Auxiliary structure generation method and device for helical tomotherapy system

Abstract

The application discloses an auxiliary structure generation method and device for a helical tomotherapy system, relates to the technical field of medical information, and in particular to the field of TOMO radiotherapy information processing, and the method comprises the following steps: generating an initial auxiliary structure according to the geometric characteristics of a target region and the spatial position relationship between the target region and a critical organ; associating and configuring the auxiliary structure with a reverse optimization target, and the association and configuration are used for realizing a linkage relationship between the parameter information of the auxiliary structure and the weight of an optimization target function; in a reverse optimization process, an optimization state index is collected, and it is determined whether a plan optimization stage is switched according to the optimization state index; and in the case that the plan optimization stage is detected to be switched, a reconstruction operation of the auxiliary structure is triggered. The application solves the technical problem that in the prior art, the auxiliary structure is stored as a static data object in a Tomo radiotherapy plan, cannot be dynamically adjusted according to an optimization stage, and leads to low calculation efficiency of a radiotherapy plan data processing system.

Description

Technical Field

[0001] This application relates to the field of medical information technology, especially to the field of TOMO radiotherapy information processing, and specifically to a method and apparatus for generating auxiliary structures for a helical tomotherapy system. Background Technology

[0002] In the inverse planning process of helical tomotherapy, in order to improve the dose conformity of the target area, control the radiation dose to organs at risk (OAR), and suppress the unintended spread of low-dose areas, it is usually necessary to construct a variety of auxiliary structures such as ring structures, avoidance structures, and shell structures around the target area, and introduce these auxiliary structures as optimization objectives or constraints into the inverse optimization process.

[0003] In related technologies, radiotherapy planning systems generally support the automatic generation of ring-shaped or shell-shaped structures based on geometric expansion rules, and some automated planning systems also reuse fixed auxiliary structure configurations through templates. However, in existing radiotherapy planning systems, auxiliary structures are stored in the planning file as static data objects, and their geometric parameters remain unchanged throughout the entire optimization data processing flow. This static data structure leads to a large number of invalid iterative calculations when the optimization algorithm processes multi-stage optimization tasks, increasing the processor's computational load and memory consumption, and affecting the data throughput efficiency of the radiotherapy planning system.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a method and apparatus for generating auxiliary structures in a helical tomotherapy system, which at least solves the technical problem in the prior art where auxiliary structures are stored as static data objects in the Tomo radiotherapy plan and cannot be dynamically adjusted according to the optimization stage, resulting in low computational efficiency of the radiotherapy plan data processing system.

[0006] According to one aspect of the embodiments of this application, a method for generating auxiliary structures for a helical tomotherapy system is provided, comprising: generating an initial auxiliary structure based on the geometric features of the target area and the spatial positional relationship between the target area and organs at risk, wherein the auxiliary structure is a spatial region used to constrain dose distribution during radiotherapy planning optimization; associating the auxiliary structure with an inverse optimization objective, wherein the association configuration is used to establish a linkage between the parameter information of the auxiliary structure and the weights of the optimization objective function, the optimization objective function being used to quantify the degree to which the radiotherapy plan meets clinical requirements; during the inverse optimization process, collecting optimization status indicators and determining whether a switch in the planning optimization stage has occurred based on the optimization status indicators; and triggering a reconstruction operation of the auxiliary structure if a switch in the planning optimization stage is detected.

[0007] Optionally, the auxiliary structure includes at least one of the following: a gradient control structure for constraining the rate of dose decrease outside the target area; a multi-layer dose control structure for hierarchically controlling the dose distribution at different spatial distances; and a rejection region structure for controlling the extent to which the inhibitory amount spills out of the target area between the target area and the organs at risk.

[0008] Optionally, an initial auxiliary structure is generated based on the geometric features of the target area and the spatial relationship between the target area and the organ at risk. This includes: determining the volume of the target area based on its geometric features; determining the number of layers of the auxiliary structure based on the volume of the target area; determining the minimum distance between the target area and the organ at risk based on the spatial relationship between the target area and the organ at risk; determining the type combination of the auxiliary structure based on the minimum distance; determining the thickness parameter of the auxiliary structure based on the current planning optimization stage, wherein the thickness parameter is used to constrain the distance range of the auxiliary structure extending outward from the outer surface of the target area; and generating the initial auxiliary structure based on the number of layers of the auxiliary structure, the type combination of the auxiliary structure, and the thickness parameter of the auxiliary structure.

[0009] Optionally, determining the number of auxiliary structure layers based on the target region's volume includes: when the target region's volume is detected to be less than a first volume threshold, determining the number of auxiliary structure layers to be a first range; when the target region's volume is detected to be greater than or equal to the first volume threshold and less than a second volume threshold, determining the number of auxiliary structure layers to be a second range; when the target region's volume is detected to be greater than or equal to the second volume threshold, determining the number of auxiliary structure layers to be a third range; wherein the maximum number of layers in the first range is less than the minimum number of layers in the second range, and the maximum number of layers in the second range is less than the minimum number of layers in the third range.

[0010] Optionally, the combination of auxiliary structures is determined based on the minimum distance between the target area and the organ at risk, including: generating at least a rejection region structure as an auxiliary structure when the minimum distance is detected to be less than a first distance threshold; generating at least a gradient control structure and a multi-layer dose control structure to form an auxiliary structure when the minimum distance is detected to be greater than or equal to the first distance threshold and less than a second distance threshold; and generating a gradient control structure to form an auxiliary structure when the minimum distance is detected to be greater than or equal to the second distance threshold.

[0011] Optionally, the auxiliary structure is configured in association with the inverse optimization objective, including: when the auxiliary structure is detected to include at least one of a gradient control structure and a multi-layer dose control structure, assigning a dose gradient constraint objective to each structure, wherein the dose gradient constraint objective is used to limit the rate at which the dose outside the target area decreases with increasing distance; when the auxiliary structure is detected to include a repulsion region structure, assigning a maximum dose constraint objective to the repulsion region structure, wherein the maximum dose constraint objective is used to limit the highest dose value within the repulsion region structure; and configuring at least one of the dose gradient constraint objective and the maximum dose constraint objective in association with the inverse optimization objective, so that the constraint strength of the auxiliary structure is adjusted as the planned optimization stage changes.

[0012] Optionally, the inverse optimization objective represents the dose index that needs to be optimized during the inverse optimization of the radiotherapy plan, including at least one of target dose coverage, organ at risk dose limitation threshold, and extra-target dose gradient threshold; the weights of the optimization objective function are relative importance coefficients assigned to each optimization objective in multi-objective inverse optimization, used to balance the priorities among different optimization objectives.

[0013] Optionally, during the reverse optimization process, optimization status indicators are collected, and the switching of the planned optimization stage is determined based on the optimization status indicators. This includes: during the reverse optimization process, monitoring at least the following optimization status indicators: the change in target area coverage, the change in volume of organs at risk exceeding a preset dose threshold, the rate of decrease in the function value of the optimization objective function as the number of iterations increases, and the change in the uniformity of dose distribution; comparing the monitored optimization status indicators with preset stage switching conditions; if the optimization status indicators meet the stage switching conditions, it is determined that the planned optimization stage has changed; if the optimization status indicators do not meet the stage switching conditions, it is determined that the planned optimization stage has not changed.

[0014] Optionally, the optimization phase includes an initial coverage phase, a transitional optimization phase, and a convergence phase. The criteria for the initial coverage phase include: the change in target coverage is less than a first preset threshold, and the rate of decrease in the function value reduction rate is greater than a first rate threshold. The criteria for the transitional optimization phase include: the change in target coverage is greater than or equal to a first preset threshold, and the change in the volume of organs at risk exceeding a preset dose threshold is greater than a second preset threshold. The criteria for the convergence phase include: the rate of decrease in the function value is less than a second rate threshold, and the change in the uniformity index of the dose distribution is less than a third preset threshold, wherein the second rate threshold is less than the first rate threshold.

[0015] Optionally, the reconstruction operation includes: in the initial coverage phase, reducing the weights corresponding to the dose constraint targets of the gradient control structure when a gradient control structure exists; reducing the weights corresponding to the dose constraint targets of the multi-layer dose control structure and reducing the number of layers of the multi-layer dose control structure when a multi-layer dose control structure exists; reducing the weights corresponding to the dose constraint targets of the exclusion region structure when an exclusion region structure exists; in the transition optimization phase, increasing the number of layers of the multi-layer dose control structure when a multi-layer dose control structure exists; introducing an exclusion region structure when no exclusion region structure exists; increasing the gradient constraint strength of the gradient control structure when a gradient control structure exists; and in the convergence phase, adding outer gradient control structures, reducing the thickness of the multi-layer dose control structure when a multi-layer dose control structure exists, and increasing the weights of dose constraint targets related to organs at risk.

[0016] Optionally, the reconstruction operation of the auxiliary structure includes at least one of the following: adding or reducing gradient control structures; adding or reducing the number of layers of multi-layer dose control structures; adjusting the expansion distance, thickness parameter, or spatial position of the auxiliary structure, wherein the expansion distance is the distance the auxiliary structure extends outward from the outer surface of the target area; replacing some gradient control structures with multi-layer dose control structures, or adding a repulsion region structure; and reallocating the dose constraint targets corresponding to each type of structure in the auxiliary structure and the weight parameters of each dose constraint target.

[0017] According to another aspect of the embodiments of this application, an auxiliary structure generation device for a helical tomotherapy system is also provided, comprising: a generation unit, configured to generate an initial auxiliary structure based on the geometric features of the target area and the spatial positional relationship between the target area and organs at risk, wherein the auxiliary structure is a spatial region used to constrain dose distribution during radiotherapy planning optimization; an association configuration unit, configured to associate the auxiliary structure with an inverse optimization objective, wherein the association configuration is used to establish a linkage between the parameter information of the auxiliary structure and the weights of the optimization objective function, the optimization objective function being used to quantify the degree to which the radiotherapy plan meets clinical requirements; a first processing unit, configured to collect optimization status indicators during the inverse optimization process and determine whether a switch has occurred in the planning optimization stage based on the optimization status indicators; and a second processing unit, configured to trigger a reconstruction operation of the auxiliary structure when a switch in the planning optimization stage is detected.

[0018] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, which stores a computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located performs the above-described auxiliary structure generation method for a helical tomotherapy system.

[0019] According to another aspect of the embodiments of this application, an electronic device is also provided, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to perform the above-described auxiliary structure generation method for a helical tomotherapy system.

[0020] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program or instructions that, when executed by a processor, implement the above-described auxiliary structure generation method for a helical tomotherapy system.

[0021] In this embodiment, the radiotherapy planning system first generates an initial auxiliary structure based on the geometric features of the target area and the spatial relationship between the target area and organs at risk. This auxiliary structure is a spatial region used to constrain dose distribution during radiotherapy planning optimization, thus obtaining an auxiliary structure configuration that matches the individual anatomical features of the patient before optimization begins. Furthermore, the radiotherapy planning system associates the auxiliary structure with the inverse optimization objective. This association establishes a linkage between the parameter information of the auxiliary structure and the weights of the optimization objective function, allowing the constraint strength of the auxiliary structure to adjust collaboratively with changes in the optimization objective, rather than existing as a static constraint independently. Based on this, the radiotherapy planning system collects optimization status indicators during the inverse optimization process and determines whether a switch in the planning optimization stage has occurred based on these indicators. Upon detecting a switch in the planning optimization stage, it triggers a reconstruction operation of the auxiliary structure. In this way, the radiotherapy planning system can perceive the stage changes of the optimization process in real time and actively adjust the auxiliary structure itself when a stage switches. By linking the parameter information of the auxiliary structure with the weights of the optimization objective function, the radiotherapy planning system transforms the auxiliary structure from a static geometric region detached from the optimization process into a dynamic constraint tool that adjusts collaboratively with the optimization objective function. Furthermore, the radiotherapy planning system collects optimization status indicators and determines whether a planning optimization phase has switched. When a phase switch occurs, it proactively triggers the reconstruction of auxiliary structures, enabling the auxiliary structures to adaptively adjust their configuration parameters according to the optimization phase. This dynamic adjustment mechanism helps prevent the auxiliary structures from becoming invalid constraints or interference terms in the later stages of optimization, ensuring that the constraints of the auxiliary structures in each optimization phase match the current optimization focus. This, in turn, improves the convergence speed and stability of inverse optimization.

