Tomo radiotherapy plan optimization method and system based on quality assurance reverse influence

Abstract

The application discloses a Tomo radiotherapy plan optimization method and system based on quality assurance reverse influence, relates to the medical technology field, and particularly relates to the Tomo radiotherapy information processing field. The method comprises the following steps: generating a corresponding quality assurance plan based on an initial radiotherapy plan; performing quality assurance verification on the initial radiotherapy plan according to the quality assurance plan; in the case that it is detected that the quality assurance verification fails, performing at least one iteration update on the initial radiotherapy plan until a preset iteration termination condition is met, and outputting a radiotherapy plan obtained through the last iteration update as a target radiotherapy plan. The application solves the technical problem of low Tomo radiotherapy plan optimization efficiency caused by the fact that the iteration optimization of the Tomo radiotherapy plan mainly depends on the subjective experience of a physicist in the prior art.

Description

Technical Field

[0001] This application relates to the field of medical technology, and more particularly to the field of Tomo radiotherapy information processing. Specifically, it relates to a method and system for optimizing Tomo radiotherapy plans based on the reverse influence of quality assurance. Background Technology

[0002] In the field of tomotherapy, treatment planning optimization typically requires multiple rounds of iterative adjustments. In clinical practice, physicists first design an initial plan based on the prescribed dosage. After dosage calculation and quality assurance verification, if the verification results are unsatisfactory, the cause of the problem needs to be analyzed and the optimization parameters manually adjusted. Then, the optimization and verification are repeated. This process is repeated until the plan meets clinical requirements.

[0003] However, the aforementioned iterative optimization process is highly dependent on the individual experience of physicists. Different physicists may have different judgment criteria and approaches to the same problem, making it difficult to guarantee the consistency of plan quality. Furthermore, each iteration requires manual intervention to analyze and verify the results, and to determine the direction and extent of adjustments, making the entire process time-consuming. Especially for complex cases, multiple iterations are often needed to achieve satisfactory results, significantly impacting clinical work efficiency.

[0004] Therefore, existing technologies suffer from low optimization efficiency of Tomo radiotherapy plans because the iterative optimization of Tomo radiotherapy plans mainly relies on the subjective experience of physicists. Summary of the Invention

[0005] This application provides a method and system for optimizing Tomo radiotherapy plans based on the reverse influence of quality assurance, in order to at least solve the technical problem of low optimization efficiency of Tomo radiotherapy plans in the prior art, which is caused by the fact that the iterative optimization of Tomo radiotherapy plans mainly relies on the subjective experience of physicists.

[0006] According to one aspect of this application, a Tomo radiotherapy plan optimization method based on quality assurance reverse influence is provided, comprising: generating a corresponding quality assurance plan based on an initial radiotherapy plan; performing quality assurance verification on the initial radiotherapy plan according to the quality assurance plan; if the quality assurance verification fails, performing at least one iterative update on the initial radiotherapy plan until a preset iteration termination condition is met, and outputting the radiotherapy plan obtained from the last iterative update as the target radiotherapy plan; wherein each iterative update includes: generating an optimization parameter adjustment scheme based on the quality assurance verification result of the radiotherapy plan, modifying the optimization parameters of the radiotherapy plan according to the optimization parameter adjustment scheme and triggering re-optimization to obtain an updated radiotherapy plan, and generating a corresponding quality assurance plan based on the updated radiotherapy plan and performing quality assurance verification on the updated radiotherapy plan; The process involves generating optimized parameter adjustment schemes based on the quality assurance verification results of the radiotherapy plan. This includes: in response to the failure of the radiotherapy plan's quality assurance verification, obtaining multi-dimensional quality indicators for the verification; analyzing these multi-dimensional quality indicators based on problem diagnosis rules to identify the problem types and corresponding problem areas that led to the verification failure; and generating optimized parameter adjustment schemes for the radiotherapy plan based on a quantitative parameter mapping rule base, according to the problem types and problem areas. The multi-dimensional quality indicators include at least global gamma pass rate and regional gamma pass rate. Global gamma pass rate characterization: the gamma analysis pass rate between the dose distribution in the quality assurance plan and the first dose distribution, which is the dose distribution recalculated by the independent dose validation system for the quality assurance plan; Regional gamma pass rate characterization: Based on at least one preset proportion of the prescription dose, the quality assurance plan's gamma analysis pass rate between the dose distribution and the second dose distribution for the dose region; the second dose distribution is the dose distribution recalculated for the dose region by the independent dose validation system based on the quality assurance plan.

[0007] Optionally, each iteration update includes the following steps: adjusting the optimization parameters of the radiotherapy plan according to the optimization parameter adjustment scheme, and re-optimizing the modified radiotherapy plan to generate an updated radiotherapy plan; after generating the corresponding quality assurance plan based on the updated radiotherapy plan, performing quality assurance verification on the updated radiotherapy plan, and determining whether the verification result meets the preset iteration termination condition.

[0008] Optionally, based on the initial radiotherapy plan, a corresponding quality assurance plan is generated, including: loading a preset quality assurance phantom, wherein the quality assurance phantom is a phantom used for dose verification; acquiring image data and region of interest data of the quality assurance phantom; generating the quality assurance plan based on the beam geometry parameters, physical parameters of helical tomotherapy, dose information, fractionation information in the initial radiotherapy plan, and the image data and region of interest data of the quality assurance phantom, wherein the physical parameters of helical tomotherapy include at least one of beam width, pitch factor, modulation factor, collimator mode, and control point sequence.

[0009] Optionally, after generating a corresponding quality assurance plan based on the initial radiotherapy plan, a unique identifier for identifying the quality assurance plan is generated based on the initial radiotherapy plan; a bidirectional association storage is established between the quality assurance plan and the initial radiotherapy plan based on the unique identifier; wherein, the bidirectional association storage includes: storing the plan identifier of the initial radiotherapy plan in the quality assurance plan, and storing an information list of the quality assurance plan in the initial radiotherapy plan, the information list including the name of the quality assurance plan, the unique identifier of the quality assurance plan, the file path and the creation time.

[0010] Optionally, the multi-dimensional quality indicators also include at least one of the following indicators: the three-dimensional spatial coordinate distribution of gamma failure points, wherein gamma failure points represent: voxel points where the gamma analysis value is greater than the target threshold between the dose distribution of the quality assurance plan and the first dose distribution; the feature point deviation of the dose-volume histogram of the target area, including: the dose deviation corresponding to the first preset volume ratio and the maximum dose deviation; the feature point deviation of the dose-volume histogram of organs at risk; dose hotspot locations and dose coldspot locations, wherein the dose hotspot location represents the spatial coordinate region where the actual dose is higher than the planned dose and the deviation exceeds the first preset threshold, and the dose coldspot location represents the spatial coordinate region where the actual dose is lower than the planned dose and the deviation exceeds the second preset threshold.

[0011] Optionally, the problem type includes one or more of the following: a first problem type, used to characterize that the dose corresponding to a first preset volume percentage of the target area is lower than a preset percentage threshold of the prescription dose, or that the dose corresponding to a first preset volume percentage of the target area in the quality assurance plan is lower than the dose corresponding to a first preset volume percentage of the target area in the radiotherapy plan and exceeds a preset difference threshold; a second problem type, used to characterize that the dose or maximum dose corresponding to a second preset volume percentage of the target area exceeds a preset percentage threshold of the prescription dose, or that the maximum dose of the target area in the quality assurance plan is higher than the maximum dose of the target area in the radiotherapy plan and exceeds a preset difference threshold; a third problem type, used to characterize that the dose index of the organ at risk exceeds the clinical limit, or that the difference between the quality assurance plan and the radiotherapy plan exceeds a preset difference threshold; a fourth problem type, used to characterize that the concentration of gamma failure points exceeds a preset spatial distribution threshold; the problem area includes the target area, the organ at risk, or the concentrated area of ​​gamma failure points, wherein the concentrated area of ​​gamma failure points is the area where the density of gamma failure points is greater than a preset density.

[0012] Optionally, the quantization parameter mapping rule base includes one or more of the following rules: A first rule includes: increasing the lower constraint weight of the target region when a first problem type is detected, and simultaneously decreasing the constraint weight of adjacent organs at risk when the constraint weight of adjacent organs at risk exceeds a preset ratio; A second rule includes: increasing the upper constraint weight of the target region or decreasing the modulation factor when a second problem type is detected; A third rule includes: increasing the constraint weight of the organ at risk corresponding to the third problem type or increasing the organ protection level of that organ when a third problem type is detected; A fourth rule includes: adjusting the constraint weight of the concentrated region of gamma failure points when a fourth problem type is detected; A fifth rule includes: simultaneously increasing the constraint weight of the target region and the constraint weight of organs at risk when the global gamma pass rate is detected to be lower than a preset threshold.

[0013] Optionally, after generating an optimized parameter adjustment scheme for the radiotherapy plan based on the quantitative parameter mapping rule base and according to the problem type and problem area, the system checks whether the constraint weights of the updated radiotherapy plan are within the preset range of minimum and maximum weights; whether the planned dose of the updated radiotherapy plan is within the preset lower limit and preset upper limit of the prescription dose; and whether the beam parameters of the updated radiotherapy plan are within the limits of the machine's physical parameters.

[0014] Optionally, when triggering a re-optimization of the modified radiotherapy plan, if it is detected that the optimization parameter adjustment scheme involves updating the constraint weights in the radiotherapy plan, the task identifier and dose influence matrix of the previous optimization are reused, and the flux map optimization update operation is re-executed; if it is detected that the optimization parameter adjustment scheme involves updating the beam parameters in the radiotherapy plan, the dose influence matrix is ​​recalculated and the flux map optimization update operation is executed.

[0015] Optionally, the iteration termination condition includes any of the following conditions: the first condition includes: the global gamma pass rate of quality assurance verification is greater than or equal to a preset pass rate threshold; the second condition includes: the feature point deviation of the dose-volume histogram of the target area and the organ at risk is less than a preset percentage threshold; the third condition includes: the parameter adjustment range of two consecutive iterations is less than a preset range threshold; the fourth condition includes: the preset maximum number of iterations has been reached.

[0016] Optionally, if the decrease in global gamma pass rate exceeds a preset threshold after two consecutive iterations, the system automatically rolls back to the historical parameter configuration with the highest global gamma pass rate and generates a prompt message.

