BNCT treatment plan generation method, device, medium and terminal based on barrier algorithm

Abstract

The present application provides a BNCT treatment plan generation method, device, medium and terminal based on a barrier algorithm, the method comprising: obtaining multiple irradiation fields; calculating the unit time dose distribution of each irradiation field respectively; using the irradiation time combination of the multiple irradiation fields as a generation target, and determining the evaluation index and constraint conditions of the generation target; obtaining the generation target of the minimum value of the evaluation index under the constraint of the barrier algorithm to obtain the optimal irradiation time combination of the multiple irradiation fields; determining the irradiation time of each irradiation field based on the optimal irradiation time combination, and generating a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution. The present application can generate a BNCT treatment plan with high dose uniformity, which can reduce damage to surrounding healthy tissues while ensuring sufficient radiation dose, and effectively improve the safety of radiotherapy.

Description

Technical Field

[0001] The present application belongs to the field of radiation technology and relates to a BNCT treatment plan generation method, device, medium and terminal based on a barrier algorithm. Background Art

[0002] Boron Neutron Capture Therapy (BNCT) is a precision radiotherapy that combines targeted drugs with neutron irradiation, selectively killing tumor cells while sparing normal cells. Compared to photon radiotherapy, BNCT effectively reduces radiation exposure to normal cells, minimizing the side effects of radiotherapy. Photon radiotherapy typically requires multiple doses, resulting in a long treatment cycle. However, BNCT requires only one or two doses to achieve the required tumor-killing dose, making treatment time very patient-friendly and reducing the burden of treatment. Compared to other radiotherapy techniques, BNCT offers significant advantages in treatment time, precision, and side effect control, and holds broad application prospects.

[0003] In BNCT treatment, evaluation indicators help to effectively evaluate the treatment of BNCT treatment plans. For example, when dose uniformity is used as an evaluation indicator, this evaluation indicator is crucial to the effectiveness of radiotherapy, which may affect the tumor control rate and the side effects of radiotherapy. Uniform dose distribution helps to ensure complete elimination of the tumor, reduce the risk of recurrence, and can reduce high-dose areas in the patient's body, thereby reducing damage to normal tissues, such as side effects such as dermatitis or urethral toxicity, and improving the safety of radiotherapy. This is very important in the treatment of breast cancer, prostate cancer, and head and neck cancer. Therefore, how to efficiently obtain a BNCT treatment plan that meets the evaluation indicators is a technical problem that needs to be solved urgently by those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a BNCT treatment plan generation method and device based on a barrier algorithm, a storage medium and a terminal, which are used to solve the technical problem of how to efficiently obtain a BNCT treatment plan that meets the evaluation indicators.

[0005] In a first aspect, the present application provides a method for generating a BNCT treatment plan based on a barrier algorithm, the method comprising:

[0006] Acquire multiple irradiation fields;

[0007] respectively calculating the unit time dose distribution of each irradiation field;

[0008] Combining the irradiation times of the multiple irradiation fields as a generation target, and determining evaluation indicators and constraint conditions for the generation target;

[0009] obtaining the generation target of the minimum value of the evaluation index under the constraint condition based on the barrier algorithm to obtain the optimal irradiation time combination of the multiple irradiation fields;

[0010] The irradiation time of each irradiation field is determined based on the optimal irradiation time combination, so as to generate a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution.

[0011] In one embodiment of the present application, the constraint conditions include tumor target radiation constraint, organ at risk radiation constraint and irradiation time constraint: wherein the tumor target radiation constraint and the organ at risk radiation constraint are determined by an average radiation dose.

[0012] In one embodiment of the present application, obtaining the generation target with the minimum evaluation index value under the constraint of the constraint condition based on the barrier algorithm includes:

[0013] Acquire a generation model based on a barrier algorithm, wherein the generation model is used to obtain a generation target of a minimum value of the evaluation index that satisfies a constraint condition;

[0014] Iteratively solving the generation model until an approximate error of the solution is not less than a preset threshold, thereby obtaining an optimal irradiation time combination for the multiple irradiation fields;

[0015] The preset threshold is the result of comparing the number of the constraint conditions with the hyperparameters of the barrier algorithm.

[0016] In one embodiment of the present application, the method further includes:

[0017] When the solution approximation error is less than a preset threshold, the barrier algorithm hyperparameters are updated to update the preset threshold, and the generation model is updated based on the updated barrier algorithm hyperparameters to re-iterate the solution; wherein, the comparison result of the barrier algorithm hyperparameters and the updated parameters is used as the updated barrier algorithm hyperparameters.