[0022] Furthermore, in this embodiment, the radiotherapy planning system establishes a linkage between the parameter information of the auxiliary structure and the weights of the optimization objective function. This transforms the auxiliary structure from a static data object detached from the optimization process into a dynamic data structure that adjusts in tandem with the optimization objective function. This dynamic adjustment mechanism helps prevent the auxiliary data structure from becoming an invalid constraint or interference term in the later stages of optimization, reducing the number of invalid iterations in the optimization algorithm and lowering the processor load. This improves the data throughput and response speed of the radiotherapy planning system, thereby solving the technical problem in the prior art where the auxiliary structure is stored as a static data object in Tomo radiotherapy planning, unable to be dynamically adjusted according to the optimization stage, resulting in low computational efficiency of the radiotherapy planning data processing system. Attached Figure Description

[0023] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0024] Figure 1 This is a schematic diagram of an optional auxiliary structure generation method for a helical tomotherapy system according to an embodiment of this application;

[0025] Figure 2 This is a flowchart of an optional auxiliary structure generation method for a helical tomotherapy system according to an embodiment of this application;

[0026] Figure 3 This is a schematic diagram of an optional auxiliary structure generation device for a helical tomotherapy system according to an embodiment of this application. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] According to an embodiment of this application, a method embodiment for generating auxiliary structures for a helical tomotherapy system is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] It should be noted that the information collected in this application (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of this data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding access points are provided for users to choose to authorize or refuse. For example, interfaces are set up between this system and relevant users or organizations, providing users with corresponding access points to choose to agree to or refuse automated decision-making results; if the user chooses to refuse, the process proceeds to the expert decision-making stage.

[0031] According to the embodiments of this application, a radiotherapy planning system can be used as the execution subject of the auxiliary structure generation method for a helical tomotherapy system in the embodiments of this application. The system can be a software system or an embedded system combining software and hardware. Of course, the execution subject of the method in the embodiments of this application can also be other forms of execution subject, such as devices, equipment, etc. It should be known by those skilled in the art that this application does not particularly limit the specific form of the execution subject.

[0032] Figure 1 This is a schematic diagram of an auxiliary structure generation method for a helical tomotherapy system according to an embodiment of this application, as shown below. Figure 1 As shown, the method includes the following steps:

[0033] Step S101: Based on the geometric features of the target area and the spatial relationship between the target area and the organs at risk, an initial auxiliary structure is generated, wherein the auxiliary structure is a spatial region used to constrain the dose distribution during the radiotherapy planning optimization process.

[0034] Optionally, the geometric features of the target region can represent its volume and morphological complexity. For example, a radiotherapy planning system can extract the three-dimensional contour information of the target region from acquired patient structure set data and calculate the total volume occupied by the target region and the degree of irregularity in its morphology. The radiotherapy planning system can extract the three-dimensional contour data of the target region from the patient structure set data, which is stored in DICOM RT Structure Set format. The system converts this data into a computable data object through a data parsing module.

[0035] Alternatively, organs at risk can refer to healthy tissues or organs located near the target area that are highly sensitive to radiation and require dose limits during radiotherapy. For example, in radiotherapy for lung tumors, the spinal cord, heart, and esophagus can all be considered organs at risk.

[0036] Optionally, the spatial relationship between the target area and the organ at risk includes at least the minimum distance between them, which may refer to the shortest spatial distance from the outer surface of the target area to the outer surface of a certain organ at risk. Additionally, the spatial relationship between the target area and the organ at risk may also include the contact area between them, the overlapping volume between them, or their relative spatial orientation.

[0037] Optionally, the initial auxiliary structure can refer to a set of spatial regions that the system can automatically create before inverse optimization begins. These regions are not actual anatomical tissue of the patient, but rather artificially constructed geometries to guide dose distribution. The auxiliary structure serves as the spatial region used to constrain dose distribution during radiotherapy planning optimization. In subsequent inverse optimization, the system can assign specific dose constraint targets to each spatial region, such as maximum dose constraints or dose gradient constraints, thereby guiding the optimization algorithm to control dose output within that spatial region in the desired manner. The auxiliary structure can be represented as a ring-shaped structure, a shell-like structure, or an avoidance structure, among other spatial regions.

[0038] For example, a radiotherapy planning system can automatically determine the number of layers, type combinations, and thickness parameters of initial auxiliary structures based on the target volume and the minimum distance between the target and the organ at risk. For instance, for a large target volume (greater than 150 cubic centimeters), the system can generate more auxiliary structures, such as 3 to 5 layers. For a smaller target volume (less than 50 cubic centimeters), the system can generate fewer auxiliary structures, such as 1 to 2 layers. Furthermore, when the minimum distance between the target and an organ at risk is very small (less than 5 millimeters), the system can generate auxiliary structures with strong repulsive properties in the direction between the target and the organ at risk; when the minimum distance is large (greater than 20 millimeters), the system can generate only auxiliary structures for controlling the dose gradient.

[0039] Compared with related technologies that rely on repeated adjustments based on human experience or use fixed rules for generation, associating the generation of auxiliary structures with objective anatomical data of the target area and organs at risk facilitates the automation and personalization of auxiliary structure configuration. This can reduce the workload of manual modeling, improve the rationality of auxiliary structure configuration, and match the initial auxiliary structure with the patient's specific anatomical characteristics.

[0040] For example, the radiotherapy planning system can acquire the patient's planning target volume (PTV) and adjacent spinal cord and bilateral parotid glands as organs of risk. The system calculates the target volume to be 120 cubic centimeters, which is considered medium-sized, thus determining that two shell-like structures need to be generated as auxiliary structures. The system further calculates the minimum distance between the target volume and the spinal cord to be 15 millimeters, also considered medium-sized, thus determining the auxiliary structure type combination to be a ring structure plus a shell structure, excluding avoidance structures. Based on the determined number of layers and type combination, the system can automatically generate the initial ring structure and shell structure around the target volume through morphological dilatation operations.

[0041] For example, the radiotherapy planning system can acquire a small intracranial target area with a volume of 35 cubic centimeters, as well as the optic chiasm and brainstem as organs of risk. The system calculates that the minimum distance between the target area and the optic chiasm is only 3 millimeters, which is considered very close. Based on the volume of the small intracranial target area and the distance between the target area and the optic chiasm, the system can automatically determine that the number of auxiliary structures is 1, and the type combination is a combination including a high-weight avoidance structure. The system then automatically generates an avoidance structure in the adjacent region between the target area and the optic chiasm. This avoidance structure can be configured to guide the optimizer to reduce the dose output in the adjacent region between the target area and the optic chiasm during the optimization process to protect the optic chiasm from over-irradiation. The system can also generate single-layer ring structures in other directions of the target area to control the baseline dose gradient.

[0042] Step S102 involves associating the auxiliary structure with the inverse optimization objective. This association configuration is used to establish a linkage between the parameter information of the auxiliary structure and the weights of the optimization objective function. The optimization objective function is used to quantify the degree to which the radiotherapy plan meets clinical requirements.

[0043] Optionally, the inverse optimization objective can refer to a series of dosimetric indicators that need to be approximated or achieved through optimization algorithms during the inverse optimization process of radiotherapy planning, such as dosimetric indicators like target dose coverage, organ at risk dose limitation threshold, or extra-target dose gradient threshold.

[0044] Optionally, the associated configuration can represent binding a specific auxiliary structure to the dose constraint function that the auxiliary structure is supposed to perform. For example, a radiotherapy planning system can assign dose gradient constraint targets to ring structures and shell structures to limit the rate at which the dose outside the target area decreases with increasing distance; and assign a maximum dose constraint target to avoidance structures to limit the highest dose value within the avoidance structure.

[0045] Optionally, the parameter information of the auxiliary structure may include its geometric parameters, such as the number of layers, thickness, expansion distance, and spatial location. The weights of the objective function can refer to the relative importance coefficients assigned to each objective in the inverse optimization of multiple objectives, used to balance the priorities among different objectives. For example, when the constraint weight of a certain auxiliary structure is high, the optimizer can allocate more computational power to satisfy that constraint.

[0046] Optionally, the linkage relationship indicates that the parameter information of the auxiliary structure and the weights of the optimization objective function are not independent static relationships, but can change collaboratively with the optimization process. For example, when the radiotherapy planning system reconstructs the auxiliary structure according to changes in the optimization stage, such as increasing the number of shell structure layers, adjusting the thickness, or introducing avoidance structures, the system can simultaneously reallocate the weights of the optimization objective function associated with the auxiliary structure, so that the constraint strength of the auxiliary structure matches the focus of the current optimization stage. For example, in the early stage of optimization, the system can set the constraint weights related to the auxiliary structure to a lower level to avoid affecting target coverage; in the later stage of optimization, the system can increase these constraint weights related to the auxiliary structure to finely control the dose gradient.

[0047] By establishing a correlation configuration between the auxiliary structure and the inverse optimization objective, the auxiliary structure is no longer a static geometric constraint, but a dynamic constraint whose influence can be automatically adjusted as the weight of the optimization objective changes. This helps to reduce the interference of invalid constraints on the inverse optimization process, improve the optimization convergence efficiency, and ensure that the configuration of the auxiliary structure is consistent with the current optimization objective, thereby enhancing the stability and effectiveness of the inverse optimization process.

[0048] For example, a radiotherapy planning system can associate dose gradient constraint objectives with each of the automatically generated three-shell structures. For instance, the three-shell structure may include an innermost shell structure, a middle shell structure, and an outermost shell structure arranged sequentially from the inside out. The system can associate a constraint objective with the innermost shell structure: the intra-shell dose does not exceed 90% of the prescribed dose; with the middle shell structure: the intra-shell dose does not exceed 70% of the prescribed dose; and with the outermost shell structure: the intra-shell dose does not exceed 50% of the prescribed dose. Furthermore, the system can establish a linkage between these dose gradient constraint objectives and the gradient sub-item weights in the optimization objective function: in the initial coverage phase, these gradient sub-item weights can be set to 0.1; in the transition optimization phase, the gradient sub-item weights can be adjusted to 0.5; and in the convergence phase, the gradient sub-item weights can be further increased to 1.0. When switching optimization phases, the system can automatically update these weight values, so that the constraint strength of the shell structure dynamically increases with each optimization phase.

[0049] For example, an avoidance structure can be created between the target area and the spinal cord. The radiotherapy planning system can associate a maximum dose constraint target with this avoidance structure, i.e., the maximum dose within the avoidance structure does not exceed 45 Gy. The system can configure this constraint target in conjunction with the weight of the OAR dose constraint term in the optimization objective function. In the early stages of optimization, this weight can be set to 0.2 to avoid excessively restricting the dose entering the target area; when the optimization enters the transition phase and the target coverage has reached more than 90%, the system automatically increases the weight to 0.8, enabling the optimizer to strongly suppress the dose within the avoidance structure, thereby protecting the spinal cord from excessive irradiation.