[0017] According to another aspect of the embodiments of this application, a Tomo radiotherapy plan optimization system based on quality assurance reverse influence is also provided, comprising: a quality assurance plan automatic generation unit, used to generate a corresponding quality assurance plan based on an initial radiotherapy plan; a quality assurance verification unit, used to perform quality assurance verification on the initial radiotherapy plan according to the quality assurance plan; and an iterative control unit, used to perform at least one iterative update on the initial radiotherapy plan when the quality assurance verification fails, until a preset iteration termination condition is met, and output the radiotherapy plan obtained from the last iterative update as the target radiotherapy plan; wherein each iterative update includes: generating an optimization parameter adjustment scheme based on the quality assurance verification result of the radiotherapy plan, modifying the optimization parameters of the radiotherapy plan according to the optimization parameter adjustment scheme and triggering re-optimization to obtain an updated radiotherapy plan, and generating a corresponding quality assurance plan based on the updated radiotherapy plan and performing quality assurance verification on the updated radiotherapy plan; wherein, the optimization parameters are generated based on the quality assurance verification result of the radiotherapy plan. The adjustment plan includes: in response to the failure of radiotherapy plan quality assurance verification, obtaining multi-dimensional quality indicators for quality assurance verification; analyzing the multi-dimensional quality indicators based on problem diagnosis rules to obtain the problem types and corresponding problem areas that led to the failure of quality assurance verification; and generating an optimized parameter adjustment plan for the radiotherapy plan based on the problem type and problem area, using a quantitative parameter mapping rule base. The multi-dimensional quality indicators include at least global gamma pass rate and regional gamma pass rate. The global gamma pass rate is characterized by the gamma analysis pass rate between the dose distribution in the quality assurance plan and the first dose distribution, where the first dose distribution is the dose distribution recalculated by the independent dose verification system for the quality assurance plan. The regional gamma pass rate is characterized by the gamma analysis pass rate between the dose distribution of the quality assurance plan for the dose area divided based on at least one preset ratio of the prescription dose and the second dose distribution, where the second dose distribution is the dose distribution recalculated by the independent dose verification system for the dose area based on the quality assurance plan.

[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 Tomo radiotherapy planning optimization method based on quality assurance reverse influence.

[0019] According to another aspect 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 Tomo radiotherapy planning optimization method based on quality assurance reverse effects.

[0020] According to another aspect 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 Tomo radiotherapy planning optimization method based on quality assurance reverse effects.

[0021] As can be seen from the above, firstly, when the quality assurance verification fails, this application does not rely on manual intervention, but automatically performs at least one iteration update on the radiotherapy plan until the preset iteration termination condition is met. This step allows the entire optimization process to proceed automatically without manual intervention, thereby helping to eliminate reliance on the physicist's subjective experience and transforming repetitive manual operations into an automated process. Secondly, this application also defines the specific content of each iteration update: based on the quality assurance verification results of the radiotherapy plan, an optimization parameter adjustment scheme is generated; the optimization parameters of the radiotherapy plan are modified according to the scheme, triggering re-optimization to obtain an updated radiotherapy plan; and a corresponding quality assurance plan is generated based on the updated radiotherapy plan before quality assurance verification is performed. This step forms a complete closed loop of "verification failure, automatic generation of adjustment scheme, automatic parameter modification, automatic re-optimization, and automatic verification." Each iteration is driven by objective quality assurance verification results rather than subjective experience to adjust parameters, and the adjustment scheme has clear quantitative basis.

[0022] In summary, this application transforms the manual parameter tuning process, which originally relied on the subjective experience of physicists, into an automated and standardized iterative optimization process through an automatic iterative update mechanism and a closed-loop feedback driven by quality assurance verification results. This solves the technical problem of low efficiency in Tomo's radiotherapy planning optimization. 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 of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of an optional Tomo radiotherapy planning optimization method based on the reverse impact of quality assurance, according to an embodiment of this application; Figure 2 This is a flowchart of the Tomo radiotherapy planning optimization system according to an embodiment of this application. Detailed Implementation

[0024] 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.

[0025] 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.

[0026] It should also be noted that the information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, and displayed data) collected in this application 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 related data all comply with the relevant laws, regulations, and standards of the relevant regions, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding operation entry points are provided for users to choose to authorize or refuse. For example, this system has interfaces with relevant users or organizations. Before obtaining relevant information, a request to obtain the information needs to be sent to the aforementioned user or organization through the interface, and the relevant information is obtained only after receiving consent from the aforementioned user or organization.

[0027] According to the embodiments of this application, a Tomo radiotherapy planning optimization system based on quality assurance reverse influence (hereinafter referred to as the radiotherapy planning optimization system) can be used as the execution subject of the Tomo radiotherapy planning optimization method based on quality assurance reverse influence 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. Those skilled in the art should know that this application does not particularly limit the specific form of the execution subject.

[0028] According to an embodiment of this application, an embodiment of a Tomo radiotherapy planning optimization method based on quality assurance reverse influence 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.

[0029] Figure 1 This is a flowchart of an optional Tomo radiotherapy planning optimization method based on the reverse impact of quality assurance, according to an embodiment of this application. Figure 1 As shown, the method includes the following steps: Step S101: Generate a corresponding quality assurance plan based on the initial radiotherapy plan.

[0030] For example, the radiotherapy planning optimization system generates a corresponding quality assurance plan based on the initial radiotherapy plan. The quality assurance plan is a validation plan used to verify the accuracy of the treatment plan's dosage; its parameters are derived from the initial radiotherapy plan, but dosage calculations are performed on a standardized quality assurance phantom (rather than patient anatomy). By automatically generating the quality assurance plan, the radiotherapy planning optimization system ensures parameter consistency between the validation plan and the treatment plan, providing a standardized evaluation object for subsequent quality assurance validation.

[0031] In some embodiments, the radiotherapy planning optimization system loads a preset quality assurance phantom and acquires its image data and region of interest (ROI) data. Based on the beam geometry parameters in the initial radiotherapy plan, the physical parameters of helical tomotherapy (including at least one of beam width, pitch factor, modulation factor, collimator mode, and control point sequence), dose information, and fractionation information, the system combines the image data and ROI data from the quality assurance phantom to generate a corresponding quality assurance plan. The system automatically copies the complete parameters of the treatment plan into the quality assurance plan, eliminating the need for manual setting of Tomo-specific parameters and ensuring parameter consistency.

[0032] In other embodiments, after generating the quality assurance plan, the radiotherapy planning optimization system also establishes a bidirectional association between the quality assurance plan and the initial radiotherapy plan. The system generates a unique identifier for the quality assurance plan based on the initial radiotherapy plan, stores the plan identifier of the initial radiotherapy plan in the quality assurance plan, and simultaneously stores an information list of the quality assurance plan in the initial radiotherapy plan. This information list includes the name, unique identifier, file path, and creation time of the quality assurance plan. Through this bidirectional association, the system can quickly trace and locate the corresponding quality assurance plan in subsequent iterations, facilitating automated process management.

[0033] Step S102: Perform quality assurance verification on the initial radiotherapy plan in accordance with the quality assurance plan.

[0034] For example, the radiotherapy planning optimization system performs quality assurance verification on the initial radiotherapy plan based on the generated quality assurance plan. Quality assurance verification refers to the process of evaluating the accuracy of the treatment plan dosage by comparing and analyzing the dose distribution of the treatment plan on a standardized phantom with the calculation results of an independent verification system. Through quality assurance verification, the radiotherapy planning optimization system determines whether the current radiotherapy plan meets the accuracy requirements for clinical implementation, providing a basis for subsequent iterative optimization.

[0035] In some embodiments, the radiotherapy planning optimization system performs dose calculations on the quality assurance plan, generating three-dimensional dose distribution data on the quality assurance phantom. The system acquires the three-dimensional dose distribution recalculated by an independent dose verification system for the same quality assurance plan as a reference dose, performs three-dimensional gamma analysis on both, and calculates the global gamma pass rate. The system also performs dose-volume histogram comparison analysis, comparing the dose-volume histogram data of the quality assurance plan in the target area and organs at risk with the corresponding data of the initial radiotherapy plan point by point, calculating the differences in key dose indicators (such as maximum dose, average dose, and dose corresponding to a specific volume). The system integrates the gamma analysis results and the dose-volume histogram comparison results to generate the quality assurance verification results.

[0036] In some embodiments, the radiotherapy planning optimization system also performs a secondary independent dose verification during the quality assurance verification process. The system submits the digital image communication protocol file of the quality assurance plan to the independent dose verification system via an interface. This system recalculates the dose distribution on the quality assurance phantom using an independent dose calculation engine different from that of the treatment planning system. The radiotherapy planning optimization system receives the three-dimensional gamma analysis results returned by the independent dose verification system, including global gamma pass rate, regional gamma pass rate (divided into high-dose, medium-dose, and low-dose zones according to the prescribed dose ratio), and the three-dimensional spatial coordinate distribution of gamma failure points. The system uses these multi-dimensional verification results as input data for subsequent problem diagnosis, providing a quantitative basis for automated parameter adjustment.

[0037] Step S103: If the quality assurance verification fails, perform at least one iteration update on the initial radiotherapy plan until the preset iteration termination condition is met, and output the radiotherapy plan obtained from the last iteration update as the target radiotherapy plan.

[0038] Each iteration update is based on the radiotherapy plan obtained from the previous iteration update. Each iteration update includes: generating an optimization parameter adjustment scheme based on the quality assurance verification results of the radiotherapy plan; modifying the optimization parameters of the radiotherapy plan according to the optimization parameter adjustment scheme and triggering re-optimization to obtain the updated radiotherapy plan; and generating a corresponding quality assurance plan based on the updated radiotherapy plan and performing quality assurance verification on the updated radiotherapy plan.

