executable candidate set, "resource matching" is used to allocate available resources to candidate tasks according to resource status R, and "conflict detection and resolution" is used to resolve conflicts between multiple candidate tasks in terms of resource occupation, mutual exclusion execution, or execution order. The three functions are distinct from each other and are executed sequentially in time order.

[0054] During the scheduling process, conflicts between tasks are detected. These conflicts include, but are not limited to: Resource conflict: Multiple tasks simultaneously occupy the same equipment, instrument, navigation resources, or computing resources; Execution conflict: Multiple tasks are mutually exclusive and cannot be executed simultaneously; Sequence conflict: The order of task execution is inconsistent with the task relationships or prerequisites.

[0055] When a conflict is detected, the system resolves the conflict based on task priority, resource status, and constraints. Conflict resolution methods include: adjusting task execution order, delaying conflicting tasks, reallocating resources, pausing low-priority tasks, or triggering rescheduling.

[0056] Through the above mechanism, embodiments of the present invention enable the automatic handling of conflicts between multiple candidate tasks during surgical execution through scheduling strategies, rather than relying entirely on manual judgment.

[0057] In step S500, task execution strategy A is generated based on the conflict resolution results, resource matching results, constraint C, real-time status X, and scheduling parameters θ. The specific implementation process is as follows: The execution strategy A includes task execution order, resource allocation results, execution conditions, control requirements, and exception handling requirements. Specifically, the task execution order indicates the sequential execution relationship of each task; the resource allocation results indicate the equipment, instruments, execution terminals, or computing resources corresponding to each task; the execution conditions indicate the safety constraints, accuracy constraints, sequence constraints, or time constraints that need to be met before the task starts; the control requirements indicate the requirements for path calling, equipment control, navigation monitoring, or status acquisition during task execution; and the exception handling requirements indicate the delay, pause, reallocation, or rescheduling methods to be taken when path deviation, equipment malfunction, resource occupancy anomalies, or task failure occur. In one implementation, task execution strategy A can be expressed as: A=F(T,R,C,X;θ) Where F represents the constraint-driven scheduling function, T represents the task set, R represents the resource status, C represents the constraint condition, X represents the real-time status, and θ represents the scheduling parameters. It should be noted that T_e is the set of executable candidate tasks or intermediate scheduling results formed after constraint filtering, resource matching, and conflict resolution of the task set T. When generating execution strategy A, F can call the filtering results, resource matching results, and conflict resolution results corresponding to T_e.

[0058] It should be noted that if the real-time status changes during execution, an updated execution strategy A′ can be generated through the dynamic rescheduling step S600.

[0059] It should be noted that, in the specific implementation process, the constraint-driven scheduling method for surgical tasks also includes: S600, when the real-time state X changes and the triggering condition is met, the scheduling function F is recalculated to generate an updated task execution strategy A′.

[0060] like Figure 2 As shown, during the surgical procedure, when the system's real-time state X changes, the system determines whether the dynamic rescheduling conditions are met. Triggering conditions include, but are not limited to: execution deviation exceeding a preset threshold, resource status changes, new task addition, task cancellation, equipment status abnormality, navigation status abnormality, or critical task execution failure. When a change in real-time state X causes a task that was already in the executable task set T_e to no longer meet safety constraints, accuracy constraints, resource constraints, or sequence constraints, the system uses this situation as one of the dynamic rescheduling triggering conditions and re-executes the constraint filtering, resource matching, and conflict detection processes.

[0061] When the triggering condition is met, the system reacquires T, R, C, X, and θ, and recalculates the scheduling function F to generate an updated execution strategy A′. A′ is used to replace or modify the original execution strategy A, enabling the task execution process to adjust according to real-time state changes. When the scheduling parameters have been updated to θ′ by subsequent parameter optimization methods, subsequent scheduling functions can also generate an updated task execution strategy based on A′=F(T,R,C,X;θ′).

[0062] The task is executed according to scheduling strategy A, and the execution results are recorded during the process. The execution results include execution time, task completion status, path deviation, resource usage, exception information, and status change information. This data can be used for subsequent status updates, rescheduling trigger determination, or as input data for subsequent parameter optimization methods.

[0063] The effectiveness of the embodiments of the present invention is verified below using specific verification data. This verification data is only used to illustrate the technical effects of the present invention in task priority calculation, constraint filtering, resource matching, conflict resolution, and dynamic rescheduling, and is not intended as a conclusion for clinical efficacy evaluation.

[0064] Verification Experiment 1: In the oral implant surgery task scheduling scenario, 30 consecutive single-tooth implantation procedures were selected for testing to verify the effectiveness of the present invention in generating execution strategy A under fixed task set and fixed resource conditions.