[0050] Step S103: During the reverse optimization process, collect optimization status indicators and determine whether the planned optimization phase has switched based on the optimization status indicators.

[0051] Optionally, the inverse optimization process can refer to the process by which a radiotherapy planning system continuously adjusts treatment parameters such as beam intensity and subfield weights through iterative algorithms to gradually approach the optimal solution of the objective function. Optimization status indicators can refer to monitorable numerical parameters used to quantify the progress of inverse optimization and the degree of change in plan quality. For example, optimization status indicators may include changes in target coverage, changes in the volume of organs at risk exceeding a preset dose threshold, the rate of decrease in the objective function value as the number of iterations increases, and changes in the uniformity of dose distribution. Specifically, changes in target coverage can represent the magnitude of change in the percentage of target dose coverage between two monitoring periods; the rate of decrease in the objective function value can represent the magnitude of the decrease in the objective function value per unit number of iterations; and the uniformity of dose distribution is used to quantify the flatness of the dose distribution within the target area.

[0052] Optionally, the planning optimization phase can refer to the time interval with different optimization focuses and convergence characteristics during the reverse optimization process. Related technologies lack phase awareness in their optimization processes, making it impossible to determine whether the current stage is the early coverage phase, the intermediate transition phase, or the late convergence phase. This application's embodiments determine whether the planning optimization phase has changed by comparing collected optimization status indicators with preset phase switching conditions. For example, when the function value reduction rate is consistently below a certain threshold, it can be determined that the optimization has entered the convergence phase; when the target coverage change is less than a certain threshold and the function value reduction rate is relatively fast, it can be determined that it is in the initial coverage phase. By sensing its own optimization progress, the radiotherapy planning system can objectively and quantitatively determine the current optimization stage, which is beneficial for subsequent dynamic reconstruction of auxiliary structures and helps trigger reconstruction actions at the correct time, avoiding premature or late adjustments due to misjudgment, thereby improving the reliability of the entire adaptive optimization process.

[0053] For example, during the reverse optimization process, the radiotherapy planning system can calculate the rate of reduction of the objective function value after each iteration cycle. The system can preset a threshold of 1%, meaning that when the rate of reduction of the objective function value is below 1% for three consecutive iteration cycles, the optimization phase is considered to have switched to the convergence phase. In the 25th iteration, the system detects that the rate of reduction of the objective function value has dropped to 0.8%, and in the subsequent two cycles it is 0.7% and 0.6%, respectively, both below the 1% threshold. Based on this, the system determines that the planning optimization phase has switched and prepares to trigger the subsequent auxiliary structure reconstruction operation.

[0054] For example, the radiotherapy planning system can simultaneously monitor changes in target coverage and changes in overdose volume in organs at risk. The system's preset joint triggering condition can be: when the change in target coverage is less than 0.5% and the change in overdose volume in organs at risk is less than 0.1%, the optimization phase is determined to switch from the transition optimization phase to the convergence refinement phase. During the optimization process, the system calculates that the average change in target coverage over the most recent five iterations is 0.3%, and the average change in overdose volume in organs at risk is 0.05%, both indicators meeting the preset conditions. Based on this, the system can determine that the optimization phase has changed, triggering subsequent auxiliary structure reconstruction operations, such as adding an outer ring structure to further suppress low-dose spillover.

[0055] Step S104: If a switch to the planning optimization phase is detected, a reconstruction operation of the auxiliary structure is triggered.

[0056] Optionally, detecting a change in the optimization phase can mean that the radiotherapy planning system determines that the current optimization phase has changed by comparing optimization status indicators with preset threshold conditions. For example, the system can determine that the optimization has entered the convergence phase from the transition phase based on the objective function value reduction rate changing from above a preset threshold to below a preset threshold.

[0057] Optionally, the trigger can indicate that the system automatically initiates the auxiliary structure reconstruction process in response to the detection results of the planned optimization phase switch, without manual intervention. The reconstruction operation can refer to one or more adjustments to the existing auxiliary structure. These adjustments can include: increasing or decreasing the number of ring structures, increasing or decreasing the number of shell structure layers, adjusting the expansion distance or thickness parameters of the auxiliary structure, replacing some ring structures with shell structures, adding avoidance structures, or reallocating the dose constraint targets and weight parameters corresponding to each auxiliary structure.

[0058] Optionally, the specific content of the reconstruction operation is related to the current planned optimization phase. In the initial coverage phase, reconstruction operations triggered after a phase switch can include: reducing the weights corresponding to the dose constraint targets of the gradient control structure, reducing the number of layers in the multi-layer dose control structure, or reducing the weights corresponding to the dose constraint targets of the avoidance structure, which helps to prioritize target dose coverage. In the transition optimization phase, triggered reconstruction operations can include: increasing the number of layers in the multi-layer dose control structure, introducing avoidance structures, or increasing the gradient constraint strength of the ring structure, which helps to simultaneously control the dose gradient and protect organs at risk. In the convergence phase, triggered reconstruction operations can include: adding an outer ring structure to control low-dose spillover, reducing the thickness of the shell structure to achieve finer dose sculpting, or increasing the weights of constraints related to organs at risk.

[0059] Radiotherapy planning systems enable auxiliary structures to be adjusted in tandem with the optimization process, such as focusing on target coverage, dose gradient control, or organ protection at different stages. This helps avoid interference from static auxiliary structures in later optimization stages, which may occur in related technologies, and improves the convergence efficiency of inverse optimization and the quality of the final plan.

[0060] For example, if the radiotherapy planning system detects that the target coverage has reached 92% and the overdose volume of the organ at risk begins to fluctuate, it can determine that the planning optimization phase is switching from the initial coverage phase to the transitional optimization phase. In response to the switch from the initial coverage phase to the transitional optimization phase, the system can automatically trigger the following reconfiguration operations: increasing the number of shell-like structures from the current 1 layer to 3 layers; introducing a new avoidance structure between the target area and the adjacent spinal cord; and increasing the gradient constraint weight of the ring structure from 0.2 to 0.6. After reconfiguration, the system continues to perform inverse optimization based on the updated auxiliary structure configuration.

[0061] For example, if the radiotherapy planning system detects that the rate of decrease in the objective function value has been below 0.8% for four consecutive iterations, and the change in the uniformity index of the dose distribution is less than 0.1%, then the planning optimization phase is switched from the transition optimization phase to the convergence phase. In response to the switch from the transition optimization phase to the convergence phase, the system can automatically trigger the following reconstruction operations: adding an outer ring structure around the target area to suppress unintended diffusion in low-dose regions; finely adjusting the thickness of the current shell structure from 8 mm to 3 mm; and increasing the weight of organ-related constraints from 0.5 to 0.9. After reconstruction, the system can continue optimization until the preset convergence conditions are met.

[0062] In some alternative embodiments, the auxiliary structure includes at least one of the following: a gradient control structure for constraining the rate of dose decrease outside the target area; a multi-layer dose control structure for hierarchically controlling the dose distribution at different spatial distances; and a repulsion region structure for controlling the extent to which the amount of inhibitor spills out of the target area between the target area and the organ at risk.

[0063] Optionally, the gradient control structure can represent a ring-shaped spatial region closely adjacent to or at a small distance from the target area, which helps to constrain the rate at which the dose outside the target area decreases with increasing distance. In helical tomotherapy, the gradient control structure can be generated by performing an outward expansion operation on the outer contour of the target area and subtracting the original target area contour. For example, a ring-shaped region with a thickness of 5 mm can be created, and a dose gradient constraint target can be assigned to this ring-shaped region, forcing the dose within the ring-shaped region to rapidly decrease from 100% of the prescribed dose at the target boundary to a lower level.

[0064] Optionally, a multi-layered dose control structure can represent a set of multi-shell-like spatial regions distributed at different distances around the target area, which is beneficial for layered control of dose distribution at different spatial distances. Unlike single-layered gradient control structures, multi-layered dose control structures can finely manage the dose decay curve through multiple levels of constraints. For example, a three-layered shell structure can be created: the first layer, 0 mm to 5 mm from the target area, is used to control steep gradients; the second layer, 5 mm to 15 mm from the target area, is used to control medium-range doses; and the third layer, 15 mm to 30 mm from the target area, is used to suppress low-dose tailing effects.

[0065] Optionally, a rejection region structure can represent a specific spatial region constructed between the target area and adjacent organs at risk, used to determine the extent to which inhibitory dose spills out of the target area between the target area and the organs at risk. Unlike gradient control structures, rejection regions are not arranged around the target area, but are strategically placed in the adjacent or intersecting regions between the target area and the organs at risk. The radiotherapy planning system can assign a maximum dose constraint target to this rejection region structure, thereby guiding the optimizer to reduce dose output within the rejection region structure.

[0066] By functionally differentiating and defining auxiliary structures, it is beneficial to provide clear tool types for subsequent automatic generation and dynamic reconstruction. Gradient control structures are used to create steep dose cliffs, multi-layer dose control structures are used to manage dose decay curves in a tiered manner, and exclusion region structures are used to provide strong directional protection when there is conflict between the organ at risk and the target area. Clear functional divisions facilitate the flexible combination of different auxiliary structures according to different clinical scenarios such as target size and proximity to the organ at risk, thereby achieving better dose conformity and organ at risk protection.

[0067] For example, in a helical tomotherapy plan for prostate cancer, there are no adjacent organs of danger around the target volume. The radiotherapy planning system can automatically generate a gradient control structure with a thickness of 8 mm, which is a ring-shaped region, and assign dose gradient constraint targets to the gradient control structure such that the dose at any point within the ring-shaped region does not exceed 70% of the prescribed dose. Furthermore, the system can generate a three-layer, multi-layer dose control structure, for example, including: a first shell located 8 mm to 15 mm from the target boundary, a second shell located 15 mm to 25 mm from the target boundary, and a third shell located 25 mm to 40 mm from the target boundary. Gradient constraint targets can be assigned to the three shell regions respectively: the dose of the first shell does not exceed 50% of the prescribed dose, the dose of the second shell does not exceed 35% of the prescribed dose, and the dose of the third shell does not exceed 20% of the prescribed dose. Through combined configuration, the system can form a control network with a gradual transition from high to low doses around the target volume.

[0068] For example, in helical tomotherapy planning for head and neck tumors, the minimum distance between the target volume and the brainstem is only 2 mm. The radiotherapy planning system can generate a repulsion region structure in the vicinity of the target volume and the brainstem. This repulsion region structure can be configured as a wedge-shaped spatial region that covers the entire adjoint surface between the target volume and the brainstem. The system assigns a maximum dose constraint target to this repulsion region structure: the maximum dose within the repulsion region structure should not exceed 50 Gy. This constraint can be set to high priority. During inverse optimization, this repulsion region structure can effectively avoid high-dose areas, creating a steep dose drop in the target volume towards the brainstem, which helps protect the brainstem dose from falling below the clinically limited threshold. Furthermore, the system can still generate gradient control structures and multi-layer dose control structures normally in other directions around the target volume to ensure dose conformity in other directions around the target volume.

[0069] In some optional embodiments, an initial auxiliary structure is generated based on the geometric features of the target area and the spatial relationship between the target area and the organ at risk. This includes: the radiotherapy planning system determining the volume of the target area based on its geometric features; determining the number of layers of the auxiliary structure based on the volume of the target area; determining the minimum distance between the target area and the organ at risk based on the spatial location information of the target area and the organ at risk; determining the type combination of the auxiliary structure based on the minimum distance; determining the thickness parameter of the auxiliary structure based on the current planning optimization stage, wherein the thickness parameter is used to constrain the distance range of the auxiliary structure extending outward from the outer surface of the target area; and generating the initial auxiliary structure based on the number of layers of the auxiliary structure, the type combination of the auxiliary structure, and the thickness parameter of the auxiliary structure.