[0039] For example, when the radiotherapy planning optimization system detects a failure in quality assurance verification, it performs at least one iterative update on the initial radiotherapy plan until a preset iteration termination condition is met, and outputs the radiotherapy plan obtained from the last iteration update as the target radiotherapy plan. Iterative update refers to a complete closed-loop process of "parameter adjustment - re-optimization - quality assurance verification." Each iteration is based on the quality assurance verification result of the radiotherapy plan obtained from the previous iteration update, driving the parameter adjustment for the next iteration update. The iteration termination condition is a preset stopping criterion to ensure optimization efficiency and plan quality, including at least one of passing quality assurance verification, meeting key performance indicators, or reaching the maximum number of iterations. The radiotherapy planning optimization system, through an automated iterative mechanism, replaces the inefficient process of repeated manual adjustments, achieving automated closed-loop optimization from the initial radiotherapy plan to the target radiotherapy plan.

[0040] In some embodiments, the radiotherapy planning optimization system performs multiple iterative updates to the radiotherapy plan. During the first iteration, the system generates an optimization parameter adjustment scheme based on the quality assurance verification results of the initial radiotherapy plan. After modifying the optimization parameters of the initial radiotherapy plan, it triggers re-optimization, resulting in the first-round updated radiotherapy plan. The system then generates a corresponding quality assurance plan based on the first-round updated radiotherapy plan and performs quality assurance verification to determine if the verification results meet the iteration termination conditions. For example, if the global gamma pass rate of the first-round updated radiotherapy plan reaches 95% or higher, the system determines that the iteration termination condition is met, stops the iteration, and outputs the first-round updated radiotherapy plan as the target radiotherapy plan. If the condition is not met, the system automatically enters the second iteration, repeating the above process. Based on the quality assurance verification results of the first-round updated radiotherapy plan, a new round of optimization parameter adjustment schemes is generated, resulting in the second-round updated radiotherapy plan, which is then verified again. This automated iterative mechanism requires no manual intervention; each iteration is driven by objective verification results.

[0041] In other embodiments, the radiotherapy planning optimization system monitors convergence and divergence risks during iterative updates. The system records the quality assurance verification results of each iteration. For example, if the global gamma pass rate decreases by more than a preset threshold (e.g., 5%) after two consecutive iterations, the system is determined to be in a divergent state, automatically rolls back to the historical parameter configuration with the highest global gamma pass rate, and generates a prompt message. The system also checks whether a preset maximum number of iterations (e.g., 5 or 10) has been reached. If so, it stops iterative updates and outputs the radiotherapy plan with the highest global gamma pass rate in the current iteration as the target radiotherapy plan. This iterative control mechanism, while ensuring optimization effectiveness, helps avoid invalid iterations and parameter divergence, improving the safety and controllability of the iterative process.

[0042] In some embodiments, each iteration update includes the following steps: In response to the failure of the radiotherapy plan's quality assurance verification, multi-dimensional quality indicators for quality assurance verification are obtained. Based on problem diagnosis rules, the multi-dimensional quality indicators are analyzed to identify the problem types and corresponding problem areas that led to the failure of quality assurance verification. Based on a quantitative parameter mapping rule base, an optimization parameter adjustment scheme for the radiotherapy plan is generated according to the problem type and problem area. According to the optimization parameter adjustment scheme, the optimization parameters of the radiotherapy plan are modified, and the modified radiotherapy plan is re-optimized to generate an updated radiotherapy plan. After generating the corresponding quality assurance plan based on the updated radiotherapy plan, quality assurance verification is performed on the updated radiotherapy plan, and it is determined whether the verification result meets the preset iteration termination conditions.

[0043] For example, each iteration update in the radiotherapy planning optimization system includes the following core steps: In response to a failed quality assurance verification, the system first obtains the multi-dimensional quality indicators of the verification; then, it analyzes the multi-dimensional quality indicators based on problem diagnosis rules to identify the problem types and corresponding problem areas that caused the verification failure; next, based on a quantitative parameter mapping rule base, it generates an optimization parameter adjustment plan according to the identified problem types and problem areas; subsequently, it modifies the optimization parameters of the radiotherapy plan according to the plan and triggers re-optimization to generate an updated radiotherapy plan; finally, it generates a quality assurance plan based on the updated radiotherapy plan and performs verification to determine whether the iteration termination conditions are met. This process transforms the traditional analysis and adjustment process, which relies on human experience, into an automated and standardized calculation process.

[0044] In some embodiments, the radiotherapy planning optimization system performs iterative updates to address the issue of insufficient target dose coverage. The system extracts multi-dimensional quality indicators from the quality assurance verification results, including target D95 (corresponding to the dose corresponding to the first preset volume ratio mentioned above) deviation and target average dose deviation. Based on problem diagnosis rules, the system identifies a target D95 below the prescribed dose and a deviation exceeding a preset threshold, diagnosing it as insufficient target coverage. The problem area is located as the target area and its adjacent organs at risk. The system calls the corresponding rules in the quantitative parameter mapping rule library to calculate the adjustment amount of the dose constraint weight under the target area, and simultaneously checks the constraint weights of adjacent organs at risk, adjusting them according to a preset ratio. After modifying the optimization parameters according to the adjustment plan, the system triggers re-optimization, generates an updated radiotherapy plan, and proceeds to the next round of verification.

[0045] In other embodiments, the radiotherapy planning optimization system performs iterative updates to address the issue of excessive doses to organs at risk. The system extracts key indicator deviations for organs at risk from the quality assurance verification results, such as excessive maximum dose to the spinal cord or excessive V20 dose to the lungs. Based on problem diagnosis rules, the system identifies an organ at risk whose dose indicator exceeds clinical limits, and the difference between the quality assurance plan and the radiotherapy plan exceeds a preset threshold, diagnosing it as an excessive dose to an organ at risk, and locating the problem area as that organ. The system calls the corresponding rules in the quantification parameter mapping rule base to calculate the increase in the constraint weight for that organ at risk, and simultaneously increases the protection level if organ protection settings are available for that organ. After modifying the optimization parameters according to the adjustment plan, the system triggers re-optimization, generates an updated radiotherapy plan, and proceeds to the next round of verification. Through this automated iterative update, the system can progressively approximate a radiotherapy plan that meets clinical requirements without manual intervention.

[0046] In some embodiments, generating a corresponding quality assurance plan based on an initial radiotherapy plan includes: loading a preset quality assurance phantom, wherein the quality assurance phantom is a phantom used for dose verification; acquiring image data and region of interest data of the quality assurance phantom; and generating a quality assurance plan based on the beam geometry parameters, physical parameters of helical tomotherapy, dose information, fractionation information, and image data and region of interest data of the quality assurance phantom in the initial radiotherapy plan, wherein the physical parameters of helical tomotherapy include at least one of beam width, pitch factor, modulation factor, collimator mode, and control point sequence.

[0047] For example, the radiotherapy planning optimization system automatically generates a corresponding quality assurance plan based on the initial radiotherapy plan. The quality assurance phantom is a standard phantom used for dose verification, such as the Cheese Phantom cylindrical phantom, which has a known geometry and known radiation attenuation characteristics, and is used to replace the patient's anatomy for dose measurement and verification. Region of interest data refers to the contour data of the virtual target area and virtual organs at risk pre-set on the quality assurance phantom. The radiotherapy planning optimization system loads the pre-set quality assurance phantom, maps the treatment parameters from the initial radiotherapy plan onto the phantom, and generates a quality assurance plan that can be used for dose verification.

[0048] In some embodiments, the radiotherapy planning optimization system automatically generates a quality assurance plan for Tomo Helical (spiral mode) radiotherapy plans. The system loads a preset Cheese Phantom cylindrical phantom, acquires its CT image data and region of interest data, including the virtual target contours and virtual sensor positions at different depths within the phantom. The system extracts beam geometry parameters (gantry angle, collimator angle, etc.), Tomo Helical mode-specific physical parameters (beam width, pitch factor, modulation factor, Jaw mode, control point sequence), dose information, and fractionation information from the initial radiotherapy plan. The system completely copies and applies these parameters to the quality assurance plan, automatically creating the plan without requiring manual setting of Tomo-specific parameters, thus avoiding the risk of parameter inconsistencies caused by manual operation.

[0049] In other embodiments, the radiotherapy planning optimization system can also automatically generate quality assurance plans for Tomo Direct mode (fixed-angle mode) radiotherapy plans. After loading a preset quality assurance phantom, the system extracts Tomo Direct mode-specific physical parameters from the initial radiotherapy plan, including the gantry angle list for the fixed-angle beam, the beam width for each angle, and the modulation factor. The system automatically converts the control point sequence in the treatment plan into the corresponding beam control points in the quality assurance plan, ensuring the geometric accuracy of beam irradiation. After generating the quality assurance plan, the system also automatically performs parameter validity checks, checking whether the beam width is within the machine's supported range (10mm / 25mm / 50mm), whether the pitch factor is within the range of 0.10 to 0.50, and whether the modulation factor is within the range of 1.0 to 999.99. The system automatically prompts for parameter anomalies based on the verification results, ensuring that the generated quality assurance plan conforms to the machine's physical constraints.

[0050] In some embodiments, after generating a corresponding quality assurance plan based on the initial radiotherapy plan, a unique identifier for identifying the quality assurance plan is generated based on the initial radiotherapy plan; a bidirectional association storage between the quality assurance plan and the initial radiotherapy plan is established based on the unique identifier; wherein, the bidirectional association storage includes: storing the plan identifier of the initial radiotherapy plan in the quality assurance plan, and storing an information list of the quality assurance plan in the initial radiotherapy plan, the information list including the name of the quality assurance plan, the unique identifier of the quality assurance plan, the file path and the creation time.

[0051] For example, after generating a corresponding quality assurance plan based on the initial radiotherapy plan, the radiotherapy planning optimization system establishes a bidirectional association storage mechanism between the two. A unique identifier is a globally unique identifier generated in the system for each quality assurance plan, used to distinguish quality assurance plans created at different times and for different plan versions. Bidirectional association storage means that the quality assurance plan and the initial radiotherapy plan mutually record each other's key information, forming a traceable relationship. Through bidirectional association storage, the radiotherapy planning optimization system can quickly locate and backtrack the corresponding quality assurance plan during subsequent iterations, facilitating automated process management and version traceability.