[0018] In one embodiment of the present application, the method includes: if the generative model has no feasible solution, adjusting the constraint conditions to re-acquire the generative model and continuing to iteratively solve the problem.

[0019] In one embodiment of the present application, generating a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution includes:

[0020] Obtaining a minimum value of radiation to the tumor target area and a maximum value of radiation to the organ at risk based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution;

[0021] Performing a generation and determination operation based on the minimum value of the tumor target area radiation and the maximum value of the organ-at-risk radiation;

[0022] When the generation judgment operation is passed, a BNCT treatment plan is generated based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution;

[0023] When the generation judgment operation fails, the constraint condition is adjusted to re-acquire the generation model and continue to iterate and solve.

[0024] In one embodiment of the present application, the generating and judging operation includes: judging whether the maximum radiation value of the organ at risk is less than the maximum tolerated dose; if not, the generating and judging operation fails; if so, continuing to judge whether the minimum radiation value of the tumor target area is not less than the prescribed dose;

[0025] If so, the generation judgment operation is passed; if not, the generation judgment operation is failed.

[0026] In a second aspect, the present application further provides a BNCT treatment plan generation device based on a barrier algorithm, the device comprising:

[0027] An acquisition module, used for acquiring multiple irradiation fields;

[0028] A calculation module, used for respectively calculating the unit time dose distribution of each irradiation field;

[0029] A target module, configured to combine the irradiation times of the plurality of irradiation fields as a generated target, and determine evaluation indicators and constraint conditions for the generated target;

[0030] A search module, configured to search for the generation target of the minimum value of the evaluation index under the constraint of the constraint condition based on a barrier algorithm, so as to obtain an optimal irradiation time combination for the multiple irradiation fields;

[0031] A generating module is configured to determine the irradiation time of each irradiation field based on the optimal irradiation time combination, so as to generate a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution.

[0032] In a third aspect, the present application further provides a storage medium storing a computer program, which, when executed by a processor, implements the BNCT treatment plan generation method based on the barrier algorithm as described above.

[0033] In a fourth aspect, the present application also provides a terminal comprising a processor and a memory, wherein the memory is communicatively connected to the processor; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal executes the BNCT treatment plan generation method based on the barrier algorithm as described above.

[0034] As described above, the barrier algorithm-based BNCT treatment plan generation method, device, medium, and terminal described in this application have the following beneficial effects:

[0035] This application uses the irradiation time combination of multiple irradiation fields as the generation target, and sets evaluation indicators and constraints to obtain the generation target that meets the constraints and the minimum value of the evaluation indicators, that is, the optimal irradiation time combination, so as to generate a BNCT treatment plan that meets the evaluation indicators, improve the treatment effect and reduce the side effects of radiotherapy. At the same time, this application introduces the barrier algorithm into the BNCT treatment plan generation scenario, and blocks the solutions that do not meet the constraints outside the barrier by obtaining a generation model, iteratively solves to quickly obtain the optimal irradiation time combination, and based on this, efficiently generates a BNCT treatment plan that meets the evaluation indicators, so as to ensure that the tumor receives the prescribed dose while making the dose distribution as uniform as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic diagram showing the principle of BNCT treatment described in the examples of the present application is shown.

[0037] Figure 2 A flow chart of the BNCT treatment plan generation method based on the barrier algorithm described in an embodiment of the present application is shown.

[0038] Figure 3 A flow chart of the BNCT treatment plan generation method based on the barrier algorithm described in an embodiment of the present application is shown.

[0039] Figure 4 A schematic diagram of an image of a logarithmic barrier function with different hyperparameter values ​​described in an embodiment of the present application is shown.

[0040] Figure 5 A schematic diagram of the process of obtaining the optimal solution of the generative model described in an embodiment of the present application is shown.

[0041] Figure 6 A logical judgment diagram for generating a BNCT treatment plan according to an embodiment of the present application is shown.

[0042] Figure 7 A schematic structural diagram of a BNCT treatment plan generation device based on a barrier algorithm according to an embodiment of the present application is shown.

[0043] Figure 8 The structural schematic diagram of the terminal described in the embodiments of the present application is shown. Specific embodiments

[0044] The following uses specific specific examples to illustrate the implementation manners of the present application. Those skilled in the art can easily understand other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0045] It should be noted that in the following description, reference is made to the accompanying drawings, which describe several embodiments of the present application. It should be understood that other embodiments can also be used, and mechanical composition, structure, electrical, and operational changes can be made without departing from the spirit and scope of the present application. The following detailed description should not be considered restrictive, and the scope of the embodiments of the present application is only defined by the claims of the published patent. The terms used here are only for describing specific embodiments and are not intended to limit the present application. Spatially related terms, such as "upper", "lower", "left", "right", "below", "beneath", "lower part", "above", "upper part", etc., can be used in the text to facilitate the description of the relationship between one element or feature shown in the figure and another element or feature.