[0070] Optionally, the target volume can refer to the total volume occupied by the target area, calculated by the radiotherapy planning system through three-dimensional voxel accumulation after extracting the target contour from the patient's structure set.

[0071] Optionally, the number of auxiliary structures can refer to the number of multi-layer dose control structures generated around the target area. The radiotherapy planning system determines the number of auxiliary structures based on the target volume. The basic rule for determining the number of layers is: the larger the target volume, the more auxiliary structures are required. Because large-volume target areas such as giant soft tissue sarcomas often have irregular shapes, the dose is more likely to spread uncontrollably and unexpectedly at the distal end. Therefore, more layers of control structures are needed to constrain the dose distribution step by step, which is beneficial for achieving better conformity. Conversely, the dose gradient around small-volume target areas such as small intracranial lesions is easier to control, and too many control layers can increase the unnecessary computational burden of optimization. For example, when the detected target volume is less than 50 cubic centimeters, the system can determine that the number of auxiliary structures is 1 to 2 layers; when the target volume is greater than or equal to 150 cubic centimeters, the system can determine that the number of auxiliary structures is 3 to 5 layers.

[0072] Optionally, the minimum distance between the target area and the organ at risk can refer to the shortest spatial distance from the outer surface of the target area to the outer surface of a certain organ at risk, used to quantify the proximity between the target area and the organ at risk. The radiotherapy planning system determines the type combination of auxiliary structures based on this minimum distance, that is, deciding which of the gradient control structures, multi-dose control structures, and rejection region structures to use. When the minimum distance is very small, for example, less than 5 mm, a conventional gradient control structure may not be sufficient to provide adequate protection, and the system needs to generate at least a rejection region structure as an auxiliary structure; when the minimum distance is moderate, for example, between 5 mm and 20 mm, the system can generate both gradient control structures and multi-dose control structures to form an auxiliary structure; when the minimum distance is large, for example, greater than or equal to 20 mm, the system can generate only gradient control structures to form an auxiliary structure.

[0073] Optionally, the current planning optimization stage can refer to the stage in the inverse optimization process, such as the initial coverage stage, the transition optimization stage, or the convergence stage. The thickness parameter is used to constrain the distance the auxiliary structure extends outward from the outer surface of the target area. For example, a ring structure with a thickness parameter of 10 mm can characterize the region formed by extending 10 mm outward from the target area boundary. The system determines the thickness parameter of the auxiliary structure based on the current planning optimization stage: in the initial coverage stage, a thicker thickness parameter can be used to ensure the stability of the initial optimization; in later stages, the thickness parameter can be gradually reduced for finer dose adjustments.

[0074] By decomposing the complex problem of generating auxiliary structures into independent decision dimensions based on volume, distance, and stage, the radiotherapy planning system facilitates the refined and personalized automatic generation of auxiliary structure configurations, thereby avoiding the unreasonable constraints brought about by the one-size-fits-all fixed rules in related technologies.

[0075] For example, a radiotherapy planning system can acquire a lung target volume of 120 cubic centimeters, which is a medium size between 50 and 150 cubic centimeters. Based on this, the system can determine that two layers of auxiliary structures need to be generated. The system further calculates that the minimum distance between the target volume and the spinal cord is 15 millimeters, which is a medium distance between 5 and 20 millimeters. Based on this, the system can determine that the type combination of auxiliary structures is a gradient control structure plus a multi-layer dose control structure, excluding rejection region structures. The system is currently in the initial coverage phase, so the thickness parameter of the first auxiliary structure can be set to 8 millimeters, and the thickness parameter of the second auxiliary structure can be set to 15 millimeters. Based on the determined number of layers, type combination, and thickness parameters, the system can automatically generate the initial auxiliary structures through morphological dilatation calculations.

[0076] For example, the radiotherapy planning system can acquire a small intracranial target region with a volume of 35 cubic centimeters, which is less than 50 cubic centimeters. Based on this, the system can determine that only one auxiliary structure needs to be generated. The system calculates that the minimum distance between the target region and the optic chiasm is only 2 millimeters, less than 5 millimeters, which is considered a very close proximity. Based on this, the system determines that the type combination of auxiliary structures needs to include a rejection region structure. The system is currently in a transitional optimization phase, so the thickness parameter of the rejection region structure can be set to 5 millimeters. Based on the determined number of layers, type combination, and thickness parameters, the system automatically generates the rejection region structure in the vicinity of the target region and the optic chiasm, while simultaneously generating gradient control structures in other directions.

[0077] In some optional embodiments, determining the number of auxiliary structures based on the target volume includes: when the radiotherapy planning system detects that the target volume is less than a first volume threshold, determining the number of auxiliary structures to be a first range of layers; when the target volume is detected to be greater than or equal to the first volume threshold and less than a second volume threshold, determining the number of auxiliary structures to be a second range of layers; when the target volume is detected to be greater than or equal to the second volume threshold, determining the number of auxiliary structures to be a third range of layers; wherein the maximum number of layers in the first range of layers is less than the minimum number of layers in the second range of layers, and the maximum number of layers in the second range of layers is less than the minimum number of layers in the third range of layers.

[0078] Optionally, the first volume threshold and the second volume threshold are two preset critical values ​​used to classify the target area size. For example, the first volume threshold can be set to 50 cubic centimeters, and the second volume threshold can be set to 150 cubic centimeters. The first layer range, the second layer range, and the third layer range correspond to the auxiliary structure layer ranges for small-volume, medium-volume, and large-volume target areas, respectively. For example, the first layer range can be 1 to 2 layers, the second layer range can be 2 to 3 layers, and the third layer range can be 3 to 5 layers. The smaller the target area volume, the fewer auxiliary structure layers are required; the larger the target area volume, the more auxiliary structure layers are required. Therefore, the maximum number of layers in the first layer range is less than the minimum number of layers in the second layer range, and the maximum number of layers in the second layer range is less than the minimum number of layers in the third layer range, which is beneficial for forming a clear hierarchical layer progression relationship.

[0079] Optionally, after acquiring the volume value of the target area, the radiotherapy planning system compares this volume value with a first volume threshold and a second volume threshold. When the volume of the target area is detected to be less than the first volume threshold, for example, less than 50 cubic centimeters, the system determines that the target area belongs to a small volume target area and determines the number of auxiliary structures within a first range, such as 1 to 2 layers. When the volume of the target area is detected to be greater than or equal to the first volume threshold and less than the second volume threshold, for example, greater than or equal to 50 cubic centimeters and less than 150 cubic centimeters, the system determines that the target area belongs to a medium volume target area and determines the number of auxiliary structures within a second range, such as 2 to 3 layers. When the volume of the target area is detected to be greater than or equal to the second volume threshold, for example, greater than or equal to 150 cubic centimeters, the system determines that the target area belongs to a large volume target area and determines the number of auxiliary structures within a third range, such as 3 to 5 layers.

[0080] By introducing specific volume thresholds and corresponding slice ranges, radiotherapy planning systems facilitate precise matching of the number of auxiliary structure slices to the target size. For small target volumes, using fewer control slices avoids over-constraint and promotes faster convergence; for large target volumes, using more control slices effectively suppresses unwanted low-dose diffusion, improving the conformity of large target planning. A quantitative, volume-integrated dynamic slice selection strategy helps achieve a better balance between optimization efficiency and planning quality.

[0081] For example, a radiotherapy planning system can acquire a target region for an intracranial metastasis, and calculate the volume of this target region to be 25 cubic centimeters using three-dimensional voxel aggregation. The system compares this volume value with a first volume threshold, determining that 25 cubic centimeters is less than 50 cubic centimeters. According to preset rules, the system determines the number of auxiliary structures to be within a first range, such as 1 to 2 layers. Considering that the target region has a regular shape and no adjacent organs at risk, the system ultimately selects to generate a 1-layer gradient control structure as the auxiliary structure to control the rapid drop in dose outside the target region.

[0082] For example, a radiotherapy planning system can acquire a target region for a large soft tissue sarcoma in the pelvis, calculating its volume to be 280 cubic centimeters. The system compares this volume value with a first volume threshold and a second volume threshold, determining that 280 cubic centimeters is greater than 150 cubic centimeters. According to preset rules, the system can determine the number of auxiliary structures to be within a third range, such as 3 to 5 layers. Considering the large volume and irregular shape of the target region, the system can ultimately select to generate four multi-layer dose control structures, located in four shell-like regions at distances of 0 mm to 8 mm, 8 mm to 18 mm, 18 mm to 30 mm, and 30 mm to 45 mm from the target region boundary, respectively, to progressively constrain dose distribution and suppress unintended diffusion of low doses to the distal pelvis.

[0083] In some optional embodiments, the combination of auxiliary structures is determined based on the minimum distance between the target area and the organ at risk, including: when the radiotherapy planning system detects that the minimum distance is less than a first distance threshold, generating at least a rejection region structure as an auxiliary structure; when the minimum distance is detected to be greater than or equal to the first distance threshold and less than a second distance threshold, generating at least a gradient control structure and a multi-layer dose control structure to form an auxiliary structure; and when the minimum distance is detected to be greater than or equal to the second distance threshold, generating a gradient control structure to form an auxiliary structure.

[0084] Optionally, the first distance threshold and the second distance threshold are preset distance thresholds used to classify the risk level between the target area and the organ at risk. For example, the first distance threshold can be set to 5 mm, and the second distance threshold can be set to 20 mm. The minimum distance can refer to the shortest spatial distance between the outer surface of the target area and the outer surface of a certain organ at risk. The core logic of this claim is to dynamically configure different functional combinations of auxiliary structures according to the distance between the organ at risk and the target area to achieve a differentiated protection strategy. When the organ at risk is close to the target area (the minimum distance is less than the first distance threshold), conventional gradient control structures may not be sufficient to provide adequate protection; when the distance between the organ at risk and the target area is moderate, an effective dose buffer can be formed by combining structures; when the distance between the organ at risk and the target area is far, its radiation risk is very low, and there is no need to introduce complex protective structures.

[0085] Optionally, after calculating the minimum distance between the target area and a critical organ, the radiotherapy planning system compares this minimum distance with a first distance threshold and a second distance threshold. The first scenario is: when the minimum distance is less than the first distance threshold, the system determines that the critical organ is within a dangerous distance range, and at least a rejection region structure is generated as an auxiliary structure. The rejection region structure can create a region with a strong inhibitory dose overflow between the target area and the critical organ, guiding the optimizer to significantly reduce the dose output within this region. The second scenario is: when the minimum distance is greater than or equal to the first distance threshold and less than the second distance threshold, the system determines that the critical organ is within a moderate distance range, and at least a gradient control structure and a multi-layer dose control structure are generated as auxiliary structures to control the dose gradient around the target area while protecting the critical organ. The third scenario is: when the minimum distance is greater than or equal to the second distance threshold, the system determines that the critical organ is within a safe distance range, and only a gradient control structure is generated as an auxiliary structure.

[0086] Radiotherapy planning systems, through differentiated configuration strategies based on distance thresholds, facilitate adaptive protection of organs at risk, avoiding over-constraint or under-protection issues caused by using the same structural combinations in all cases. This helps to balance the protection of organs at risk with the avoidance of unnecessary optimization constraints.