[0065] (1) Experimental conditions: The same CBCT equipment and planning software were used; the same group of doctors performed the operation according to the same operating procedures; the comparison object was the fixed sequence execution method. The fixed sequence execution method means that each task is executed in sequence according to the preset steps, and the execution order is not dynamically adjusted according to task priority, resource status changes or conflicts.

[0066] (2) Method: The fixed-order execution method executes the steps sequentially according to preset steps; the method of this invention, according to steps S100 to S500, obtains the task set T, resource status R, constraint condition C, real-time status X, and scheduling parameter θ, performs priority calculation, constraint filtering, resource matching, and conflict resolution to generate execution strategy A. Evaluation indicators include average task execution time and number of process interruptions. Among them, the average task execution time refers to the average time required from the task set T entering the scheduling process to the completion of the corresponding task and the formation of an execution record; the number of process interruptions refers to the number of times the task process is paused due to waiting for resources, task conflicts, mismatched execution order, or manual confirmation.

[0067] (3) Results: The average task execution time decreased from 41.8 min to 34.6 min, a reduction of approximately 17%; the number of process interruptions decreased from 2.5 times / case to 1.4 times / case. Among them, 41.8 min and 34.6 min are the arithmetic mean of the task execution time corresponding to the fixed sequence execution method and the method of the present invention in 30 single-seed planting processes, respectively; 2.5 times / case and 1.4 times / case are the average number of process interruptions caused by waiting for resources, task conflicts, mismatched execution order, or manual confirmation in 30 processes, respectively.

[0068] (4) Conclusion: The above results show that, under the same task set and resource conditions, the present invention can reduce the waiting and interruption caused by fixed sequential execution by generating execution strategy A through constraint-driven scheduling, thereby improving task execution efficiency.

[0069] Verification Experiment 2: Resource conflicts and execution conflicts are simulated in multi-task execution scenarios to verify the effectiveness of the conflict detection and resolution mechanism of the present invention.

[0070] (1) Experimental setup: The experiment included scenarios such as multiple tasks competing for the same device resources, multiple tasks having mutually exclusive execution relationships, and some tasks experiencing resource status changes. The test data is only used to illustrate the technical effects of this invention in task conflict identification and conflict resolution.

[0071] (2) Method: The S400 step of the present invention is used to detect and resolve conflicts, including identifying resource conflicts, execution conflicts and sequence conflicts, and adjusting execution priority, delaying execution or reallocating resources according to task priority, resource status R and constraint conditions C, to form an execution strategy A after conflict resolution.

[0072] (3) Results: The conflict resolution success rate increased from approximately 65% ​​to 92%; the number of manual interventions decreased from an average of 2.0 times / group to 1.2 times / group, a reduction of approximately 40%. Among them, the conflict resolution success rate refers to the proportion of the system that completes task order adjustment or resource reallocation without violating safety constraints, accuracy constraints, and sequence constraints after a conflict is detected; the number of manual interventions refers to the number of times the operator performs manual confirmation or adjustment due to resource competition, mutual exclusion execution, or sequence conflict.

[0073] (4) Conclusion: The above results show that the present invention can incorporate task conflict identification and conflict resolution into a unified scheduling process, reducing the need for manual processing.

[0074] Verification Experiment 3: By setting execution disturbances, such as instrument offset of 0.5-1.0 mm or local execution state changes, the dynamic rescheduling capability of the present invention is verified.

[0075] (1) Experimental setup: During task execution, a state change is introduced to alter the real-time state X, and it is observed whether the system triggers a scheduling update. The test data is only used to illustrate the dynamic rescheduling capability of this invention after a real-time state change.

[0076] (2) Method: When the execution deviation exceeds the preset threshold, the system re-acquires T, R, C, X, and θ, and recalculates the scheduling function to generate an updated execution strategy: A′=F(T,R,C,X;θ). Evaluation metrics include scheduling response time and execution error recovery time. Scheduling response time refers to the time required for the system to generate execution strategy A′ after detecting a change in real-time state X; execution error recovery time refers to the time required from detecting the execution deviation to the system completing path or task order adjustment.

[0077] (3) Results: The scheduling response time was shortened from 3.2 s to 1.9 s; the execution error recovery time was shortened from 4.0 s to 2.8 s, a reduction of about 30%.

[0078] (4) Conclusion: The above results show that when the real-time state X changes, the present invention can improve the system’s response to execution disturbances by recalculating the scheduling function to generate the update strategy A′.

[0079] Verification Experiment 4: The test was conducted under conditions of simultaneous security constraints, accuracy constraints, resource constraints, and sequence constraints to verify the ability of the present invention to form an executable candidate task set T_e and an execution strategy A under multiple constraints.

[0080] (1) Experimental setup: Set restrictions on task execution range, path error, device resource usage, and task pre-order. The test data is only used to illustrate the technical effects of this invention in task selection, resource matching, and strategy generation under multiple constraints.