[0046] Furthermore, as used herein, the singular forms "a", "an", and "the" are also intended to include the plural forms unless the context indicates otherwise. It should be further understood that the terms "comprising", "including" indicate the presence of the described features, operations, elements, components, items, types, and / or groups, but do not exclude the presence, occurrence, or addition of one or more other features, operations, elements, components, items, types, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or meaning any one or any combination.

[0047] BNCT is a targeted particle therapy method that selectively kills tumor cells and preserves normal cells. Its principle is shown in Figure 1. During the treatment process, first, a targeting drug carrying 10 B, such as BPA (Boronophenylalanine) or BSH (Sodium Borocaptate), is introduced into the patient's body by intravenous injection. At this time, the targeting drug will selectively accumulate in tumor cells. When the tumor cells accumulate a sufficient number of 10 B atoms, the tumor is irradiated with a beam of epithermal neutrons (0.5 eV < En < 10 keV) or thermal neutrons (En < 0.5 eV). The10 The B (n,α)7Li fission reaction will produce α particles (4He) and recoil 7Li+ nuclei. The reaction equation is shown in (1):

[0048]

[0049] From formula (1), we can see that 10 B as the target nucleus absorbs neutrons to form an excited compound nucleus ¹¹B, which then directly splits into lithium (Li) and helium nuclei (α particles), and de-excites by releasing γ rays. In fact, α particles and 7 The linear energy transfer (LET) of Li+ nuclei is very high, 164 keV / μm and 151 keV / μm respectively, and the α particles and 7 The range of Li+ nuclei is very short, 9μm and 5μm respectively, which is smaller than the diameter of most cells (about 10μm). This means that the fission reaction occurring in tumor cells has almost no impact on cells in surrounding tissues. BNCT can kill tumor cells without accidentally harming surrounding normal cells.

[0050] In the process of generating a BNCT treatment plan, setting reasonable evaluation indicators can effectively evaluate the treatment of the treatment plan. Therefore, how to efficiently generate a BNCT treatment plan that meets the evaluation indicators is a technical problem that needs to be solved by those skilled in the art.

[0051] In order to at least solve the above technical problems, the embodiments of the present application provide a BNCT treatment plan generation method, device, medium and terminal based on the barrier algorithm, which can quickly generate a BNCT treatment plan that meets the evaluation indicators and improve the BNCT treatment effect.

[0052] Figure 2 FIG. 1 shows a flow chart of a BNCT treatment plan generation method based on a barrier algorithm according to an embodiment of the present application, as shown in FIG. Figure 2 As shown, the BNCT treatment plan generation method based on the barrier algorithm provided in the embodiment of the present application includes steps S1 to S5.

[0053] S1. Acquire multiple irradiation fields.

[0054] In some embodiments, after the tumor target volume is determined, the positions of the tumor target area contour points can be outlined using a DICOM-RT Struct file, and the arithmetic average of these positions is taken as the geometric center point A of the tumor target volume. The distance between the skin contour points outlined in the DICOM-RT Struct file and point A is then calculated. The point with the smallest distance is used as the starting point, and the vector pointing to point A is the irradiation angle of the skin surface closest to the geometric center point A. The irradiation angle is rotated 360° in the horizontal plane with point A as the center, and any angle is selected and recorded as the irradiation angle of an irradiation field, thereby allowing the selection of irradiation fields with multiple different irradiation angles.

[0055] Furthermore, since the irradiation field usually needs to be selected on the side of the skin closer to the tumor and needs to avoid the treatment bed, the final multiple irradiation fields can be obtained by removing these unreasonable angles.

[0056] S2. Calculate the dose distribution per unit time of each irradiation field respectively.

[0057] In some embodiments, the dose distribution per unit time for each irradiation field can be simulated and calculated using a Monte Carlo method. For example, a DICOM-RT Image file obtained from a CT scan is converted into a voxel model suitable for Monte Carlo calculations. The voxel model is then populated with corresponding boron concentrations based on the organs outlined in the DICOM-RT Struct file to simulate the boron distribution in a patient after BSA or BSH injection. Subsequently, based on the multiple irradiation fields selected in step S1, the angle and position of the corresponding source are added to the input card to perform a Monte Carlo simulation calculation per unit time for each irradiation field, obtaining the physical dose of different radiation components. After the calculation is completed, the physical dose of different radiation components within each voxel grid is weighted and summed according to the relative biological effectiveness (RBE) and comparative biological effectiveness (CBE) to obtain the equivalent photon dose distribution, i.e., the dose distribution per unit time for each irradiation field.