[0087] For example, in a helical tomotherapy plan for head and neck tumors, the radiotherapy planning system can calculate a minimum distance of 2 mm between the target area and the brainstem. The system compares this minimum distance with a first distance threshold and determines that 2 mm is less than the first distance threshold. According to preset rules, the system generates at least one rejection region structure as an auxiliary structure in the adjacent area between the target area and the brainstem. This rejection region structure is configured as a wedge-shaped spatial region covering the entire adjacent surface between the target area and the brainstem. The system assigns a maximum dose constraint target to this rejection region structure: the maximum dose within the rejection region structure does not exceed 50 Gy, so as to force the dose to form a depression in this region, which is beneficial to protect the brainstem and reduce over-irradiation.

[0088] For example, in a helical tomotherapy plan for lung tumors, the radiotherapy planning system calculates a minimum distance of 15 mm between the target area and the spinal cord. The system compares this minimum distance with a first distance threshold and a second distance threshold, determining that 15 mm is greater than or equal to the first distance threshold and less than the second distance threshold. According to preset rules, the system generates at least a gradient control structure and a multi-layer dose control structure to form an auxiliary structure. The system generates a 5 mm thick ring structure as the gradient control structure, and simultaneously generates two shell-like structures as multi-layer dose control structures, located in two regions 5 mm to 15 mm and 15 mm to 30 mm from the target area boundary, respectively, to control the dose gradient distribution around the target area while protecting the spinal cord.

[0089] In some optional embodiments, the auxiliary structure is configured in association with the inverse optimization objective, including: when the radiotherapy planning system detects that the auxiliary structure includes at least one of a gradient control structure and a multi-layer dose control structure, assigning a dose gradient constraint objective to each structure, wherein the dose gradient constraint objective is used to limit the rate at which the dose outside the target area decreases with increasing distance; when the auxiliary structure is detected to include a repulsion region structure, assigning a maximum dose constraint objective to the repulsion region structure, wherein the maximum dose constraint objective is used to limit the highest dose value within the repulsion region structure; and configuring at least one of the dose gradient constraint objective and the maximum dose constraint objective in association with the inverse optimization objective, such that the constraint strength of the auxiliary structure is adjusted as the planning optimization phase changes.

[0090] Optionally, the dose gradient constraint target can refer to a functional relationship used to limit the rate of dose decrease outside the target area with increasing distance. This constraint target can be defined in the form of "the dose at a distance of X millimeters from the target area does not exceed Y% of the prescribed dose." For example, a dose gradient constraint target can be assigned to a gradient control structure located 0 to 5 millimeters outside the target area: the dose at any point within the structure does not exceed 80% of the prescribed dose. The radiotherapy planning system assigns a dose gradient constraint target to each structure if it detects that the auxiliary structures include at least one of gradient control structures or multi-layer dose control structures.

[0091] Optionally, the maximum dose constraint target can refer to a functional relationship used to limit the highest dose value within a specific structure, and can be defined in the form of "the maximum dose within the structure does not exceed Z Gy". For example, a maximum dose constraint target can be assigned to an avoidance structure as: the maximum dose within the avoidance structure does not exceed 45 Gy. The radiotherapy planning system assigns a maximum dose constraint target to the rejection region structure when it detects that the auxiliary structure includes a rejection region structure.

[0092] Optionally, at least one of the dose gradient constraint objective and the maximum dose constraint objective can be associated with the inverse optimization objective, so that the constraint strength of the auxiliary structure can be adjusted as the planning optimization stage changes. The constraint strength can represent the weight of a certain constraint objective in the optimization objective function. The larger the weight, the more computing power the optimizer invests to satisfy the constraint. For example, the inverse optimization objective function can be referred to by formula (1):

[0093] Formula (1);

[0094] in, It can represent the value of the optimization objective function that quantifies how well the radiotherapy plan meets clinical requirements. Indicates the index identifier of multiple optimization objective items participating in reverse optimization. This represents the relative importance coefficient assigned to each optimization objective in multi-objective inverse optimization, used to balance the priorities among different optimization objectives. This indicates the dose indicators that need to be optimized during the reverse optimization of radiotherapy plans, including target coverage, dose constraints on organs at risk, and dose gradient constraints from auxiliary structures.

[0095] For example, gradient sub-terms associated with gradient control structures Configurable weights Maximum dose sub-item related to avoidance structure Configurable weights The radiotherapy planning system dynamically adjusts these weights based on the current planning optimization phase: in the initial coverage phase, the weights can be adjusted accordingly. and The weight is set relatively low, such as 0.1, to avoid interfering with target dose coverage; during the transition optimization phase, the weight can be increased to a medium level, such as 0.5; during the convergence phase, the weight can be further increased, such as 1.0, to finely control the dose gradient and protect organs at risk.

[0096] By linking the constraint strength of auxiliary structures to the optimization stage, the radiotherapy planning system enables the auxiliary structures to play different roles in different optimization stages. This helps avoid the optimization interference caused by the constant constraint strength of auxiliary structures in related technologies, and helps improve the convergence efficiency of inverse optimization and the quality of the final plan.

[0097] For example, the radiotherapy planning system detects a gradient control structure within the auxiliary structure. The system assigns a dose gradient constraint objective to this gradient control structure: "The dose at any point within this annular region shall not exceed 70% of the prescribed dose." The system associates this dose gradient constraint objective with the gradient sub-item in the inverse optimization objective and sets an initial weight of 0.2. When optimization enters the transition phase, for example, when target coverage has reached over 90%, the system automatically adjusts this weight from 0.2 to 0.7, making the optimizer place greater emphasis on the constraint of this gradient control structure, thereby creating a steeper dose drop around the target area. When optimization enters the convergence phase, the system further adjusts the weight to 1.0 for final dose gradient refinement.

[0098] For example, the radiotherapy planning system detects a rejection region structure within the auxiliary structure. The system assigns a maximum dose constraint target to this rejection region structure as: "The maximum dose within this rejection region structure shall not exceed 45 Gy". The system associates this maximum dose constraint target with the OAR dose constraint term in the inverse optimization objective. In the initial stage of inverse optimization, the weight of this constraint is set to 0.1 to avoid over-restricting the dose into the target area. When the system detects that the target coverage has reached 95% and the optimization has entered the convergence phase, the system automatically increases the weight of this constraint from 0.1 to 0.9, causing the optimizer to strongly suppress the dose within the rejection region structure, ensuring that the maximum spinal cord dose is strictly below the clinically limited threshold of 45 Gy.

[0099] In some optional embodiments, the inverse optimization objective represents the dose index that needs to be optimized to achieve during the inverse optimization of the radiotherapy plan, including at least one of target dose coverage, organ at risk dose limitation threshold, and extra-target dose gradient threshold; the weights of the optimization objective function are relative importance coefficients assigned to each optimization objective in multi-objective inverse optimization, used to balance the priorities among different optimization objectives.

[0100] Optionally, the inverse optimization objective can represent a series of dosimetric indicators that need to be approximated or achieved by the optimization algorithm during the inverse optimization of the radiotherapy plan. These indicators specifically include at least one of target dose coverage, organ-at-risk dose limitation threshold, and extra-target dose gradient threshold.

[0101] Optionally, target dose coverage is one of the core metrics for evaluating the quality of radiotherapy planning, representing the percentage of the target volume that receives or exceeds the prescribed dose. For example, a clinical requirement is that the prescribed dose covers 95% of the target volume, meaning that at least 95% of the target volume receives the prescribed dose.

[0102] Optionally, the organ at risk dose limit threshold is an upper limit of dose set by clinicians to protect normal tissues. For example, the maximum point dose to the spinal cord must not exceed 45 Gy. This threshold is used to constrain the radiation dose to organs at risk.

[0103] Optionally, the dose gradient threshold outside the target area is used to represent the required rate at which the dose outside the target area decreases with increasing distance; for example, the dose at a distance of 5 mm from the target boundary shall not exceed 70% of the prescribed dose. This threshold is used to constrain the steepness of the dose drop outside the target area.

[0104] Optionally, the weights of the objective function can refer to the relative importance coefficients assigned to each optimization objective in multi-objective inverse optimization, used to balance the priorities among different optimization objectives. In the inverse optimization of helical tomotherapy, the objective function can take the form of a multi-objective weighted function, for example, it can be expressed as... .in, Indicates the first Sub-targets, such as target coverage sub-target, dose-constrained target for organs at risk, and dose gradient-constrained target. This indicates the weight assigned to the sub-objective. Different optimization objectives may conflict; for example, increasing target coverage may increase the radiation dose to nearby organs at risk. By assigning different weight values ​​to different sub-objectives, the optimizer can understand which objective is more important at the current stage, thus making reasonable trade-offs.

[0105] For example, a radiotherapy planning system can configure inverse optimization objectives for prostate cancer planning. The first sub-objective is target volume dose coverage: requiring at least 95% of the planned target volume to receive a prescribed dose of 78 Gy. The second sub-objective is the organ at risk dose limitation threshold: requiring the rectum's V70 to be less than 15%, i.e., the percentage of volumes receiving 70 Gy or more is less than 15%. The system assigns a weight of 1.0 to the target volume coverage sub-objective and a weight of 0.8 to the rectal dose limitation sub-objective. During inverse optimization, the optimizer, based on the relative magnitudes of these two weights, suppresses the rectal dose below the threshold as much as possible while prioritizing target volume coverage.

[0106] For example, the radiotherapy planning system configures inverse optimization objectives for brain tumor planning. In addition to the target coverage sub-objective, the system can also configure an extra-target dose gradient threshold sub-objective: requiring that the dose at any point within a ring-shaped area 5 mm from the planned target boundary does not exceed 60% of the prescribed dose. The system weights this dose gradient constraint sub-objective with the target coverage sub-objective. Initially, the gradient constraint weight is set to 0.3. Once the target coverage is achieved, the system dynamically adjusts this weight to 0.9, causing the optimizer to later emphasize the steepness of the dose drop around the target area to protect surrounding normal brain tissue.

[0107] In some optional embodiments, during the reverse optimization process, optimization status indicators are collected, and the switching of the planned optimization stage is determined based on the optimization status indicators. This includes: during the reverse optimization process, monitoring at least the following optimization status indicators: the change in target coverage, the change in volume of organs at risk exceeding a preset dose threshold, the rate of decrease in the function value of the optimization objective function as the number of iterations increases, and the change in the uniformity of dose distribution; comparing the monitored optimization status indicators with preset stage switching conditions; if the optimization status indicators meet the stage switching conditions, determining that the planned optimization stage has changed; if the optimization status indicators do not meet the stage switching conditions, determining that the planned optimization stage has not changed.

[0108] Optionally, the change in target coverage can be used to represent the magnitude of the change in target dose coverage between two monitoring periods or two iterations. For example, if the target coverage was 85% in the previous period and 86% in the current period, the change is 1%. A small change that gradually decreases and tends to stabilize can indicate that the target coverage has basically met the target.

[0109] Optionally, the change in volume of an organ at risk exceeding a preset dose threshold can be expressed as the percentage change in the volume of an organ at risk exceeding a preset dose limit threshold between two monitoring periods. For example, if the volume of the spinal cord exceeding 45 Gy in the previous period was 5%, and in the current period it is 4.8%, then the change is 0.2%.

[0110] Optionally, the rate of decrease in the function value of the objective function as the number of iterations increases can represent the magnitude of the decrease in the objective function value per unit number of optimization iterations. A gradually decreasing rate of decrease can indicate that the optimization has entered the convergence phase. The change in the dose distribution uniformity index can represent the magnitude of change in the uniformity of dose distribution within the target area between two monitoring periods.