[0052] In some embodiments, the radiotherapy planning optimization system creates a unique identifier for the first generated quality assurance plan and adds a Referenced Treat Plan Sop Ins UID field to the quality assurance plan's data record to store the plan identifier of the initial radiotherapy plan. Simultaneously, the system maintains a QA (Quality Assurance) Plan Info List in the initial radiotherapy plan's data record. This list contains information such as the name, unique identifier, file path, and creation time of the generated quality assurance plan. When the system iterates and updates the same initial radiotherapy plan multiple times, each generated quality assurance plan is added to this information list, forming a historical record sorted by creation time. This bidirectional association storage facilitates quick retrieval of all associated quality assurance plans from an initial radiotherapy plan and also allows for the reverse tracing of the source initial radiotherapy plan from a quality assurance plan.

[0053] In other embodiments, the radiotherapy planning optimization system utilizes bidirectional associative storage to achieve automatic traceability of the iteration process. When the system generates an updated radiotherapy plan after re-optimization, it generates a corresponding quality assurance plan based on the updated plan and establishes a bidirectional associative storage between the updated plan and the new quality assurance plan. The system retains a reference chain to the original initial radiotherapy plan in the updated plan, making the plan version relationship throughout the iteration process a traceable tree structure. For example, when the system needs to roll back to a historical version, it can quickly locate the plan data of the target version by traversing the plan identifier and creation time in the bidirectional associative storage. This mechanism supports version management, quality traceability, and anomaly recovery during the iteration process, providing a data foundation for automated closed-loop optimization.

[0054] In some embodiments, the multidimensional quality metrics include at least two of the following: Global gamma pass rate, which represents the gamma analysis pass rate between the dose distribution in the quality assurance plan and the first dose distribution, where the first dose distribution is the dose distribution recalculated by the independent dose validation system for the quality assurance plan; Regional gamma pass rate, wherein the regional gamma pass rate characterizes: the gamma analysis pass rate between the dose distribution and the second dose distribution of the quality assurance plan for the dose region divided based on at least one preset ratio of the prescription dose; the second dose distribution is the dose distribution recalculated for the dose region by the independent dose validation system based on the quality assurance plan; The three-dimensional spatial coordinate distribution of gamma failure points, wherein gamma failure points are characterized as voxel points where the gamma analysis value is greater than the target threshold between the dose distribution of the quality assurance plan and the first dose distribution; The feature point deviations of the dose-volume histogram of the target area include: the dose deviation corresponding to the first preset volume ratio and the maximum dose deviation; Characteristic point bias in the dose-volume histogram of organs at risk; Dosage hotspot locations and dosage coldspot locations, wherein a dosage hotspot location represents a spatial coordinate region where the actual dose is higher than the planned dose and the deviation exceeds a first preset threshold, and a dosage coldspot location represents a spatial coordinate region where the actual dose is lower than the planned dose and the deviation exceeds a second preset threshold.

[0055] For example, the radiotherapy planning optimization system extracts multi-dimensional quality indicators from the quality assurance verification results to comprehensively evaluate the dose accuracy of the treatment plan. Global gamma pass rate is a comprehensive indicator measuring the consistency of the entire dose distribution. By comparing the dose distribution of the quality assurance plan on the phantom with the dose distribution recalculated by the independent dose verification system, the percentage of voxels with gamma analysis values ​​less than 1 can be calculated. Regional gamma pass rate is calculated by dividing the dose space into high-dose, medium-dose, and low-dose zones according to the prescribed dose ratio, and then calculating the gamma pass rate for each zone to identify the specific areas where dose deviations occur. The three-dimensional spatial coordinate distribution of gamma failure points records the voxel locations where gamma analysis values ​​are greater than the target threshold, helping to locate the spatial distribution characteristics of dose deviations.

[0056] In some embodiments, the radiotherapy planning optimization system extracts multi-dimensional quality indicators related to gamma analysis. The system obtains the global gamma pass rate (e.g., 95.6%) returned by an independent dose verification system, and simultaneously obtains the regional gamma pass rates for three dose regions divided according to the prescribed dose ratio (high-dose region greater than 80% of the prescribed dose, medium-dose region 30% to 80% of the prescribed dose, and low-dose region less than 30% of the prescribed dose). The system also extracts the three-dimensional spatial coordinates of all gamma failure points, forming a failure point cloud data. By comprehensively analyzing these indicators, the system can determine whether the dose deviation is global or local, and whether the deviation mainly occurs in the high-dose region or the low-dose region, providing precise spatial location information for subsequent problem diagnosis.

[0057] In other embodiments, the radiotherapy planning optimization system can also extract multi-dimensional quality indicators and dose spatial distribution characteristic indicators related to the dose-volume histogram. For example, the system acquires the D95 (dose corresponding to 95% volume) deviation and Dmax (maximum dose) deviation of the target area, as well as the deviations of key indicators of organs at risk (such as spinal cord Dmax, lung V20, etc.). The system also identifies spatial coordinate regions where the actual dose is higher than the planned dose and the deviation exceeds a first preset threshold as dose hotspots by comparing the dose distribution of the quality assurance plan on the phantom with the planned dose distribution calculated by the treatment planning system, and identifies spatial coordinate regions where the actual dose is lower than the planned dose and the deviation exceeds a second preset threshold as dose coldspots. The system uses these difference indicators and spatial location information as input data for subsequent problem diagnosis to identify specific problem types such as insufficient target coverage, excessively high hotspots, or excessive dose to organs at risk.

[0058] In some embodiments, the problem type includes one or more of the following: The first problem type is used to characterize that the dose corresponding to the first preset volume percentage of the target area is lower than the preset percentage threshold of the prescription dose, or that the dose corresponding to the first preset volume percentage of the target area in the quality assurance plan is lower than the dose corresponding to the first preset volume percentage of the target area in the radiotherapy plan and exceeds the preset difference threshold. The second type of problem is characterized by the dose or maximum dose corresponding to the second preset volume percentage of the target area exceeding the preset percentage threshold of the prescription dose, or the maximum dose of the target area in the quality assurance plan being higher than the maximum dose of the target area in the radiotherapy plan and exceeding the preset difference threshold. The third type of problem is used to characterize the dose index of organs at risk exceeding the clinical limit, or the difference between the quality assurance plan and the radiotherapy plan exceeding the preset difference threshold. The fourth problem type is used to characterize the concentration of gamma failure points exceeding a preset spatial distribution threshold; Problem areas include the target area, areas of concentration of organs at risk or gamma failure points, where areas of concentration of gamma failure points are areas where the density of gamma failure points is greater than a preset density.

[0059] For example, the radiotherapy planning optimization system automatically identifies the types and regions of problems that failed quality assurance verification based on extracted multi-dimensional quality indicators and problem diagnosis rules. The first problem type characterizes insufficient target volume dose coverage, i.e., the dose corresponding to the first preset volume percentage (e.g., 95% volume) of the target volume is lower than the preset percentage threshold (e.g., 95%) of the prescribed dose, or the indicator in the quality assurance plan is lower than the corresponding indicator in the radiotherapy plan and exceeds a preset difference threshold (e.g., 3%). The second problem type characterizes excessive target volume dose, i.e., the dose corresponding to the second preset volume percentage (e.g., 2% volume) of the target volume or the maximum dose exceeds the preset percentage threshold (e.g., 110%) of the prescribed dose. The third problem type characterizes excessive dose to organs at risk. The fourth problem type characterizes the spatially concentrated distribution of gamma failure points. Problem regions include the target volume, organs at risk, or concentrated areas of gamma failure points.

[0060] In some embodiments, the radiotherapy planning optimization system can identify the first problem type: insufficient target dose coverage. The system obtains D95 deviation data for the target area from multi-dimensional quality indicators, finding that the target area D95 is 92% of the prescribed dose, lower than the preset percentage threshold of 95%, and that the target area D95 in the quality assurance plan is 4% lower than the target area D95 in the radiotherapy plan, exceeding the preset difference threshold of 3%. The system also analyzes the constraint weight data of adjacent organs at risk to determine if there is an over-protection of adjacent organs at risk leading to dose "squeezing out." The system locates the problem area as the target area and its adjacent organs at risk, and passes this diagnostic result to the parameter mapping rule base, triggering an increase in the dose constraint weights under the target area.

[0061] In other embodiments, the radiotherapy planning optimization system identifies a third problem type: excessive dose to organs at risk. The system obtains spinal cord Dmax data from multi-dimensional quality indicators and finds that the spinal cord Dmax is 46 Gy, exceeding the clinical limit of 45 Gy. Furthermore, the spinal cord Dmax in the quality assurance plan is 6% higher than that in the radiotherapy plan, exceeding a preset difference threshold of 5%. The system locates the problem area as the organ at risk and passes this diagnosis to the parameter mapping rule base, triggering an increase in the constraint weight of the organ at risk. The system also identifies a fourth problem type. It calculates the local density of gamma failure points from the three-dimensional spatial coordinate distribution of the failure points and finds that the density of gamma failure points in a certain spatial region exceeds a preset density threshold. This indicates a focused dose deviation in that region, and the problem area is located as a concentrated area of ​​gamma failure points, triggering an adjustment of the constraint weight of the relevant region of interest in that region.

[0062] In some embodiments, the quantization parameter mapping rule base includes one or more of the following rules: The first rule includes: when the first problem type is detected, increasing the lower constraint weight of the target area, and simultaneously decreasing the constraint weight of the adjacent endangered organs when the constraint weight of the adjacent endangered organs exceeds a preset ratio. The second rule includes: increasing the upper constraint weight of the target region or decreasing the modulation factor when the second problem type is detected; The third rule includes: when a third problem type is detected, increasing the constraint weight of the organ at risk corresponding to the third problem type, or raising the organ protection level of the organ at risk. The fourth rule includes: adjusting the constraint weights of the concentrated region of gamma failure points when the fourth problem type is detected; The fifth rule includes: when the global gamma pass rate is detected to be lower than a preset threshold, the constraint weights of the target area and the constraint weights of organs at risk are simultaneously increased.

[0063] For example, a radiotherapy planning optimization system can automatically generate optimized parameter adjustment schemes based on a pre-stored quantitative parameter mapping rule base, according to the identified problem type and problem region. The quantitative parameter mapping rule base is a set of "condition-action" rules that map the diagnosed problem type to specific parameter adjustment strategies. Each rule defines the triggering condition and the corresponding adjustment action, including the target parameter, adjustment direction, and adjustment magnitude, and is associated with safety boundary checks. Through this rule base, the radiotherapy planning optimization system transforms the complex clinical decision-making process into a quantifiable and repeatable computational process, replacing the physicist's subjective experience and judgment.