[0058] S3. Combining the irradiation times of the multiple irradiation fields as a generation target, and determining evaluation indicators and constraint conditions of the generation target.

[0059] In fact, the present application generates a suitable BNCT treatment plan by determining the irradiation time of multiple irradiation fields. Therefore, the embodiment of the present application takes the irradiation time combination as the generation target and sets evaluation indicators in order to generate a BNCT treatment plan that meets the evaluation indicators.

[0060] In fact, dose uniformity is one of the important indicators for evaluating the quality of treatment plans. Uniform dose distribution ensures that all areas within the gross tumor volume (GTV) receive sufficient radiation dose, thereby maximizing the tumor control probability (TCP). Increasing the radiation dose received by the GTV can improve its tumor control rate, and uniform dose distribution is the basis for achieving this goal. It can also reduce damage to normal tissue. By limiting high-dose areas within the GTV, the radiation dose to surrounding normal tissues can be minimized, thereby reducing radiotherapy-related side effects. In other words, a good diagnosis and treatment plan must ensure that the target volume receives an adequate dose while reducing radiation to normal tissue.

[0061] However, when using BNCT technology for treatment, if the tumor is located deep, the neutron flux at the tumor is low, and the total dose is therefore low. In order to ensure that all tumor cells receive the prescribed dose, the irradiation time needs to be extended. If a single irradiation field is used, not only will the dose distribution within the tumor area be very uneven, but the normal tissue between the tumor and the neutron source will also receive a high dose due to the increased irradiation time, which may lead to serious complications. If the number of irradiation fields is increased, and the neutron beam is irradiated from different directions, the dose can be "spread" over more normal tissue, reducing the maximum and average doses within normal tissue and the probability of complications.

[0062] Therefore, in order to reduce the side effects of radiotherapy, it is necessary to improve dose uniformity as much as possible. In this way, when the average dose of each voxel grid in the tumor target area is the same, a BNCT treatment plan with better dose uniformity can increase the minimum dose of the tumor target area, ensure that all voxel grids in the tumor target area receive sufficient radiation dose, and reduce radiation to normal tissues, thereby better protecting patients and reducing the probability of radiotherapy complications.

[0063] To this end, in some embodiments, the present application sets the evaluation index as the radiation dose variance of each voxel grid within the tumor target area, and improves the dose uniformity of the BNCT treatment plan by minimizing the evaluation index, thereby ensuring the radiation dose of the tumor target area while reducing side effects on normal tissues.

[0064] Among them, the evaluation index can be expressed by formula (2):

[0065]

[0066] in, is the number of voxel grids in the tumor target area; , expressed as the radiation dose / Gy delivered by all irradiation fields to the i-th voxel grid in the tumor target area in the BNCT treatment plan; n is the number of irradiation fields; Expressed as the dose rate delivered by the jth irradiation field to the i-th voxel grid in the tumor target area / Gy·min -1 ; Expressed as the irradiation time of the jth irradiation field / min; It is expressed as the average radiation dose / Gy of the voxel grid in the tumor target volume. Dose uniformity can be improved by minimizing the radiation dose variance of each voxel grid in the tumor target volume.

[0067] Furthermore, in the process of generating a BNCT treatment plan, it is necessary to adhere to a number of treatment principles in order to achieve evaluation indicators and treatment effects. For example, in order to kill all tumor cells and ensure treatment effects, all voxel grids within the tumor target area should receive the prescribed dose within the total irradiation time. For another example, in order to protect organs at risk and avoid the effects of excessive radiation doses on such organs, the maximum radiation dose received by each organ at risk should not exceed the maximum tolerated dose. In addition, the treatment time also needs to be determined in advance to avoid the effects of excessive treatment time on the patient's body. Therefore, in the process of obtaining a BNCT treatment plan that meets the evaluation indicators, it is also necessary to set constraints to achieve the generation target and meet various treatment principles and treatment effects.

[0068] In some embodiments, the present application sets tumor target area radiation constraint, organ at risk radiation constraint and irradiation time constraint as constraint conditions.

[0069] In some embodiments, the irradiation time constraint includes: the irradiation time constraint is that the sum of the irradiation times of all the irradiation fields should not be greater than a preset total irradiation time, and the irradiation time of each irradiation field should not be less than zero.