[0111] Optionally, the radiotherapy planning system compares the monitored optimization status indicators with preset stage switching conditions. Stage switching conditions are a set of pre-defined rules; for example, when the change in target coverage is less than 0.5% for three consecutive cycles and the function value shrinkage rate is less than 1% per iteration, the system determines that it has entered the convergence phase. If the optimization status indicators meet the stage switching conditions, the system determines that the planning optimization phase has changed and triggers subsequent auxiliary structure reconstruction operations. If the optimization status indicators do not meet the stage switching conditions, the system determines that the planning optimization phase has not changed, continues to perform reverse optimization, and continuously monitors various indicators.

[0112] By simultaneously monitoring multiple optimization status indicators such as coverage, overdose volume, function decay rate, and homogeneity, radiotherapy planning systems provide a more stable stage determination mechanism than single-indicator monitoring, avoiding false triggers or missed triggers caused by occasional fluctuations in a single indicator. The method of joint monitoring and comparison of multiple indicators helps to more accurately capture the true changes in the optimization stage, which is beneficial for the dynamic reconstruction of subsequent auxiliary structures.

[0113] For example, during the reverse optimization process, the radiotherapy planning system calculates the rate of reduction of the objective function value after each iteration cycle. The system's preset stage switching condition is: when the rate of reduction of the objective function value is below 0.8% for three consecutive iteration cycles, the optimization stage is considered to have switched to the convergence stage. In the 30th iteration, the system detected a rate of reduction of 0.7%, followed by 0.6% and 0.5% in the next two cycles, both below the 0.8% threshold. Based on this, the system determines that the optimization status indicator meets the stage switching condition, confirming that the planning optimization stage has changed.

[0114] For example, the radiotherapy planning system simultaneously monitors changes in target coverage and changes in overdose volume in organs at risk. The system's preset joint phase switching condition is: when the change in target coverage is less than 0.5% and the change in volume exceeding a preset dose threshold in organs at risk is less than 0.1%, the optimization phase is determined to switch from the transitional optimization phase to the convergence and refinement phase. During the optimization process, the system calculated that the average change in target coverage over the most recent five iterations was 0.3%, and the average change in overdose volume in organs at risk was 0.08%, both indicators meeting the preset conditions. Based on this, the system determines that the optimization status indicators meet the phase switching condition, confirms that the planning optimization phase has changed, and prepares to trigger subsequent auxiliary structure reconstruction operations.

[0115] For example, the change in target coverage can be represented by the symbol ΔV95; the change in volume of organs at risk exceeding the preset dose threshold can be represented by the symbol ΔOAR; the calculation of the rate of reduction of the function value of the optimized objective function can be referred to formula (2):

[0116] Formula (2);

[0117] Where R represents the rate at which the function value of the objective function is reduced. Indicates the first The value of the objective function at the next iteration. Indicates the first The objective function value at the next iteration; the change in the dose distribution uniformity index is represented by the symbol HI. The radiotherapy planning system compares the monitored optimization status indices with the preset stage switching conditions, and determines whether the current planning optimization stage has changed based on the comparison results.

[0118] The radiotherapy planning system compares the monitored optimization status indicators with preset stage switching conditions. Triggering mechanisms include single-indicator triggering and multi-indicator joint triggering. Single-indicator triggering occurs when the rate of decrease (R) of the optimization objective function is less than a preset rate threshold, at which point the radiotherapy planning system initiates the auxiliary structure reconfiguration process. Multi-indicator joint triggering occurs when the change in target coverage (ΔV95) is less than a first preset threshold (a), the change in volume (ΔOAR) of organs at risk exceeding a preset dose threshold is less than a second preset threshold (b), and the rate of decrease (R) of the optimization objective function is less than a third preset threshold (c), at which point the radiotherapy planning system determines that the planning optimization stage has entered a new stage. When preset stage change conditions are met, the radiotherapy planning system triggers the auxiliary structure reconfiguration process.

[0119] For example, when the rate of decrease R of the objective function is less than 1%, the radiotherapy planning system determines that the planning optimization phase has entered the convergence phase. As another example, when the change in target coverage ΔV95 is less than 0.5%, and the change in volume ΔOAR of organs at risk exceeding a preset dose threshold is less than a preset stability threshold, the radiotherapy planning system determines that the planning optimization phase has stabilized.

[0120] like Figure 2 As shown, firstly, the radiotherapy planning system can extract and preprocess data based on the geometric features of the target area and the spatial relationship between the target area and organs at risk. Data extraction and preprocessing refers to the system acquiring the three-dimensional contour information of the target area and organs at risk from the patient's structural dataset, and performing voxelization, morphological dilation and erosion operations, and overlap analysis on the acquired contour data. Based on the preprocessed data, an initial auxiliary structure is generated, which serves as the spatial region to constrain dose distribution during radiotherapy planning optimization. Secondly, the system associates the generated auxiliary structure with the inverse optimization objective, linking the parameters of the auxiliary structure with the weights of the objective function. Subsequently, during inverse optimization, the system continuously collects optimization status indicators and determines whether a shift in the planning optimization stage has occurred based on these indicators. If no shift has occurred, the system returns to collecting and monitoring optimization status indicators; if a shift has occurred, the system triggers a reconstruction operation of the auxiliary structure based on the change in the planning optimization stage. Finally, after completing the reconstruction operation, the system continues inverse optimization until the plan meets the convergence criteria.

[0121] In some optional embodiments, the planning optimization phase includes an initial coverage phase, a transition optimization phase, and a convergence phase. The determination criteria for the initial coverage phase include: the change in target coverage is less than a first preset threshold, and the rate of decrease of the function value reduction rate is greater than a first rate threshold. The determination criteria for the transition optimization phase include: the change in target coverage is greater than or equal to a first preset threshold, and the change in the volume of organs at risk exceeding a preset dose threshold is greater than a second preset threshold. The determination criteria for the convergence phase include: the rate of decrease of the function value is less than a second rate threshold, and the change in the uniformity index of the dose distribution is less than a third preset threshold, wherein the second rate threshold is less than the first rate threshold.

[0122] Optionally, the optimization phase can be divided into three consecutive phases: the initial coverage phase, the transitional optimization phase, and the convergence phase. The initial coverage phase can refer to the early stage of the inverse optimization process. In this initial coverage phase, the main task of the optimization algorithm is to drive the radiation dose to rapidly cover the target area, so that the target area dose reaches or approaches the prescribed dose level. At this time, the requirements for dose gradient and protection of organs at risk can be temporarily relaxed.

[0123] The transitional optimization stage can refer to the intermediate stage between initial coverage and final convergence. In this transitional optimization stage, the target coverage has basically met the requirements, and the optimization algorithm begins to focus more on how to control the dose gradient outside the target area and how to suppress the dose to organs at risk below the constraint threshold.

[0124] The convergence phase can refer to the later stage of the inverse optimization process. In this phase, all key dose indicators are close to the target value, the decrease in the objective function value becomes very slow, and the focus of optimization is to perform fine dose sculpting to further improve uniformity and conformity, and suppress unintended diffusion in low-dose regions.

[0125] Optionally, the criteria for determining the initial coverage phase include two parts: First, the change in target area coverage is less than a first preset threshold, indicating that the target area coverage has stabilized and is no longer increasing significantly; second, the rate of decrease in the function value reduction rate is greater than a first rate threshold, indicating that the objective function is decreasing rapidly and optimization is in an active state. When both conditions are met simultaneously, it indicates that optimization has entered the stable phase from the rapid coverage phase.

[0126] The criteria for determining the transition optimization phase include: the change in target coverage is greater than or equal to the first preset threshold, indicating that the coverage is still changing and has not yet fully stabilized; and the change in the volume of organs at risk exceeding the preset dose threshold is greater than the second preset threshold, indicating that the dose to organs at risk has begun to fluctuate or exceed the limit, requiring close attention.

[0127] The criteria for determining the convergence phase include: the rate of function value reduction being less than a second rate threshold, indicating that the objective function has decreased very slowly; and the change in the uniformity index of the dose distribution being less than a third preset threshold, indicating that the dose distribution has tended to stabilize. The second rate threshold is less than the first rate threshold; for example, the first rate threshold can be set to 5% per iteration cycle, and the second rate threshold can be set to 1% per iteration cycle, reflecting the transition from rapid descent to slow convergence.

[0128] The radiotherapy planning system clearly defines the three optimization stages—initial coverage stage, transition optimization stage, and convergence stage—through quantitative and multi-dimensional judgment conditions. This helps to provide clear and executable logical rules for automated stage identification and subsequent dynamic reconstruction, avoiding the subjectivity and inconsistency of relying on human experience judgment.

[0129] For example, during the inverse optimization process, the radiotherapy planning system monitors the following states: the change in target coverage averages 0.3% over the past five iterations, which is less than a first preset threshold; simultaneously, the rate of decrease in the function value reduction rate is 8% / iteration, 6% / iteration, and 5% / iteration over the past three iterations, respectively, all greater than the first rate threshold. Based on the simultaneous fulfillment of these two conditions, the system determines the current planning optimization stage as the initial coverage stage and continues to maintain the current auxiliary structure configuration, weakening the gradient control structure weights to prioritize target dose coverage.

[0130] For example, when the radiotherapy planning system detects that the function value reduction rate has dropped to 0.6% / iteration when the optimization has reached the 40th iteration, which is less than the second rate threshold; at the same time, the change in the dose distribution uniformity index has averaged 0.05% over the past five cycles, which is lower than the third preset threshold. Based on the simultaneous satisfaction of these two conditions, the system determines that the current planning optimization phase has switched to the convergence phase and triggers auxiliary structure reconstruction operations, such as adding an outer ring structure to control low-dose spillover and finely adjusting the thickness of the shell structure from 8 mm to 3 mm.

[0131] In some optional embodiments, the reconstruction operation includes: during the initial coverage phase, reducing the weights corresponding to the dose constraint targets of the gradient control structure when a gradient control structure exists; reducing the weights corresponding to the dose constraint targets of the multi-layer dose control structure and reducing the number of layers of the multi-layer dose control structure when a multi-layer dose control structure exists; reducing the weights corresponding to the dose constraint targets of the exclusion region structure when an exclusion region structure exists; during the transition optimization phase, increasing the number of layers of the multi-layer dose control structure when a multi-layer dose control structure exists; introducing an exclusion region structure when no exclusion region structure exists; increasing the gradient constraint strength of the gradient control structure when a gradient control structure exists; and during the convergence phase, adding outer gradient control structures, reducing the thickness of the multi-layer dose control structure when a multi-layer dose control structure exists, and increasing the weights of dose constraint targets related to organs at risk.

[0132] Optionally, differentiated auxiliary structure reconstruction operations can be defined for different planning optimization stages. In the initial coverage stage, the optimization focus can be on ensuring that the radiation dose can quickly and fully cover the target area. Therefore, the reconstruction operation can weaken all constraints that may hinder the dose from entering the target area. That is, when a gradient control structure exists, the weight corresponding to the dose constraint target of the gradient control structure is reduced; when a multi-layer dose control structure exists, the weight corresponding to the dose constraint target of the multi-layer dose control structure is reduced, and the number of layers of the multi-layer dose control structure is decreased; when a repulsion region structure exists, the weight corresponding to the dose constraint target of the repulsion region structure is reduced.

[0133] Optionally, during the transition optimization phase, once target coverage has been largely achieved, the optimization focus can shift to simultaneously controlling the dose gradient and protecting organs at risk. Therefore, the reconfiguration operation can enhance the control structure and introduce protective structures: increasing the number of layers in a multi-layered dose control structure when one exists; introducing a rejection region structure when none exists; and increasing the gradient constraint strength of a gradient control structure when one exists.