[0064] In some embodiments, the radiotherapy planning optimization system can adjust the lower constraint weights of the target area based on a first problem type. Upon detecting a first problem type of insufficient target dose coverage, the system triggers a first rule: increasing the lower constraint weights of the target area by multiplying the adjustment factor by the deviation magnitude. For example, the system increases the lower Dose constraint weight of the target area from 50 to 62.5. The system also checks the constraint weights of adjacent organs at risk. If their weights exceed a preset proportion (e.g., 80% of the target area weight), the constraint weights of adjacent organs at risk are simultaneously reduced by 10% to 20% to release overprotected dose space. The system also adjusts the organ protection level, reducing the protection level of adjacent organs at risk by one level. Through quantitative adjustments of the rules, the system can increase the target dose while avoiding overprotection of adjacent organs at risk.

[0065] In other embodiments, the radiotherapy planning optimization system adjusts the upper constraint weights of the target area based on a second problem type. Upon detecting a second problem type with excessively high hotspots in the target area, the system triggers a second rule: increasing the upper constraint weight of the target area or decreasing the modulation factor to weaken the beam modulation intensity. The system calculates the weight increment according to the adjustment formula, increasing the Upper Dose constraint weight of the target area from 50 to 75, while simultaneously checking if the current modulation factor is greater than 2.5; if so, it decreases it by 0.2 to 0.5. When both insufficient target coverage and excessively high hotspots exist simultaneously, the system can combine the first and second rules to balance the adjustment of the lower and upper constraints. The system also triggers a third rule based on a third problem type: increasing the constraint weight of the corresponding endangered organ or raising the organ protection level of that endangered organ. For example, the system increases the Upper Dose constraint weight of the spinal cord by a corresponding amount, while simultaneously raising the organ protection level from Level 1 to Level 2. Through the combined application of the rule base, the system can simultaneously handle multiple problem types and generate multi-dimensional optimization parameter adjustment schemes.

[0066] In some embodiments, after generating an optimized parameter adjustment scheme for the radiotherapy plan based on the quantitative parameter mapping rule base and according to the problem type and problem area, it is detected whether the constraint weights of the updated radiotherapy plan are between the preset minimum weight and maximum weight range; whether the planned dose of the updated radiotherapy plan is between the preset lower limit and preset upper limit of the prescription dose; and whether the beam parameters of the updated radiotherapy plan are within the limits of the machine's physical parameters.

[0067] For example, after generating the optimized parameter adjustment scheme, the radiotherapy planning optimization system performs a safety boundary check on the adjusted parameters to ensure that the parameter adjustment results are within a safe range. The minimum and maximum weight ranges refer to the system's preset upper and lower limits for beam weights (e.g., 0.01 to 10000), preventing excessively low weights from causing constraint failure or excessively high weights from causing optimization divergence. The preset lower and upper limits for the prescription dose refer to the reasonable range of the dose target value (e.g., 50% to 150% of the prescription dose), preventing the dose target value from deviating from the clinically reasonable range. The machine physical parameter limitation range refers to the physical constraints of the Tomo radiotherapy equipment itself (e.g., pitch factor 0.10–0.50, modulation factor 1.0–999.99, beam width 10 / 25 / 50 mm), ensuring that the adjusted beam parameters are within the equipment's executable range. Through safety boundary checks, the radiotherapy planning optimization system avoids parameter adjustments exceeding the safe range, which could lead to unexecutable plans or abnormal optimization results.

[0068] In some embodiments, the radiotherapy planning optimization system performs a safety boundary check on the constraint weights. After increasing the lower constraint weight of the target area from 50 to 75 according to a first rule, the system checks whether this weight is within a preset range of 0.01 to 10000, confirming that 75 is within the safe range. If the system increases the upper constraint weight of the target area from 8000 to 12000 according to a second rule, this weight exceeds the upper limit of 10000. The system automatically truncates the weight to the upper limit of 10000 and records the weight exceeding the limit in a log. The system also checks the adjusted dose target value to ensure that the new target value does not exceed 150% of the prescribed dose and is not lower than 50%. If the calculated compensation dose causes the new target value to exceed the safe range, the system adjusts the target value to the closest safe boundary value and generates a prompt message informing the user that the parameter has been corrected by the boundary constraints.

[0069] In some embodiments, the radiotherapy planning optimization system performs machine physical parameter limit checks on the Tomo beam parameters. The system adjusts the modulation factor according to a second rule, increasing it from 2.0 to 2.7. The system checks if this value is within the device-supported range of 1.0 to 999.99, confirming that 2.7 is within a safe range. If the system adjusts the pitch factor from 0.30 to 0.55, exceeding the physical limit range of 0.10 to 0.50, the system automatically truncates the pitch factor to the upper limit of 0.50 and prompts the user that the beam parameters have been corrected for boundary constraints. For the beam width parameter, the system verifies whether the adjusted width is a fixed value supported by the machine (10mm, 25mm, or 50mm). If the adjustment result is a non-standard value, the system automatically selects the closest standard value and provides a prompt. Through these safety boundary checks, the system ensures that all parameter adjustments are within the physically executable range, avoiding optimization failures or plan inoperability due to parameter out-of-bounds errors.

[0070] In some embodiments, when a re-optimization of the modified radiotherapy plan is triggered, if the optimization parameter adjustment scheme involves updating the constraint weights in the radiotherapy plan, the task identifier and dose influence matrix of the previous optimization are reused, and the flux map optimization update operation is re-executed; if the optimization parameter adjustment scheme involves updating the beam parameters in the radiotherapy plan, the dose influence matrix is ​​recalculated and the flux map optimization update operation is executed.

[0071] For example, when the radiotherapy planning optimization system triggers a re-optimization of the modified radiotherapy plan, it automatically selects the optimization path based on the type of optimization parameter adjustment scheme. The dose influence matrix is ​​a mathematical matrix describing the contribution of each beam unit to the dose of each voxel; it involves a large amount of computation and is a computationally intensive part of the optimization process. Flux map optimization is an iterative process of adjusting the beam intensity distribution based on the dose influence matrix to optimize the dose distribution. By distinguishing between two different types of adjustments—constraint weight modification and beam parameter modification—the radiotherapy planning optimization system selects different optimization paths, thereby improving computational efficiency while ensuring optimization accuracy.

[0072] In some embodiments, the radiotherapy planning optimization system detects that the optimization parameter adjustment scheme only involves modifications to constraint weights and does not involve changes to beam parameters. The system reuses the task identifier and the already calculated dose-effect matrix from the previous optimization task, skipping the recalculation step of the dose-effect matrix and directly proceeding to the flux map optimization update operation. For example, when the system adjusts only the constraint weights of the target area and organs at risk according to the first problem type, the dose-effect matrix remains unchanged, and the system directly re-executes flux map optimization based on the original matrix. This approach helps to shorten the computation time for re-optimization, thereby improving iteration efficiency.

[0073] In other embodiments, the radiotherapy planning optimization system detects that the optimization parameter adjustment scheme involves modifications to beam parameters, such as adjustments to the pitch factor, modulation factor, or beam width. Since changes in beam parameters affect the beam's geometry and dose deposition characteristics, the original dose influence matrix is ​​no longer applicable. The system recalculates the dose influence matrix, calculates the dose contribution of each beam element to each voxel based on the new beam parameters, and then performs an optimization update operation on the flux map. For example, when the system reduces the modulation factor from 2.0 to 1.8 according to the second problem type, the system automatically switches to a full recalculation path, which helps ensure that beam parameter changes are fully reflected in the optimization results. This path improves the accuracy of the optimization results. The system automatically selects the optimal path based on the parameter adjustment type, maximizing computational efficiency while ensuring accuracy.

[0074] In some embodiments, the iteration termination condition includes any of the following conditions: The first condition includes: the global gamma pass rate of quality assurance verification is greater than or equal to the preset pass rate threshold; The second condition includes: the deviation of the feature points of the dose-volume histograms of the target area and the organs at risk is less than a preset percentage threshold. The third condition includes: the parameter adjustment range in two consecutive iterations is less than the preset range threshold; The fourth condition includes: the preset maximum number of iterations has been reached.

[0075] For example, during the iterative update process, the radiotherapy planning optimization system determines whether the quality assurance verification results meet the preset iteration termination conditions. The iteration termination conditions are the criteria for stopping the iterative update and include quality assurance verification compliance conditions, optimization convergence conditions, and maximum iteration protection conditions. Specifically, the first condition is that the global gamma pass rate of the quality assurance verification is greater than or equal to a preset pass rate threshold (e.g., 95%), indicating that the dose distribution of the current radiotherapy plan is highly consistent with the independent verification results. The second condition is that the deviation of key feature points in the dose-volume histograms of the target area and organs at risk is less than a preset percentage threshold (e.g., 3%), indicating that the dose indicators meet clinical accuracy requirements. The third condition is that the parameter adjustment range in two consecutive iterations is less than a preset range threshold (e.g., 5%), indicating that the optimization process has converged. The fourth condition is that the preset maximum number of iterations (e.g., 5 or 10) has been reached, serving as a protection mechanism to prevent infinite iteration.

[0076] In some embodiments, the radiotherapy planning optimization system determines whether to terminate the iteration based on a first condition or a second condition. After each iteration update, the system calculates the global gamma pass rate. If the pass rate reaches 95% or higher, the system determines that the first condition is met, stops the iteration, and outputs the current radiotherapy plan as the target radiotherapy plan. If the global gamma pass rate has not yet reached 95%, the system further checks whether the target area D95 deviation, maximum dose deviation, and deviations of key indicators for each organ at risk are all less than 3%. If these are met, the system determines that the second condition is met, stops the iteration, and outputs the plan. For example, after an iteration, the global gamma pass rate is 94%, but the target area D95 deviation is 2%, the spinal cord Dmax deviation is 1.5%, and the lung V20 deviation is 2.5%. All key deviations are less than 3%, and the system determines that the plan quality has reached a clinically acceptable level and stops the iteration. This condition ensures timely termination when the indicators meet the standards, avoiding unnecessary optimization calculations.