[0070] It's important to note that while BNCT treatment plans are essentially a priori, tumor target radiation and OAR radiation are actually a posteriori. Specifically, specific dose distributions based on the BNCT treatment plan can only be obtained after it is generated. Therefore, the tumor target radiation constraints and OAR radiation constraints are determined using an average radiation dose approximation.

[0071] In some embodiments, the radiation constraint of the tumor target area set by the average dose approximation is that the minimum value of the average radiation dose of each voxel grid in the tumor target area under all the irradiation fields should not be less than the prescription dose.

[0072] In some embodiments, the radiation constraint of organs at risk set by the average dose approximation is that the average radiation dose of each organ at risk in all the irradiation fields should not be greater than the maximum tolerated dose.

[0073] In some embodiments, the constraints set above are as shown in formula (3):

[0074]

[0075] in, It is represented as the average dose rate vector of all voxel grids in the tumor target area under n different irradiation fields; Expressed as the average dose rate of all voxel grids in the tumor target area under the jth irradiation field / Gy·min -1 ; It is expressed as the irradiation time of the irradiation field; Expressed as prescription dose / Gy; It is expressed as a matrix of average dose rates of the irradiation field to organs at risk; Expressed as the average dose rate of the jth irradiation field to the kth organ at risk / Gy·min -1 ; Expressed as the dose rate delivered by the jth irradiation field to the i-th voxel grid in the k-th organ at risk / Gy·min -1 ; N represents the number of voxel grids of the organ at risk; t lim Expressed as the upper limit of the total exposure time of all irradiation fields / min; It is expressed as the maximum tolerated dose to organs at risk / Gy.

[0076] In order to ensure that the dose can be accurately delivered to the tumor target area and the patient cannot move arbitrarily during irradiation, the total irradiation time of the treatment plan cannot be set too long. In some embodiments, t lim Set to 60 minutes.

[0077] Furthermore, it should be noted that the above embodiment illustrates only one implementation of constraints. In other words, constraints can be set based on treatment objectives and actual circumstances. For example, the average dose to the tumor target or the average dose to an organ at risk could be used as a constraint, or the threshold required for the specific dose to a particular organ at risk could be used as a constraint. This application does not impose any limitations on this.

[0078] S4. Obtaining the generation target of the minimum value of the evaluation index under the constraint condition based on the barrier algorithm to obtain the optimal irradiation time combination of the multiple irradiation fields.

[0079] As previously described, this application generates a BNCT treatment plan by setting evaluation indicators and constraints to achieve a generation target that satisfies both. In other words, this application essentially solves a constrained optimization problem. Specifically, under the constraints, the generation target that minimizes the evaluation indicator is the desired optimal irradiation time combination. In the above embodiment, the optimal irradiation time combination should achieve optimal dose uniformity.

[0080] To this end, the present application introduces a barrier algorithm to obtain the generation target of the minimum value of the evaluation index under the constraints of the constraints. In some embodiments, the generation target of obtaining the minimum value of the evaluation index under the constraints of the constraints based on the barrier algorithm can be expressed by formula (4):

[0081]

[0082] in, It is expressed as the irradiation time of j irradiation fields, that is, the irradiation time combination of irradiation fields, ; This is the evaluation index shown in Formula 2, which is the objective function to be minimized; represents the wth constraint condition, and there are m constraints in total. For example, in the above embodiment, there are actually 4 constraints in total.

[0083] In fact, due to the inequality constraints in equation (4), the Newton method cannot be used to directly calculate the gradient of the objective function and iteratively solve it. Therefore, in some embodiments, the present application introduces a logarithmic barrier algorithm to convert equation (4) into an unconstrained approximation problem, and then uses the Newton method to solve and gradually approach the optimal solution, that is, to obtain the optimal irradiation time combination for multiple irradiation fields. Figure 3 FIG. 1 shows a flow chart of a BNCT treatment plan generation method based on a barrier algorithm according to an embodiment of the present application, as shown in FIG. Figure 3 As shown, obtaining the generation target with the minimum evaluation index value under the constraint of the constraint condition based on the barrier algorithm includes steps S41 to S42.

[0084] S41. Acquire a generation model based on a barrier algorithm, where the generation model is used to acquire a generation target of a minimum value of the evaluation index that satisfies a constraint condition.

[0085] Barrier algorithms usually need to meet the following conditions:

[0086] (1) When the barrier function is within the definition domain of the original problem in equation (4), the value of the barrier function is small enough and will not affect the objective function in equation (4). to minimize the deviation caused by approximate conversion.