[0134] Optionally, during the convergence phase, the optimization focus can shift to finely sculpting the dose distribution, suppressing unintended low-dose diffusion, and ensuring that the dose to organs at risk strictly meets clinical requirements. Therefore, refactoring operations can include: adding outer gradient control structures, such as adding a more outer ring structure outside the current ring structure to tighten the boundary of the low-dose region; reducing the thickness of multi-layer dose control structures when they exist to achieve finer dose control; and increasing the weight of dose constraint targets related to organs at risk to ensure that the dose to organs at risk strictly meets clinically restrictive thresholds.

[0135] Radiotherapy planning systems employ a phased, differentiated reconfiguration strategy to synchronize the configuration of auxiliary structures with the inherent requirements of the optimization objective. For example, constraints are weakened in the initial coverage phase to accelerate target coverage; constraints are strengthened and structures are built in the transition optimization phase to establish effective dose gradients and protective barriers; and fine-tuning of dose is performed in the convergence phase to improve the overall quality of the plan. This dynamic adjustment strategy, tightly coupled with the optimization phases, helps to maximize the potential of each optimization stage, avoids constraint interference problems that may arise from static auxiliary structures in related technologies, thereby improving the convergence efficiency of inverse optimization and the dose conformity of the final radiotherapy plan.

[0136] For example, the radiotherapy planning system detects that the target coverage has reached 92% and that the overdose volume of the organ at risk has begun to fluctuate, thus determining that the planning optimization phase is switching from the initial coverage phase to the transitional optimization phase. In response to this phase switch, the system automatically triggers the following reconfiguration operations: increasing the number of layers in the multi-layer dose control structure from the current 1 layer to 3 layers; introducing a new exclusion region structure between the target area and the adjacent spinal cord; and increasing the gradient constraint weight of the gradient control structure from 0.2 to 0.6. After reconfiguration, the system continues to perform inverse optimization based on the updated auxiliary structure configuration.

[0137] For example, the radiotherapy planning system detects that the rate of decrease in the objective function value has been below 0.8% for four consecutive iterations, and the change in the dose distribution uniformity index is less than 0.1%, thus determining that the planning optimization phase is switching from the transition optimization phase to the convergence phase. In response to this phase switch, the system automatically triggers the following reconstruction operations: adding an outer gradient control structure outside the current ring structure to suppress unintended diffusion in low-dose regions; finely adjusting the thickness of the current shell structure from 8 mm to 3 mm; and increasing the weight of organ-related dose constraint targets from 0.5 to 0.9. After reconstruction, the system continues optimization until the preset convergence conditions are met.

[0138] In some optional embodiments, the reconstruction operation of the auxiliary structure includes at least one of the following: adding or reducing gradient control structures; adding or reducing the number of layers of multi-layer dose control structures; adjusting the expansion distance, thickness parameter, or spatial position of the auxiliary structure, wherein the expansion distance is the distance the auxiliary structure extends outward from the outer surface of the target area; replacing some gradient control structures with multi-layer dose control structures, or adding repulsion region structures; and reallocating the dose constraint targets corresponding to each type of structure in the auxiliary structure and the weight parameters of each dose constraint target.

[0139] Optionally, the reconstruction operation of auxiliary structures can be defined as an operation menu containing multiple adjustment methods. The first type of operation is structural adjustment operation: adding or reducing gradient control structures, that is, dynamically adding new ring structures or removing existing ring structures according to optimization needs; increasing or decreasing the number of layers in multi-layer dose control structures; replacing some gradient control structures with multi-layer dose control structures, or adding exclusion region structures. For example, if it is found during the optimization process that dose control in a certain direction requires more refined layered management, the ring structure in that direction can be replaced with a shell structure, or a new avoidance structure can be added between the target area and the organ at risk.

[0140] Optionally, the second type of operation is a parameter adjustment operation: adjusting the expansion distance, thickness parameter, or spatial position of the auxiliary structure. The expansion distance represents the numerical value of the distance the auxiliary structure extends outward from the outer surface of the target area; for example, adjusting the expansion distance of a ring structure from 5 mm to 8 mm can represent moving the ring structure outward by 3 mm. The thickness parameter represents the thickness of the auxiliary structure itself; for example, adjusting the thickness of a shell-like structure from 10 mm to 6 mm. The spatial position represents the layout of the auxiliary structure in three-dimensional space; for example, adjusting an avoidance structure from the left side of the target area to the right side to address asymmetrical distribution of organs at risk.

[0141] Optionally, the third type of operation is a weight reallocation operation: reallocating the dose constraint objectives corresponding to each type of structure in the auxiliary structure and the weight parameters of each dose constraint objective. For example, after the optimization enters the convergence phase, the system can increase the weight of the dose gradient constraint objective of the gradient control structure, while decreasing the weight of the target coverage constraint objective, so that the optimization focus shifts from target dose coverage to sculpting the dose edge.

[0142] Compared to related technologies that can only adjust a single parameter, the embodiments of this application can provide a flexible and rich set of reconfiguration operations, enabling the radiotherapy planning system to select the most appropriate adjustment method based on complex and ever-changing optimization conditions and clinical needs. Multiple reconfiguration operations can be used individually or in combination to address various anatomical scenarios and dosimetric challenges. For example, when a severe, unintended dose spread is detected in a certain direction, gradient control structures in that direction can be added, and their expansion distance can be adjusted simultaneously; when a dose to an organ at risk approaches a threshold, a new rejection region structure can be added, and a high-weighted maximum dose constraint target can be assigned to the rejection region structure.

[0143] For example, during the transition phase of inverse optimization, the radiotherapy planning system detects significant diffusion of the low-dose region to the right of the target area, while the dose gradient control to the left is good. The system can trigger a refactoring operation by: first, increasing the thickness parameter of the gradient control structure to the right from 6 mm to 12 mm to provide a wider dose attenuation buffer band; second, replacing the existing gradient control structure to the left with a multi-layered dose control structure, such as generating a two-shell structure, to achieve finer layered dose control in that direction. After completing the refactoring, the system continues with inverse optimization.

[0144] For example, after the inverse optimization enters the convergence phase, the radiotherapy planning system detects that the maximum dose to the spinal cord is approaching the clinical limit threshold of 45 Gy. The system triggers a reconfiguration operation: a new exclusion region structure is added between the target area and the spinal cord, and a maximum dose constraint target is assigned to this exclusion region structure: the maximum dose within this structure should not exceed 44 Gy. Simultaneously, the weight of this constraint target is reallocated from the default 0.5 to 0.95, making it the highest priority constraint at this stage. Furthermore, the system can also reallocate the weight of the dose gradient constraint target of the gradient control structure from 0.4 to 0.7 to further enhance the steepness of the dose drop outside the target area. After completing the weight reallocation and adding the new structure, the system continues optimization until the spinal cord dose is suppressed below the safe threshold.

[0145] like Figure 3 As shown, according to another aspect of the embodiments of this application, an auxiliary structure generation device for a helical tomotherapy system is also provided, including: a generation unit, an associated configuration unit, a first processing unit, and a second processing unit.

[0146] The system includes the following components: a generation unit, which generates an initial auxiliary structure based on the geometric features of the target area and the spatial relationship between the target area and organs at risk; an association configuration unit, which associates the auxiliary structure with the inverse optimization objective; a first processing unit, which collects optimization status indicators during the inverse optimization process and determines whether a switch has occurred in the planning optimization phase; and a second processing unit, which triggers a reconstruction operation of the auxiliary structure when a switch in the planning optimization phase is detected.

[0147] Optionally, the auxiliary structure includes at least one of the following: a gradient control structure for constraining the rate of dose decrease outside the target area; a multi-layer dose control structure for hierarchically controlling the dose distribution at different spatial distances; and a rejection region structure for controlling the extent to which the inhibitory amount spills out of the target area between the target area and the organs at risk.

[0148] Optionally, the generation unit includes: a first processing subunit, used to determine the volume of the target area based on its geometric features, and to determine the number of layers of the auxiliary structure based on the volume of the target area; a second processing subunit, used to determine the minimum distance between the target area and the organ at risk based on the spatial location information of the target area and the organ at risk, and to determine the type combination of the auxiliary structure based on the minimum distance; a third processing subunit, used to determine the thickness parameter of the auxiliary structure based on the current planning optimization stage, wherein the thickness parameter is used to constrain the distance range of the auxiliary structure extending outward from the outer surface of the target area; and a fourth processing subunit, used to generate an initial auxiliary structure based on the number of layers of the auxiliary structure, the type combination of the auxiliary structure, and the thickness parameter of the auxiliary structure.

[0149] Optionally, the first processing subunit includes: a first processing module, configured to determine the number of auxiliary structures as a first range of layers when the volume of the detected target region is less than a first volume threshold; a second processing module, configured to determine the number of auxiliary structures as a second range of layers when the volume of the detected target region is greater than or equal to the first volume threshold and less than a second volume threshold; and a third processing module, configured to determine the number of auxiliary structures as a third range of layers when the volume of the detected target region is greater than or equal to the second volume threshold; wherein the maximum number of layers in the first range of layers is less than the minimum number of layers in the second range of layers, and the maximum number of layers in the second range of layers is less than the minimum number of layers in the third range of layers.

[0150] Optionally, the second processing subunit includes: a fourth processing module, configured to generate at least a repulsion region structure as an auxiliary structure when the minimum distance detected is less than a first distance threshold; a fifth processing module, configured to generate at least a gradient control structure and a multi-layer dose control structure to form an auxiliary structure when the minimum distance detected is greater than or equal to the first distance threshold and less than a second distance threshold; and a sixth processing module, configured to generate a gradient control structure to form an auxiliary structure when the minimum distance detected is greater than or equal to the second distance threshold.

[0151] Optionally, the associated configuration unit includes: a fifth processing subunit, configured to assign a dose gradient constraint target to each structure when at least one of a gradient control structure and a multi-layer dose control structure is detected in the auxiliary structure, wherein the dose gradient constraint target is used to limit the rate at which the dose outside the target area decreases with increasing distance; a sixth processing subunit, configured to assign a maximum dose constraint target to the repulsion region structure when a repulsion region structure is detected in the auxiliary structure, wherein the maximum dose constraint target is used to limit the highest dose value within the repulsion region structure; and a seventh processing subunit, configured to associate at least one of the dose gradient constraint target and the maximum dose constraint target with the inverse optimization target, so that the constraint strength of the auxiliary structure is adjusted as the planning optimization stage changes.

[0152] Optionally, the inverse optimization objective represents the dose index that needs to be optimized during the inverse optimization of the radiotherapy plan, including at least one of target dose coverage, organ at risk dose limitation threshold, and extra-target dose gradient threshold; the weights of the optimization objective function are relative importance coefficients assigned to each optimization objective in multi-objective inverse optimization, used to balance the priorities among different optimization objectives.

[0153] Optionally, the first processing unit includes: an eighth processing subunit, used to monitor at least the following optimization status indicators during the reverse optimization process: the change in target area coverage, the change in volume of organs at risk exceeding a preset dose threshold, the rate of decrease in the function value of the optimization objective function as the number of iterations increases, and the change in the uniformity of dose distribution; a ninth processing subunit, used to compare the monitored optimization status indicators with preset stage switching conditions; a tenth processing subunit, used to determine that the planned optimization stage has changed when the optimization status indicators meet the stage switching conditions; and an eleventh processing subunit, used to determine that the planned optimization stage has not changed when the optimization status indicators do not meet the stage switching conditions.