[0081] (2) Method: The S200 constraint screening step of the present invention is used to eliminate or delay tasks that do not meet the safety constraints, accuracy constraints, basic resource constraints or sequence constraints, forming a set of executable candidate tasks T_e; then, through S300 resource matching and S400 conflict detection and resolution, resource allocation and conflict handling are performed on the candidate tasks that meet the conditions, and further execution strategy A is formed.

[0082] (3) Results: The task constraint satisfaction rate in the execution strategy increased from 82% to 96%; task execution conflicts decreased by about 35%. Among them, the constraint satisfaction rate refers to the proportion of each task in the generated execution strategy A that simultaneously satisfies the safety constraints, precision constraints, resource constraints and sequence constraints; the reduction in task execution conflicts refers to the percentage decrease in the number of resource conflicts, execution mutual exclusion and sequence conflicts compared with the fixed sequence execution method.

[0083] (4) Conclusion: The above results show that the present invention can generate a scheduling strategy with high executability when multiple constraints exist at the same time.

[0084] This invention also provides a constraint-driven scheduling system for surgical tasks, comprising: The scheduling input acquisition module is used to acquire the task set T, resource status R, constraint conditions C, real-time status X, and scheduling parameters θ; The priority calculation and constraint filtering module is used to calculate task priorities based on task set T, constraint C, real-time status X, and scheduling parameter θ; and to filter task set T according to constraint C to obtain executable candidate task set T_e. The resource matching module is used to allocate available resources to each task in the executable candidate task set T_e according to the resource status R; The conflict detection and resolution module is used to detect resource conflicts, execution conflicts, and sequence conflicts for multiple tasks in the executable candidate task set T_e after resource matching is completed, and to resolve conflicts based on task priority, resource status R, and constraint conditions C when a conflict is detected. The strategy generation module is used to generate task execution strategy A based on conflict resolution results, resource matching results, constraint C, real-time status X, and scheduling parameters θ.

[0085] It should be noted that, in the specific implementation process, the constraint-driven scheduling system also includes a dynamic rescheduling module, which is used to recalculate the scheduling function F and generate an updated task execution strategy A′ when the real-time state X changes and the triggering condition is met.

[0086] It is understood that the constraint-driven scheduling system for surgical tasks provided in this embodiment of the invention corresponds to the constraint-driven scheduling method for surgical tasks described above. The explanations, examples, and beneficial effects of the relevant content can be found in the corresponding content of the constraint-driven scheduling method for surgical tasks, and will not be repeated here.

[0087] This invention also provides a computer-readable storage medium storing a computer program for constraint-driven scheduling for surgical tasks, wherein the computer program causes a computer to execute the constraint-driven scheduling method for surgical tasks as described above.

[0088] This invention also provides an electronic device, including: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including a constraint-driven scheduling method for performing surgical tasks as described above.

[0089] In summary, compared with existing technologies, it has the following beneficial effects: 1. Based on the already obtained set of surgical tasks, this embodiment of the invention integrates the task set, resource status, constraints, real-time status and scheduling parameters to generate a task execution strategy that can be called by the surgical execution system. By constructing a scheduling function model, multiple tasks in the surgical process are uniformly organized and sorted, so that the task execution process is transformed from fixed process-driven to scheduling control-driven.

[0090] 2. The embodiments of the present invention coordinate the execution sequence of parallel and related tasks during surgery by combining task constraint screening, dynamic priority calculation and refined resource matching. This abandons the inherent pattern of fixed-sequence serial execution in traditional surgery and can reasonably stagger the waiting periods of each task, reducing the risk of process pauses and interruptions caused by task waiting, resource occupation or mismatch in execution order, thereby improving the efficiency of intraoperative task scheduling.

[0091] 3. In the scheduling decision-making process, this invention simultaneously integrates four core clinical constraints: safety constraints, accuracy constraints, resource constraints, and task sequence constraints. It constructs a multi-constraint collaborative scheduling model, uniformly evaluating task execution priorities and executability under different constraints, and resolving conflicts when preset conditions are met. Compared to traditional scheduling methods that only consider a single constraint, this invention can adapt to complex and ever-changing clinical surgical conditions, outputting scheduling strategies that take into account multi-dimensional clinical requirements, effectively improving the clinical executability and overall rationality of the scheduling plan.

[0092] 4. This invention addresses multi-task parallel scenarios during surgery by establishing a comprehensive conflict detection mechanism. It can automatically identify resource contention, execution mutual exclusion, and timing conflicts during task execution. Simultaneously, it employs conflict resolution strategies such as priority adjustment, delayed execution, and resource reallocation to form a scheduling process from conflict identification to the generation of handling suggestions. Under preset constraints, it reduces reliance on manual intervention and assists in handling various scheduling conflicts.