[0077] In other embodiments, the radiotherapy planning optimization system determines whether to terminate the iteration based on a third or fourth condition. The system records the parameter adjustment range for each iteration. When the constraint weight adjustment range for two consecutive iterations is less than 5%, it determines that the optimization has converged to the current parameter configuration, the system stops iterating, and outputs the current radiotherapy plan as the target radiotherapy plan. For example, if the adjustment ranges in two consecutive iterations are 4% and 3%, respectively, both less than 5%, the system determines that continuing the iteration will not yield significant improvement and automatically terminates the iteration. The system also checks whether the preset maximum number of iterations (e.g., 10) has been reached. If so, the iteration is forcibly terminated, and the radiotherapy plan with the highest global gamma pass rate in the current iteration is output as the target radiotherapy plan. Simultaneously, a prompt message is generated to inform the user that the maximum number of iterations has been reached. This protection mechanism prevents infinite loops due to repeated parameter oscillations or convergence difficulties, ensuring the controllability of the iteration process.

[0078] In some embodiments, if the decrease in global gamma pass rate for two consecutive iterations exceeds a preset decrease threshold, the system automatically rolls back to the historical parameter configuration with the highest global gamma pass rate and generates a prompt message.

[0079] For example, the radiotherapy planning optimization system monitors the changing trend of global gamma pass rate during iterative updates. When it detects that the decrease in global gamma pass rate exceeds a preset threshold for two consecutive iterations, it determines that the optimization process has diverged and automatically rolls back to the historical parameter configuration with the highest global gamma pass rate. The preset threshold is used to determine the risk boundary of optimization divergence (e.g., 5%). When the pass rate decrease exceeds this threshold for two consecutive iterations, it indicates that the parameter adjustment direction is inappropriate, and continued iteration may lead to further deterioration of the plan quality. The rollback operation refers to restoring the parameters of the current radiotherapy plan to the version with the highest global gamma pass rate in the historical iteration, avoiding further waste of computational resources in the wrong optimization direction. Through divergence detection and automatic rollback mechanisms, the radiotherapy planning optimization system automatically corrects erroneous optimization directions without manual intervention.

[0080] In some embodiments, the radiotherapy planning optimization system detects a divergence trend during a series of iterative updates. For example, the system records the global gamma pass rate after each iteration. After the first iteration, the pass rate is 92%. After the second iteration, the pass rate drops to 85%, a decrease of 7%, exceeding a preset decrease threshold of 5%. The system continues to monitor the third iteration, where the pass rate further decreases to 79%, a decrease of 6% compared to the second iteration, again exceeding the threshold. The system determines that the pass rate decrease exceeds 5% for two consecutive iterations, triggering the divergence detection protection mechanism. The system automatically rolls back to the parameter configuration with the highest global gamma pass rate in the historical iterations, i.e., the radiotherapy plan after the first iteration, which has a global gamma pass rate of 92%. The system also generates a prompt message informing the user that optimization divergence has been detected, that the system has automatically rolled back to the historical parameter configuration with the highest global gamma pass rate, and suggesting that the user check the prescription dose or adjust the optimization strategy.

[0081] In other embodiments, the radiotherapy planning optimization system performs a rollback and generates detailed diagnostic information after detecting divergence. The system records the specific parameter adjustments that triggered the divergence determination, including parameter modifications that led to a decrease in throughput. For example, the system detects that a decrease in the modulation factor from 2.0 to 1.5 in the second iteration caused a significant drop in throughput, and a further decrease to 1.2 in the third iteration led to further deterioration. The system determines that this parameter adjustment direction is inappropriate, rolls back to the parameter configuration before the adjustment, and includes this diagnostic conclusion in the prompt message: "A decrease in the modulation factor was detected, leading to a continuous decline in gamma throughput; the system has automatically rolled back to the parameter configuration before the adjustment." After the rollback, the system can choose to pause automatic iteration, wait for user confirmation or strategy adjustment, or restart the iteration using a different parameter adjustment direction. Through this mechanism, the system can automatically identify and correct erroneous optimization directions during the optimization process, avoiding deterioration of plan quality and improving the stability and safety of the optimization process.

[0082] According to another aspect of the embodiments of this application, a Tomo radiotherapy planning optimization system based on quality assurance reverse influence is also provided, comprising: a quality assurance plan automatic generation unit, used to generate a corresponding quality assurance plan based on an initial radiotherapy plan; a quality assurance verification unit, used to perform quality assurance verification on the initial radiotherapy plan according to the quality assurance plan; and an iterative control unit, used to perform at least one iterative update on the initial radiotherapy plan when the quality assurance verification is detected to have failed, until a preset iterative termination condition is met, and output the radiotherapy plan obtained from the last iterative update as the target radiotherapy plan; wherein each iterative update includes: generating an optimization parameter adjustment scheme based on the quality assurance verification result of the radiotherapy plan, modifying the optimization parameters of the radiotherapy plan according to the optimization parameter adjustment scheme and triggering re-optimization to obtain an updated radiotherapy plan, and generating a corresponding quality assurance plan based on the updated radiotherapy plan and performing quality assurance verification on the updated radiotherapy plan.

[0083] Figure 2 This is a flowchart of the Tomo radiotherapy planning optimization system according to an embodiment of this application, as follows: Figure 2 As shown, it includes the following steps.

[0084] Step S201: Perform Tomo treatment plan reverse optimization. The radiotherapy planning optimization system acquires user-set input data, including CT image information (CT image data path, image size, voxel spacing, electron density table, etc.), target area and organ-at-risk structural information (structural number, name, 3D contour data, and structural type for each region of interest), optimization constraint information (dose constraints for each region of interest, including constraint type, target dose value, constraint weight, constraint volume, hard / soft constraint flags, and organ protection level settings), Tomo beam parameters (beam width, pitch factor, modulation factor, Jaw mode, irradiation mode), SGO (Stochastic Gradient Optimization) optimization control parameters (maximum number of iterations, convergence threshold, minimum modulation factor, etc.), prescribed dose information (total dose, fractional dose, fractional dose, normalization method, and reference point settings), and machine physical parameters (MLC blade width, maximum blade velocity, gantry rotation speed limit, dose rate range, etc.).

[0085] The radiotherapy planning optimization system assembles all the above input data into a standardized optimization request data packet, generates a unique task identifier for this optimization task, and sends the request to the algorithm server via a remote procedure call interface. The algorithm server performs inverse optimization calculations based on the stochastic gradient optimization algorithm, periodically returning convergence information, current penalty values, and beam weight update data during the calculation process. After optimization, the system receives the optimization results, processes the beam control point data (including gantry angle, blade opening / closing time sequence control point sequence data, cumulative jump weights, etc.), and calculates the Tomo irradiation parameters: for Tomo Helical mode, the total number of rotations is calculated based on the number of control points and the number of control points per rotation, and then the gantry rotation period and treatment bed movement speed are calculated; for Tomo Direct mode, the gantry rotation period is calculated based on the projection interval and the number of control points per rotation. This step generates an initial treatment plan that conforms to Tomo physical constraints, providing an evaluation object for subsequent quality assurance verification.

[0086] Step S202: Automatically generate a quality assurance plan based on the treatment plan. The radiotherapy planning optimization system automatically creates a corresponding quality assurance verification plan based on the treatment plan generated in step S201. First, the system automatically generates a unique identifier for the quality assurance plan, serving as a globally unique identifier within the system. Second, the system automatically loads a preset quality assurance phantom (such as a Cheese Phantom cylindrical phantom) and acquires the phantom's CT image data and radiotherapy structure data. Then, the system performs automatic parameter copying from the treatment plan to the quality assurance plan, including: beam geometry parameters (gantry angle, collimator angle, treatment bed angle), Tomo-specific physical parameters (beam width, pitch factor, modulation factor, Jaw mode, control point sequence data), prescribed dose and fractionation information, and normalization point settings. The system performs automatic parameter validity checks, including: verification of the quality assurance plan name being non-empty and unique, verification of the validity of the quality assurance phantom selection and the integrity of the CT data, verification of the existence and reasonableness of the reference points' locations, verification of the dose grid resolution range, verification of the beam geometry parameter range, and verification of the fractionation range. The system establishes a bidirectional association between the quality assurance plan and the original treatment plan: the forward association stores the unique identifier of the original treatment plan in the quality assurance plan record; the reverse association maintains a list of quality assurance plan information in the original treatment plan record, including the quality assurance plan name, unique identifier, file path, creation time, etc.; it also records the type and identifier of the quality assurance module used. This step achieves fully automated creation from treatment plan to quality assurance plan, eliminating the need for manual setting of Tomo-specific parameters.

[0087] Step S203: Perform dose calculation for the quality assurance plan. The radiotherapy planning optimization system performs dose calculations on the quality assurance plan created in step S202, using a quality assurance phantom. The system prepares the dose calculation input parameters, including the CT image index of the quality assurance phantom, the radiotherapy structure index of the quality assurance plan, the unique identifier of the quality assurance plan, and the file path. The system sets the dose calculation engine parameters: the calculation engine type is Monte Carlo algorithm, the uncertainty is set to 2.0%, and the dose grid resolution is set to 2.0 mm. The system submits a dose calculation request to the algorithm server through the algorithm service interface. The algorithm server performs the following calculations: calculates the dose influence matrix, performs Monte Carlo dose simulation, and generates three-dimensional dose distribution data. After the calculation is completed, the system receives and stores the complete dose distribution data on the quality assurance phantom, generates beam dose and total dose files, and updates the quality assurance plan status to dose calculation complete. This step calculates the dose distribution of the treatment plan on a standardized quality assurance phantom, providing a benchmark for subsequent plan comparison and validation.

[0088] Step S204 involves performing plan comparison and secondary independent dose verification. For example, based on the dose calculation results from step S203, the radiotherapy planning optimization system performs multi-dimensional plan comparison analysis and independent dose verification, including two sub-steps: DVH (dose-volume histogram) plan comparison analysis and secondary independent dose verification.