[0087] (2) When the value of the barrier function is large enough at the boundary of the domain of the original problem in Equation (4) or outside the domain of definition, when the solution of the Newton method iteration approaches the boundary of the domain of definition or goes outside the domain of definition, the value of the objective function will become very large, ensuring that the iteration is carried out within the domain of definition.

[0088] To this end, in some embodiments, a logarithmic barrier function is used to obtain a generative model. The standard form of the logarithmic barrier function is shown in formula (5):

[0089]

[0090] Where t represents the barrier algorithm hyperparameter and t>0.

[0091] Among them, the generation model obtained based on the barrier algorithm can be expressed by formula (6):

[0092]

[0093] It can be seen that the generative model actually integrates the evaluation index of the generated target with the constraint conditions, converting the constraint conditions into a part of the whole, thereby converting the equation (4) with the inequality constraint into the equation (6) without the constraint, so as to solve the generative model and obtain the generated target with the minimum value of the evaluation index that satisfies the constraint conditions.

[0094] Furthermore, in fact, the logarithmic barrier function images of different t values ​​are shown in Figure 4. When t is large (see Figure 4 (see the black curve in Figure 4), the value of the logarithmic barrier function within the domain is small enough to have little impact on the original problem of Equation (4). However, when t is small (see the blue curve in Figure 4), the value of the logarithmic barrier function within the domain becomes larger, causing the optimal solution of the approximate transformation Equation (6) to move away from the boundary of the domain, resulting in deviation. Therefore, the value of t will cause a deviation between the optimal solutions of Equation (6) and Equation (4), so it is necessary to continuously update t in subsequent steps to ensure that the optimal solution of the generative model can be used as the optimal solution of Equation (4).

[0095] S42. Iteratively solve the generation model until the approximate error of the solution is not less than a preset threshold, and obtain the optimal irradiation time combination of the multiple irradiation fields.

[0096] The preset threshold is the result of comparing the number of the constraint conditions with the hyperparameters of the barrier algorithm.

[0097] According to the Slater condition, if the objective function in Equation (4) is convex and has a strictly feasible solution, then the optimization problem has a strong duality property. If there is only one feasible solution instead of a strictly feasible solution, then Equation (4) has one and only one solution, and there is no need to discuss its optimal solution and optimal value. Furthermore, if there is no feasible solution, then Equation (4) has no solution, and therefore no optimal solution or optimal value. For clarity, the following description of the situation where there is no optimal solution refers to the situation where there is no feasible solution.

[0098] Furthermore, if Equation (4) has an optimal solution, it must satisfy the Karush-Kuhn-Tucker (KKT) condition. The KKT condition mainly includes the following four parts:

[0099] (1) Stationarity, in the optimal solution At , the derivative of the Lagrangian function of Equation (4) is 0, indicating that it is an extreme value.

[0100] (2) Primal Feasibility, All constraints in formula (4) must be satisfied. The inequality constraint requires within its feasible domain.

[0101] (3) Dual Feasibility: The Lagrange multiplier of the inequality constraint of the Lagrange function of formula (4) must be non-negative, because the Lagrange multiplier is the solution to the dual problem, and the dual problem variables are nonnegative.

[0102] (4) Complementary Slackness At this point, each inequality constraint is "tight" ( ) or "relax" ( ).

[0103] Since there are inequality constraints in Equation (4), the KKT condition satisfied by the optimal solution of Equation (4) can be expressed by Equation (7):

[0104]

[0105] in, is the stationary condition; is the original feasibility condition; is the dual feasibility condition; It is a complementary slack condition.

[0106] Furthermore, the Lagrangian dual function of the optimal solution of Equation (4) is shown in (8):

[0107]

[0108] in, is the Lagrangian function of formula (4); is the solution to the dual problem of Eq. (4).

[0109] Similarly, if Equation (6) has an optimal solution, it should also satisfy the KKT condition, as shown in Equation (9):

[0110]

[0111] in, is the optimal solution of formula (6) at time t.

[0112] Because According to formula (7) and formula (8), we can get: It is a feasible solution to the dual problem of Equation (4).