[0093] 5. The execution strategy A generated in this embodiment of the invention can be directly applied to the task execution order, resource allocation and execution conditions, reducing the conversion links between scheduling results and actual execution, and improving the continuity of system operation.

[0094] 6. When there are changes in tasks, resources, or execution disturbances, the embodiments of the present invention can generate an updated execution strategy A′ through a dynamic rescheduling mechanism, which has better adaptability than fixed path or offline scheduling methods.

[0095] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0096] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A constraint-driven scheduling method for a surgical task, characterized in that, include: Obtain the task set, resource status, constraints, real-time status, and scheduling parameters; The task priority is calculated based on the task set, constraints, real-time status, and scheduling parameters; the task set is then filtered according to the constraints to obtain a set of executable candidate tasks. Allocate available resources to each task in the set of executable candidate tasks based on resource status; After resource matching is completed, resource conflict, execution conflict and sequence conflict detection are performed on multiple tasks in the set of executable candidate tasks. When a conflict is detected, the conflict is resolved according to task priority, resource status and constraints. A task execution strategy is generated based on the conflict resolution results, resource matching results, constraints, real-time status, and scheduling parameters.

2. The surgical task oriented constraint driven scheduling method of claim 1, wherein, The calculation of task priority based on task set, constraints, real-time status, and scheduling parameters includes: For each task in the task set Calculate priority value The formula for calculating the priority value is as follows: in, Indicates task dependency; Indicates the urgency of the task; Indicates the degree of influence of the constraint. Indicates the degree of impact of real-time status; , , and Indicates the scheduling weight parameter; Task dependency The determination is based on a comprehensive assessment of the number of preceding tasks, the number of succeeding tasks, whether the task is on a critical task path, and the impact of the task's completion on subsequent tasks within the task set. Mission urgency It can be determined based on the current surgical stage, task waiting time, task preset execution window, whether the task affects subsequent key steps, and any abnormalities or deviations reflected in the real-time status. Degree of influence of constraints The degree of restriction imposed on the task by the safety constraints, accuracy constraints, and resource constraints in the constraints is determined; Real-time status impact Determined based on the current deviation, abnormal state, or degree of resource change in real-time status X.

3. The constraint-driven scheduling method for surgical tasks as described in claim 1, characterized in that, The step of allocating available resources to each task in the set of executable candidate tasks based on resource status includes: Based on the resource status, allocate available resources to the set of executable candidate tasks and determine the corresponding equipment, instruments, execution terminals or computing resources for each task.

4. The constraint-driven scheduling method for surgical tasks as described in claim 1, characterized in that, The conflict resolution methods include: adjusting the task execution order, delaying conflicting tasks, reallocating resources, pausing low-priority tasks, or triggering rescheduling.

5. The constraint-driven scheduling method for surgical tasks as described in claim 1, characterized in that, The execution strategy includes task execution order, resource allocation results, execution conditions, control requirements, and exception handling requirements.

6. The constraint-driven scheduling method for surgical tasks as described in any one of claims 1 to 5, characterized in that, The constraint-driven scheduling method further includes: When the real-time state changes and the triggering condition is met, the scheduling function F is recalculated to generate an updated task execution strategy A′; its expression is as follows: A′=F(T,R,C,X;θ) Where F represents the constraint-driven scheduling function; T represents the current task set; R represents the current resource state; C represents the constraint condition; X represents the real-time state; and θ represents the scheduling parameters.

7. The constraint-driven scheduling method for surgical tasks as described in claim 6, characterized in that, The triggering conditions include: execution deviation exceeding a preset threshold, resource status change, new task addition, task cancellation, abnormal device status, abnormal navigation status, or failure of critical task execution.

8. A constraint-driven scheduling system for surgical tasks, characterized in that, include: The scheduling input acquisition module is used to acquire task sets, resource status, constraints, real-time status, and scheduling parameters; The priority calculation and constraint filtering module is used to calculate task priorities based on task sets, constraints, real-time status, and scheduling parameters; and to filter the task set according to constraints to obtain a set of executable candidate tasks. The resource matching module is used to allocate available resources to each task in the set of executable candidate tasks based on the resource status; The conflict detection and resolution module is used to detect resource conflicts, execution conflicts, and sequence conflicts among multiple tasks in the set of executable candidate tasks after resource matching is completed, and to resolve conflicts based on task priority, resource status, and constraints when a conflict is detected. The strategy generation module is used to generate task execution strategies based on conflict resolution results, resource matching results, constraints, real-time status, and scheduling parameters.

9. A computer-readable storage medium, characterized in that, It stores a computer program for constraint-driven scheduling for surgical tasks, wherein the computer program causes the computer to execute the constraint-driven scheduling method for surgical tasks as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including a constraint-driven scheduling method for performing surgical tasks as described in any one of claims 1 to 7.

Images