[0154] Optionally, the optimization phase includes an initial coverage phase, a transitional optimization phase, and a convergence phase. The criteria for the initial coverage phase include: the change in target coverage is less than a first preset threshold, and the rate of decrease in the function value reduction rate is greater than a first rate threshold. The criteria for the transitional optimization phase include: the change in target coverage is greater than or equal to a first preset threshold, and the change in the volume of organs at risk exceeding a preset dose threshold is greater than a second preset threshold. The criteria for the convergence phase include: the rate of decrease in the function value is less than a second rate threshold, and the change in the uniformity index of the dose distribution is less than a third preset threshold, wherein the second rate threshold is less than the first rate threshold.

[0155] Optionally, the reconstruction operation includes: in the initial coverage phase, reducing the weights corresponding to the dose constraint targets of the gradient control structure when a gradient control structure exists; reducing the weights corresponding to the dose constraint targets of the multi-layer dose control structure and reducing the number of layers of the multi-layer dose control structure when a multi-layer dose control structure exists; reducing the weights corresponding to the dose constraint targets of the exclusion region structure when an exclusion region structure exists; in the transition optimization phase, increasing the number of layers of the multi-layer dose control structure when a multi-layer dose control structure exists; introducing an exclusion region structure when no exclusion region structure exists; increasing the gradient constraint strength of the gradient control structure when a gradient control structure exists; and in the convergence phase, adding outer gradient control structures, reducing the thickness of the multi-layer dose control structure when a multi-layer dose control structure exists, and increasing the weights of dose constraint targets related to organs at risk.

[0156] Optionally, the reconstruction operation of the auxiliary structure includes at least one of the following: adding or reducing gradient control structures; adding or reducing the number of layers of multi-layer dose control structures; adjusting the expansion distance, thickness parameter, or spatial position of the auxiliary structure, wherein the expansion distance is the distance the auxiliary structure extends outward from the outer surface of the target area; replacing some gradient control structures with multi-layer dose control structures, or adding a repulsion region structure; and reallocating the dose constraint targets corresponding to each type of structure in the auxiliary structure and the weight parameters of each dose constraint target.

[0157] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, which stores a computer program, wherein when the computer program is executed, the device on which the computer-readable storage medium is located performs the above-described auxiliary structure generation method for a helical tomotherapy system.

[0158] According to another aspect of the embodiments of this application, an electronic device is also provided, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by one or more processors, the one or more processors cause the one or more processors to perform the above-described auxiliary structure generation method for a helical tomotherapy system.

[0159] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program or instructions that, when executed by a processor, implement the above-described auxiliary structure generation method for a helical tomotherapy system.

[0160] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0161] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0162] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0163] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0164] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0165] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it 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 all or part 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 a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0166] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for generating auxiliary structures for a helical tomotherapy system, characterized in that, include: Based on the geometric features of the target area and the spatial relationship between the target area and the organs at risk, an initial auxiliary structure is generated, wherein the auxiliary structure is a spatial region used to constrain the dose distribution during the radiotherapy planning optimization process; The auxiliary structure is associated with the inverse optimization objective. This association configuration links the parameter information of the auxiliary structure with the weights of the optimization objective function, which quantifies the degree to which the radiotherapy plan meets clinical requirements. The parameter information of the auxiliary structure includes its number of layers, thickness, expansion distance, and spatial location. The weights of the optimization objective function refer to the relative importance coefficients assigned to each optimization objective in the inverse optimization of multiple objectives, used to balance the priorities among different optimization objectives. During the reverse optimization process, optimization status indicators are collected, and it is determined whether the planned optimization phase has switched based on the optimization status indicators. If a switch to the planning optimization phase is detected, a reconstruction operation of the auxiliary structure is triggered.

2. The method according to claim 1, characterized in that, The auxiliary structure includes at least one of the following: Gradient control structures are used to constrain the rate of dose descent outside the target area; A multi-layer dose control structure is used to control dose distribution in layers at different spatial distances; Rejection region structure, used to define the extent to which inhibitory amounts spill out of the target region between the target region and the organ at risk.

3. The method according to claim 2, characterized in that, Based on the geometric features of the target area and the spatial relationship between the target area and the organ at risk, an initial auxiliary structure is generated, including: The volume of the target region is determined based on its geometric features, and the number of layers of the auxiliary structure is determined based on the volume of the target region. The minimum distance between the target area and the organ at risk is determined based on the spatial location information of the target area and the organ at risk, and the type combination of the auxiliary structures is determined based on the minimum distance; The thickness parameters of the auxiliary structure are determined according to the current planning optimization phase, wherein the thickness parameters are used to constrain the distance range of the auxiliary structure extending outward from the outer surface of the target area; The initial auxiliary structure is generated based on the number of layers of the auxiliary structure, the combination of types of the auxiliary structure, and the thickness parameter of the auxiliary structure.

4. The method according to claim 3, characterized in that, The number of layers of the auxiliary structure is determined based on the volume of the target region, including: When the volume of the target region is detected to be less than a first volume threshold, the number of layers of the auxiliary structure is determined to be within a first layer range; When the volume of the target region is detected to be greater than or equal to the first volume threshold and less than the second volume threshold, the number of layers of the auxiliary structure is determined to be the second layer range; When the volume of the target region is detected to be greater than or equal to the second volume threshold, the number of layers of the auxiliary structure is determined to be a third layer range; Wherein, the maximum number of layers in the first range is less than the minimum number of layers in the second range, and the maximum number of layers in the second range is less than the minimum number of layers in the third range.

5. The method according to claim 3, characterized in that, The combination of accessory structures is determined based on the minimum distance between the target area and the organ at risk, including: If the minimum distance is detected to be less than the first distance threshold, at least the repulsion region structure is generated as the auxiliary structure; If the minimum distance is detected to be greater than or equal to the first distance threshold and less than the second distance threshold, at least the gradient control structure and the multi-layer dose control structure are generated to form the auxiliary structure. If the minimum distance is detected to be greater than or equal to the second distance threshold, the gradient control structure is generated to form the auxiliary structure.

6. The method according to claim 2, characterized in that, Associating the auxiliary structure with the inverse optimization objective includes: If at least one of the gradient control structure and the multilayer dose control structure is detected in the auxiliary structure, a dose gradient constraint target is assigned to each structure, wherein the dose gradient constraint target is used to limit the rate at which the dose outside the target area decreases with increasing distance; If the auxiliary structure is found to include the repulsion region structure, a maximum dose constraint target is assigned to the repulsion region structure, wherein the maximum dose constraint target is used to limit the highest dose value within the repulsion region structure; At least one of the dose gradient constraint target and the maximum dose constraint target is associated with the inverse optimization target, so that the constraint strength of the auxiliary structure is adjusted as the planning optimization stage changes.

7. The method according to claim 1, characterized in that, The reverse optimization objective represents the dose indicators that need to be optimized and achieved during the reverse optimization of the radiotherapy plan, including at least one of target dose coverage, organ at risk dose limitation threshold, and extra-target dose gradient threshold; the weight of the optimization objective function is a relative importance coefficient assigned to each optimization objective in multi-objective reverse optimization, used to balance the priority between different optimization objectives.

8. The method according to claim 1, characterized in that, During the reverse optimization process, optimization status indicators are collected, and the switching of the planned optimization phase is determined based on these indicators, including: During the reverse optimization process, at least the following optimization status indicators shall be monitored: The changes in target coverage, the changes in volume of organs at risk exceeding the preset dose threshold, the rate at which the function value of the optimized objective function decreases with increasing iterations, and the changes in the uniformity index of dose distribution. The monitored optimization status indicators are compared with preset stage switching conditions; If the optimization status indicator is found to meet the stage switching condition, it is determined that the planned optimization stage has changed; If the optimization status indicator is found to be unsatisfactory under the stage switching conditions, it is determined that the planned optimization stage has not changed.

9. The method according to claim 8, characterized in that, The planning optimization phase includes an initial coverage phase, a transition optimization phase, and a convergence phase. The determination criteria for the initial coverage phase include: the change in the target area coverage rate is less than a first preset threshold, and the decrease rate of the function value shrinkage rate is greater than a first rate threshold. The criteria for determining the transition optimization phase include: the change in target coverage is greater than or equal to the first preset threshold, and the change in volume of the organ at risk exceeding the preset dose threshold is greater than the second preset threshold. The criteria for determining the convergence phase include: the rate of decrease of the function value is less than the second rate threshold, and the change in the uniformity index of the dose distribution is less than the third preset threshold, wherein the second rate threshold is less than the first rate threshold.

10. The method according to claim 9, characterized in that, The reconstruction operation includes: In the initial coverage phase, when a gradient control structure exists, the weight corresponding to the dose constraint target of the gradient control structure is reduced; when a multi-layer dose control structure exists, the weight corresponding to the dose constraint target of the multi-layer dose control structure is reduced, and the number of layers of the multi-layer dose control structure is decreased; when a repulsion region structure exists, the weight corresponding to the dose constraint target of the repulsion region structure is reduced. During the transition optimization phase, the number of layers in the multi-layer dose control structure is increased when the multi-layer dose control structure exists; the repulsion region structure is introduced when the repulsion region structure does not exist; and the gradient constraint strength of the gradient control structure is increased when the gradient control structure exists. During the convergence phase, an outer gradient control structure is added. When the multi-layer dose control structure exists, the thickness of the multi-layer dose control structure is reduced, thereby increasing the weight of dose constraint targets related to organs at risk.

11. The method according to claim 2, characterized in that, The reconstruction operation of the auxiliary structure includes at least one of the following: Increase or decrease the gradient control structure; Increase or decrease the number of layers in the multilayer dose control structure; Adjust the expansion distance, thickness parameter, or spatial position of the auxiliary structure, wherein the expansion distance is the distance the auxiliary structure extends outward from the outer surface of the target area; Replace some gradient control structures with multilayer dose control structures, or add exclusion region structures; The dose constraint targets corresponding to each type of structure in the auxiliary structure and the weight parameters of each dose constraint target are reallocated.

12. An auxiliary structure generation device for a helical tomotherapy system, characterized in that, include: A generation unit is used to generate an initial auxiliary structure based on the geometric features of the target area and the spatial relationship between the target area and the organs at risk, wherein the auxiliary structure is a spatial region used to constrain the dose distribution during the radiotherapy planning optimization process; The association configuration unit is used to associate the auxiliary structure with the inverse optimization objective. The association configuration establishes a linkage between the parameter information of the auxiliary structure and the weights of the optimization objective function. The optimization objective function quantifies the degree to which the radiotherapy plan meets clinical requirements. The parameter information of the auxiliary structure includes the number of layers, thickness, expansion distance, and spatial location of the auxiliary structure. The weights of the optimization objective function refer to the relative importance coefficients assigned to each optimization objective in the inverse optimization of multiple optimization objectives, used to balance the priorities among different optimization objectives. The first processing unit is used to collect optimization status indicators during the reverse optimization process and determine whether the planned optimization phase has switched based on the optimization status indicators. The second processing unit is used to trigger a reconstruction operation of the auxiliary structure when a switch in the planning optimization phase is detected.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the computer program is executed, the device containing the computer-readable storage medium performs the auxiliary structure generation method for a helical tomotherapy system as described in any one of claims 1 to 11.

14. An electronic device, characterized in that, It includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to perform the auxiliary structure generation method for a helical tomotherapy system according to any one of claims 1 to 11.

15. A computer program product, characterized in that, Includes a computer program or instructions that, when executed by a processor, implement the auxiliary structure generation method for a helical tomotherapy system as described in any one of claims 1 to 11.

Images