[0089] like Figure 2 As shown, the DVH plan comparison analysis includes: the system acquires complete DVH datasets from the original treatment plan and the quality assurance plan, which are designated as reference DVH data and comparison DVH data, respectively. Since the regions of interest (ROIs) in the treatment plan and the quality assurance plan may differ, the system uses intelligent matching by name (supporting exact matching and fuzzy matching ignoring case / spaces) to create a pairing list. For successfully paired ROIs, the system performs multi-dimensional DVH comparison analysis: dose statistics comparison (absolute and relative percentage differences in maximum, minimum, and average doses), dose-volume point comparison (differences in key indicators such as D95 and D2 (corresponding to the doses corresponding to the second preset volume ratio mentioned above), volume indicators comparison (differences in key indicators such as V20), and DVH curve morphology comparison (area difference between the two curves, maximum deviation point, and deviation magnitude). The system classifies the comparison results into four levels: Level 1 (differences in all indicators less than 2%), Level 2 (differences in all indicators less than 3%), Level 3 (differences in all indicators less than 5%), and Level 4 (differences in some indicators greater than or equal to 5%).

[0090] like Figure 2As shown, the secondary independent dose verification includes: the system prepares a verification data package, including the DICOM RT Plan file, RT Dose file, and RT Structure file of the quality assurance plan, and submits the data to the independent dose verification system via an HTTP interface. This system recalculates the quality assurance phantom dose using an independent dose calculation engine different from the treatment planning system. The system performs three-dimensional gamma analysis, comparing the dose calculated by the treatment planning system and the dose calculated by the independent verification system. The gamma analysis standard is 3% / 3mm, and the gamma pass rate (the percentage of voxels with a gamma of less than 1 out of the total number of voxels) is calculated, generating three-dimensional spatial distribution data of the gamma analysis. The system extracts the verification results, including the gamma pass rate, verification status (pass, warning, alert), and the three-dimensional spatial distribution of gamma failure points. This step comprehensively evaluates the dose accuracy of the treatment plan through two complementary methods: DVH comparison and independent gamma analysis. When the gamma pass rate is greater than or equal to 95% and the difference of all key DVH indicators is less than 3%, the quality assurance verification is considered passed, and the process ends; otherwise, proceed to step S205 for problem diagnosis.

[0091] Step S205: Multi-dimensional analysis of quality assurance results and automatic problem diagnosis. For example, when the quality assurance verification result in step S204 is unsuccessful, the radiotherapy planning optimization system automatically extracts multi-dimensional quality indicators from the verification results and performs problem diagnosis.

[0092] For example, the extraction of multi-dimensional quality indicators includes: gamma analysis indicators (global gamma pass rate, regional gamma pass rate, and three-dimensional coordinate distribution of gamma failure points); DVH difference indicators (D95 deviation of the target area, maximum dose deviation; deviation of key indicators of organs at risk, deviation magnitude and direction of each indicator); and dose spatial distribution characteristics (dose hotspot locations, dose coldspot locations, and abnormal dose gradient regions).

[0093] Based on the extracted quality indicators, the system performs automated problem diagnosis. The criteria for diagnosing insufficient target coverage are: the target area's D95 is lower than 95% of the prescribed dose, or the D95 of the target area in the quality assurance plan is more than 3% lower than the original plan. The problem is identified as an excessively low lower dose constraint weight for the target area, or excessive protection of adjacent organs at risk leading to dose displacement. The criteria for diagnosing excessive target dose are: the target area's D2 or maximum dose exceeds 110% of the prescribed dose, or the maximum dose of the target area in the quality assurance plan is more than 3% higher than the original plan. The problem is identified as an excessively low upper dose constraint weight for the target area, or an excessively high modulation factor setting. The criteria for diagnosing excessive dose to organs at risk are: a critical indicator of an organ at risk exceeds the clinical limit, or the quality assurance difference exceeds 5%. The problem is identified as insufficient constraint weight for that organ at risk, or an imbalance in the weight ratio with adjacent target areas. The criteria for diagnosing uneven dose distribution are: gamma failure points are concentrated in a specific spatial region. The problem is identified as the need to adjust the beam optimization parameters in that direction. The criteria for diagnosing poor overall quality is a global gamma pass rate below 90%, indicating that the problem stems from unreasonable global optimization parameters requiring overall adjustment. This step transforms the quality assurance verification results from a qualitative "pass / fail" judgment into a quantitative problem diagnosis, automatically identifying the problem type and location.

[0094] Step S206: Automatic mapping and adjustment of optimization parameters based on quality assurance diagnostic results. For example, the radiotherapy planning optimization system automatically calculates and executes an optimization parameter adjustment scheme based on the diagnostic results of step S205 using a parameter mapping rule base.

[0095] For example, the parameter mapping rule base adopts a "condition-action" rule format. Rule 1 (Insufficient Target Coverage) is triggered when the target area D95 deviation exceeds 3% and the deviation direction is negative. The action is to increase the lower dose constraint weight of the target area by 20% to 50%, while simultaneously checking the constraint weight of adjacent organs at risk; if it is too high, it is appropriately reduced by 10% to 20%. Rule 2 (Excessive Target Hotspot) is triggered when the maximum dose deviation of the target area exceeds 5% and the deviation direction is positive. The action is to increase the upper dose constraint weight of the target area by 30% to 60%, or reduce the current modulation factor by 0.2 to 0.5 if it is greater than 2.5. Rule 3 (Excessive Dose to Organs at Risk) is triggered when the critical indicator of an organ at risk exceeds the limit by X%. The action is to increase the corresponding constraint weight of the organ at risk by (X / 2)% to X%. If the organ at risk has organ protection settings, the protection level is increased by 1 level. Rule 4 (Local Gamma Failure) is triggered when gamma failure points are concentrated in a specific spatial region. The action is to analyze the region of interest (ROI) within that area, prioritizing an increase in the constraint weight of that ROI. If the failure region is related to the beam boundary, consider adjusting the pitch factor or Jaw mode. Rule 5 (Low Overall Gamma Passivity) is triggered when the global gamma passivity is below 90%. The action is to increase the overall constraint weight of all target areas by 10% to 15%, and the overall constraint weight of all organs at risk by 5% to 10%. Consider increasing the number of optimization iterations.

[0096] For example, the system performs quantitative calculations for parameter adjustments, using the constraint weight adjustment formula: the new weight equals the original weight multiplied by (1 plus the adjustment coefficient multiplied by the deviation amplitude); the dose target value adjustment formula: the new target value equals the original target value plus the compensation dose, and the compensation dose equals the quality assurance actual dose minus the original planned dose; Tomo beam parameter adjustments: if dose uniformity needs to be increased, the pitch is reduced; if modulation capability needs to be increased, the modulation factor is increased; if dose gradient needs to be improved, the beam width is changed. The system performs parameter safety boundary checks: the constraint weight range is 0.01 to 10000, the dose target value does not exceed 150% of the prescription dose and is not less than 50%, Tomo parameter physical limitations (pitch 0.10 to 0.50, modulation factor 1.0 to 999.99, beam width is a fixed value supported by the machine), and the single adjustment amplitude is limited to no more than 50% of the original value. The system automatically modifies and optimizes constraints through a programmatic interface, updates constraint parameters item by item according to the adjustment scheme, writes the updated constraint list back to the database, updates the optimization constraint version number, records the modification time, modification source and modification reason, and saves a snapshot of the constraints before modification to support rollback operations. This step enables an automated closed-loop mapping from quality assurance verification results to optimized parameter adjustments, transforming the subjective adjustment process that originally relied on physicists' experience into a quantifiable and repeatable automated process.

[0097] Step S207: Automatic re-optimization and iterative verification are triggered. For example, after the parameter adjustment in step S206 is completed, the radiotherapy planning optimization system automatically triggers re-optimization and enters the iterative verification loop.

[0098] For example, the system automatically selects the optimization path based on the parameter adjustment type. Path A (constraint weight modification only): This approach reuses the task identifier and dose influence matrix from the previous optimization, only re-executing the flux map optimization, resulting in a computational speed improvement of approximately 50%. Path B (beam parameter modification involved): This approach completely re-optimizes, reusing the task identifier but recalculating the dose influence matrix and flux map optimization, ensuring that beam parameter changes are fully reflected in the optimization results.

[0099] After the re-optimization is completed, the system automatically performs the dose calculation in step S203 and the quality assurance verification in step S204.

[0100] For example, the iterative control strategy includes: a maximum iteration limit of N iterations (default N=3) for feedback optimization loops, each loop including parameter adjustment, re-optimization, dose calculation, quality assurance verification, and result evaluation. Iteration stops when any one of the convergence criteria is met: Criterion 1 is a global gamma pass rate greater than or equal to 95%; Criterion 2 is a difference of less than 3% in all key DVH indicators; Criterion 3 is a parameter adjustment range of less than 5% in two consecutive iterations. The divergence detection and protection mechanism is as follows: if the global gamma pass rate decreases by more than 5% in two consecutive iterations, it is considered divergence. In the event of divergence, the system automatically rolls back to the historical parameter configuration with the highest global gamma pass rate, prompting the user that manual intervention may be necessary.

[0101] For example, the feedback history records complete information for each iteration, including the iteration number, timestamp, list of modified constraints and comparison before and after modification, optimization results (number of convergence rounds, final penalty score, computation time), quality assurance verification results (global gamma pass rate, details of DVH differences), and adjustment decision basis (triggered rule number, type of diagnosed problem), for quality traceability and experience accumulation. This step achieves complete closed-loop iterative control, ensuring the feedback optimization process is efficient, safe, and controllable through intelligent path selection (skipping dose influence matrix calculation only when constraints are modified to accelerate, and fully recalculating when beam parameters are modified to ensure accuracy), convergence determination, and divergence protection.

[0102] 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 Tomo radiotherapy planning optimization method based on quality assurance reverse influence.

[0103] 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 Tomo radiotherapy planning optimization method based on quality assurance reverse effects.

[0104] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program or instructions, which, when executed by a processor, implement the above-described Tomo radiotherapy planning optimization method based on quality assurance reverse effects.

[0105] 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.

[0106] 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.

[0107] 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 example, 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 of units or modules may be electrical or other forms.

[0108] The units described as separate components may or may not be physically separate. Similarly, the components shown as units may or may not be physical units; 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, depending on actual needs.

[0109] 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.

[0110] 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 of 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.