[0113] When strong duality holds, the optimal solution and optimal value of Equation (4) are the same as the optimal solution and optimal value of the dual problem of Equation (4). Then, the dual interval of Equation (4) can be scaled to obtain Equation (10):

[0114]

[0115] Then, combining formula (8) we can get formula (11):

[0116]

[0117] From formula (11), we can derive formula (12):

[0118]

[0119] Then let the approximate error be , the preset threshold is set to , which is the result of comparing the number of constraints m with the barrier algorithm hyperparameter t. According to (12), it can be considered that the optimal solution of formula (6) at time t is the optimal solution of formula (4). The optimal solution of formula (6) is solved by Newton's method, which is also the optimal solution of the generated model. At this time, the solution is actually the optimal irradiation time combination of multiple irradiation fields required by formula (4). If , then t needs to be updated at this time. The common method is to divide by the update parameter Increase t and update the generated model based on the updated t to iteratively solve the problem and obtain the optimal solution, and then compare and , until the conditions are met and the optimal solution is output. That is, when the solution approximation error is less than a preset threshold, the barrier algorithm hyperparameters are updated to update the preset threshold, and the generative model is updated based on the updated barrier algorithm hyperparameters to re-iterate the solution; wherein the comparison result of the barrier algorithm hyperparameters with the updated parameters is used as the updated barrier algorithm hyperparameters. Figure 5 The process of obtaining the optimal solution of the generative model in an embodiment of the present application is shown.

[0120] It should be noted that, as mentioned above, there may be a situation where there is no feasible solution to formula (4), that is, there is no optimal solution. This means that the generation target of the minimum value of the evaluation index cannot be obtained under the constraints of the constraint conditions based on the barrier algorithm. This means that the constraint conditions need to be readjusted to re-obtain the generation model and continue to iterate the solution. In some embodiments, the inability to obtain the generation target with the minimum value of the evaluation index under the constraints of the constraint conditions means that the tumor target radiation constraint cannot be met while the radiation constraint of the organ at risk is met. In this case, it is necessary to reduce the radiation constraint of the tumor target, reset the radiation constraint of the organ at risk to re-obtain the generation model, and iterate the solution to obtain the optimal solution. Among them, resetting the radiation constraint of the organ at risk is to restore the definition domain reduced by the radiation constraint of the organ at risk.

[0121] It should be noted that in the optimal time combination obtained by the generative model, there may be an irradiation field with an irradiation time of 0, which indicates that the irradiation field at this angle is actually not needed.

[0122] S5. Determine the irradiation time of each irradiation field based on the optimal irradiation time combination, so as to generate a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution.

[0123] As previously mentioned, BNCT treatment plans are a priori. Only after obtaining the optimal irradiation time combination, that is, the irradiation time of different irradiation fields, can the dose distribution of the patient's tumor target and organs at risk be calculated based on this. Specifically, the patient's dose distribution is inferred based on the irradiation time of the multiple irradiation fields and the corresponding unit time dose distribution. The dose distribution of the tumor target and organs at risk is then calculated to determine whether a new optimal solution needs to be generated. This process is then repeated and iterated to obtain a good BNCT treatment plan.

[0124] In some embodiments, generating a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution includes: obtaining the minimum value of the tumor target area radiation and the maximum value of the organ at risk radiation based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution; performing a generation judgment operation based on the minimum value of the tumor target area radiation and the maximum value of the organ at risk radiation, and when the generation judgment operation passes, generating a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution; when the generation judgment operation fails, adjusting the constraint condition to re-acquire the generation model and continue to iterate the solution.

[0125] The generation and judgment operation includes: determining whether the maximum radiation value of the organ at risk is less than the maximum tolerated dose; if not, the generation and judgment operation fails; if so, continuing to determine whether the minimum radiation value of the tumor target area is not less than the prescribed dose; if so, the generation and judgment operation passes; if not, the generation and judgment operation fails.

[0126] Figure 6 The following is a schematic diagram showing the logic judgment of generating a BNCT treatment plan in the embodiment of the present application. Figure 6 As shown, it is determined whether there is an optimal solution. If so, the minimum radiation value of the tumor target area and the maximum radiation value of the organ at risk are obtained based on the irradiation time of the multiple irradiation fields and the corresponding unit time dose distribution. If not, the radiation constraint of the tumor target area is reduced and the radiation constraint of the organ at risk is reset to re-iterate and solve the generation model.

[0127] If there is an optimal solution, the generation judgment operation is performed. First, it is determined whether the maximum radiation of the organ at risk is less than the maximum tolerated dose. If not, the generation judgment operation fails. At this time, the constraint conditions are adjusted, that is, the radiation constraint of the organ at risk is reduced to re-acquire the generation model and continue the iterative solution. If it is less than the maximum tolerated dose, it is further determined whether the minimum radiation of the tumor target area is not less than the prescribed dose. If it is not less than the prescribed dose, the generation judgment operation is passed, and the BNCT treatment plan is generated, that is, BNCT treatment is performed according to the irradiation time of multiple irradiation fields currently determined. If it is less than the prescribed dose, the constraint conditions are adjusted. At this time, it is necessary to further determine whether the tumor target area radiation constraint has been reduced in the process of obtaining the optimal solution. If the tumor target area radiation constraint has been reduced, the tumor target area radiation constraint is reduced again. If the tumor target area radiation constraint has not been reduced, the tumor target area radiation constraint is increased.