[0111] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles 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 optimizing Tomo radiotherapy plans based on the reverse influence of quality assurance, characterized in that, include: Based on the initial radiotherapy plan, a corresponding quality assurance plan is generated; The initial radiotherapy plan was validated for quality assurance purposes in accordance with the aforementioned quality assurance plan. If the quality assurance verification fails, the initial radiotherapy plan is updated at least once until the preset iteration termination condition is met, and the radiotherapy plan obtained from the last iteration update is output as the target radiotherapy plan. Each iteration update includes: generating an optimization parameter adjustment scheme based on the quality assurance verification results of the radiotherapy plan; modifying the optimization parameters of the radiotherapy plan according to the optimization parameter adjustment scheme and triggering re-optimization to obtain an updated radiotherapy plan; and generating a corresponding quality assurance plan based on the updated radiotherapy plan and performing the quality assurance verification on the updated radiotherapy plan. The process of generating an optimized parameter adjustment scheme based on the quality assurance verification results of the radiotherapy plan includes: in response to the failure of the quality assurance verification of the radiotherapy plan, obtaining multi-dimensional quality indicators of the quality assurance verification; analyzing the multi-dimensional quality indicators based on problem diagnosis rules to obtain the problem types and corresponding problem areas that led to the failure of the quality assurance verification; and generating an optimized parameter adjustment scheme for the radiotherapy plan based on the problem types and problem areas according to a quantitative parameter mapping rule library. The multi-dimensional quality indicators include at least global gamma pass rate and regional gamma pass rate. The global gamma pass rate characterizes the gamma analysis pass rate between the dose distribution in the quality assurance plan and the first dose distribution, where the first dose distribution is the dose distribution recalculated by the independent dose validation system for the quality assurance plan. The regional gamma pass rate characterizes the gamma analysis pass rate between the dose distribution and a second dose distribution in the dose region defined by the quality assurance plan based on at least one preset ratio of the prescription dose; the second dose distribution is the dose distribution recalculated by the independent dose verification system for the dose region based on the quality assurance plan.

2. The method according to claim 1, characterized in that, Each iteration update includes the following steps: Based on the optimization parameter adjustment scheme, the optimization parameters of the radiotherapy plan are modified, and the modified radiotherapy plan is re-optimized to generate an updated radiotherapy plan; After generating a corresponding quality assurance plan based on the updated radiotherapy plan, the quality assurance verification is performed on the updated radiotherapy plan, and it is determined whether the verification result meets the preset iteration termination condition.

3. The method according to claim 1, characterized in that, Based on the initial radiotherapy plan, a corresponding quality assurance plan is generated, including: Load a preset quality assurance phantom, wherein the quality assurance phantom is a phantom used for dose verification; Acquire the image data and radiation region of interest data of the quality assurance phantom; Based on the beam geometry parameters, physical parameters of helical tomotherapy, dose information, fractionation information, and image data and region of interest data of the quality assurance phantom in the initial radiotherapy plan, the quality assurance plan is generated. The physical parameters of helical tomotherapy include at least one of beam width, pitch factor, modulation factor, collimator mode, and control point sequence.

4. The method according to claim 1, characterized in that, After generating a corresponding quality assurance plan based on the initial radiotherapy plan, the method further includes: Based on the initial radiotherapy plan, a unique identifier is generated to identify the quality assurance plan; Based on the unique identifier, establish a bidirectional association between the quality assurance plan and the initial radiotherapy plan; The bidirectional associated storage includes: storing the plan identifier of the initial radiotherapy plan in the quality assurance plan, and storing an information list of the quality assurance plan in the initial radiotherapy plan, wherein the information list includes the name of the quality assurance plan, the unique identifier of the quality assurance plan, the file path, and the creation time.

5. The method according to claim 2, characterized in that, The multi-dimensional quality indicators also include at least one of the following indicators: The three-dimensional spatial coordinate distribution of gamma failure points, wherein the gamma failure point is characterized as: a voxel point where the gamma analysis value is greater than the target threshold between the dose distribution of the quality assurance plan and the first dose distribution; The feature point deviations of the dose-volume histogram of the target area include: the dose deviation corresponding to the first preset volume ratio and the maximum dose deviation; Characteristic point bias in the dose-volume histogram of organs at risk; The dose hotspot location and the dose coldspot location, wherein the dose hotspot location represents the spatial coordinate region where the actual dose is higher than the planned dose and the deviation exceeds a first preset threshold, and the dose coldspot location represents the spatial coordinate region where the actual dose is lower than the planned dose and the deviation exceeds a second preset threshold.

6. The method according to claim 5, characterized in that, The problem types include one or more of the following: The first problem type is used to characterize that the dose corresponding to the first preset volume percentage of the target area is lower than the preset percentage threshold of the prescription dose, or that the dose corresponding to the first preset volume percentage of the target area in the quality assurance plan is lower than the dose corresponding to the first preset volume percentage of the target area in the radiotherapy plan and exceeds the preset difference threshold. The second type of problem is characterized by the dose or maximum dose corresponding to the second preset volume percentage of the target area exceeding the preset percentage threshold of the prescription dose, or the maximum dose of the target area in the quality assurance plan being higher than the maximum dose of the target area in the radiotherapy plan and exceeding the preset difference threshold. The third type of problem is characterized by dose indicators that indicate organs at risk exceeding clinical limits, or by differences between the quality assurance plan and the radiotherapy plan exceeding a preset difference threshold. The fourth problem type is used to characterize the concentration of gamma failure points exceeding a preset spatial distribution threshold; The problem area includes the target area, the organ at risk, or the concentrated area of ​​the gamma failure points, wherein the concentrated area of ​​the gamma failure points is the area where the density of the gamma failure points is greater than a preset density.

7. The method according to claim 6, characterized in that, The quantization parameter mapping rule base includes one or more of the following rules: The first rule includes: when the first problem type is detected, increasing the lower constraint weight of the target area, and simultaneously decreasing the constraint weight of the adjacent endangered organ when the constraint weight of the adjacent endangered organ exceeds a preset ratio. The second rule includes: increasing the upper constraint weight of the target region or decreasing the modulation factor when the second problem type is detected; The third rule includes: when the third problem type is detected, increasing the constraint weight of the organ at risk corresponding to the third problem type, or raising the organ protection level of the organ at risk; The fourth rule includes: adjusting the constraint weights of the concentrated region of the gamma failure points when the fourth problem type is detected; The fifth rule includes: when the global gamma pass rate is detected to be lower than a preset threshold, the constraint weight of the target area and the constraint weight of the organ at risk are simultaneously increased.

8. The method according to claim 7, characterized in that, After generating an optimized parameter adjustment scheme for the radiotherapy plan based on the quantitative parameter mapping rule base, according to the problem type and the problem region, the method further includes: Detect whether the constraint weights of the updated radiotherapy plan are within the preset range of minimum and maximum weights; The system checks whether the planned dose after the radiotherapy plan is updated is between the preset lower limit and the preset upper limit of the prescription dose. The system checks whether the updated beam parameters of the radiotherapy plan are within the limits of the machine's physical parameters.

9. The method according to claim 1, characterized in that, When triggering a re-optimization of the modified radiotherapy plan, the method further includes: If it is detected that the optimization parameter adjustment scheme involves updating the constraint weights in the radiotherapy plan, the task identifier and dose influence matrix of the previous optimization are reused, and the flux map optimization update operation is re-executed. If the optimization parameter adjustment scheme is detected to involve updating the beam parameters in the radiotherapy plan, the dose influence matrix is ​​recalculated and the flux map is optimized and updated.

10. The method according to claim 5, characterized in that, The iteration termination condition includes any one of the following conditions: The first condition includes: the global gamma pass rate of the quality assurance verification is greater than or equal to a preset pass rate threshold. The second condition includes: the deviation of the feature points of the dose-volume histograms of the target area and the organs at risk is less than a preset percentage threshold. The third condition includes: the parameter adjustment range in two consecutive iterations is less than the preset range threshold; The fourth condition includes: the preset maximum number of iterations has been reached.

11. The method according to claim 5, characterized in that, The method further includes: If the global gamma pass rate decreases by more than a preset threshold after two consecutive iterations, the system will automatically roll back to the historical parameter configuration with the highest global gamma pass rate and generate a prompt message.

12. A Tomo radiotherapy planning optimization system based on the reverse influence of quality assurance, characterized in that, include: The quality assurance plan automatic generation unit is used to generate a corresponding quality assurance plan based on the initial radiotherapy plan; A quality assurance verification unit is used to perform quality assurance verification on the initial radiotherapy plan in accordance with the quality assurance plan. An iterative control unit is used to perform at least one iterative update on the initial radiotherapy plan when the quality assurance verification is detected to be unsuccessful, until a preset iteration termination condition is met, and output the radiotherapy plan obtained from the last iterative update as the target radiotherapy plan. Each iteration update includes: generating an optimization parameter adjustment scheme based on the quality assurance verification results of the radiotherapy plan; modifying the optimization parameters of the radiotherapy plan according to the optimization parameter adjustment scheme and triggering re-optimization to obtain an updated radiotherapy plan; and generating a corresponding quality assurance plan based on the updated radiotherapy plan and performing the quality assurance verification on the updated radiotherapy plan. The process of generating an optimized parameter adjustment scheme based on the quality assurance verification results of the radiotherapy plan includes: in response to the failure of the quality assurance verification of the radiotherapy plan, obtaining multi-dimensional quality indicators of the quality assurance verification; analyzing the multi-dimensional quality indicators based on problem diagnosis rules to obtain the problem types and corresponding problem areas that led to the failure of the quality assurance verification; and generating an optimized parameter adjustment scheme for the radiotherapy plan based on the problem types and problem areas according to a quantitative parameter mapping rule library. The multi-dimensional quality indicators include at least global gamma pass rate and regional gamma pass rate. The global gamma pass rate characterizes the gamma analysis pass rate between the dose distribution in the quality assurance plan and the first dose distribution, where the first dose distribution is the dose distribution recalculated by the independent dose validation system for the quality assurance plan. The regional gamma pass rate characterizes the gamma analysis pass rate between the dose distribution and a second dose distribution in the dose region defined by the quality assurance plan based on at least one preset ratio of the prescription dose; the second dose distribution is the dose distribution recalculated by the independent dose verification system for the dose region based on the quality assurance plan.

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 Tomo radiotherapy planning optimization method based on quality assurance reverse effects 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 for storing 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 Tomo radiotherapy planning optimization method based on quality assurance reverse effects as described in 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 Tomo radiotherapy planning optimization method based on quality assurance back influence as described in any one of claims 1 to 11.

Images