[0128] It should be noted that in the above embodiments, lowering the radiation constraint for the tumor target area means lowering the prescribed dose, while raising the radiation constraint for the tumor target area means raising the prescribed dose. Furthermore, lowering the radiation constraint for organs at risk means lowering the maximum tolerated dose, while raising the radiation constraint for organs at risk means raising the maximum tolerated dose.

[0129] When the generation target of the minimum value of the evaluation index that meets the constraint conditions is obtained, that is, the optimal irradiation time combination, at this time, after neutron beam irradiation according to the irradiation time of multiple irradiation fields, the dose distribution in the patient's body can obtain the optimal result in terms of the evaluation index.

[0130] Therefore, the BNCT treatment plan generation method based on the barrier algorithm provided in the embodiment of the present application can ensure that the tumor receives sufficient radiation dose while minimizing damage to surrounding healthy tissues, thereby improving the treatment effect and reducing the side effects of radiotherapy.

[0131] The scope of protection of the BNCT treatment plan generation method based on the barrier algorithm in the embodiment of the present application is not limited to the order of execution of the steps listed in this embodiment. All solutions implemented by adding, subtracting, or replacing steps in the prior art based on the principles of the present application are included in the scope of protection of the present application.

[0132] An embodiment of the present application also provides a BNCT treatment plan generation device based on a barrier algorithm. The BNCT treatment plan generation device based on a barrier algorithm can implement the BNCT treatment plan generation method based on a barrier algorithm described in the present application. However, the implementation device of the BNCT treatment plan generation device based on a barrier algorithm described in the present application includes but is not limited to the structure of the BNCT treatment plan generation device based on a barrier algorithm listed in this embodiment. Any structural deformation and replacement of the prior art made according to the principles of the present application are included in the scope of protection of the present application.

[0133] Figure 7 FIG shows a schematic diagram of the structure of the BNCT treatment plan generation device based on the barrier algorithm according to an embodiment of the present application. Figure 7 As shown, the BNCT treatment plan generation device based on the barrier algorithm includes an acquisition module 41, a calculation module 42, a target module 43, a search 44 and a generation module 45.

[0134] An acquisition module 41 is used to acquire multiple irradiation fields;

[0135] A calculation module 42 is used to calculate the unit time dose distribution of each irradiation field;

[0136] A target module 43 is configured to combine the irradiation times of the plurality of irradiation fields as a generated target, and determine evaluation indicators and constraint conditions for the generated target;

[0137] A second calculation module 44 is configured to search for the generation target of the minimum value of the evaluation index under the constraint condition based on a barrier algorithm to obtain an optimal irradiation time combination for the multiple irradiation fields;

[0138] The generating module 45 is configured to determine the irradiation time of each irradiation field based on the optimal irradiation time combination, so as to generate a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution.

[0139] It should be noted that the structures and principles of the acquisition module 41, calculation module 42, target module 43, search module 44 and generation module 45, as well as the beneficial effects achieved by the device are the same as those in the above embodiment and will not be described in detail here.

[0140] An embodiment of the present application further provides a storage medium having a computer program stored thereon, wherein when the program is executed by a processor, all steps of the barrier algorithm-based BNCT treatment plan generation method of the embodiment are implemented.

[0141] Among them, the specific steps of the BNCT treatment plan generation method based on the barrier algorithm and the beneficial effects obtained by applying the readable storage medium provided in the embodiment of the present application are the same as those in the above embodiment and will not be repeated here.

[0142] Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments can be performed by instructing a processor through a program. The program can be stored in a computer-readable storage medium, which is a non-transitory medium, such as random access memory, read-only memory, flash memory, a hard disk, a solid-state drive, a magnetic tape, a floppy disk, an optical disc, or any combination thereof. The storage medium can be any available medium accessible by a computer, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0143] An embodiment of the present application also provides a terminal. Figure 8 A schematic diagram of the structure of the terminal according to an embodiment of the present application is shown. Figure 8As shown, the terminal 700 of the embodiment of the present application includes: at least one processor 701, a memory 702, at least one network interface 704 and a user interface 706. In addition, the various components in the terminal 700 are coupled together through a bus system 705. It can be understood that the bus system 705 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 705 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, Figure 7 The various buses are all labeled as bus systems. The user interface 706 may include a display, keyboard, mouse, trackball, click gun, keys, buttons, touch pad or touch screen.

[0144] It can be understood that the memory 702 can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. This application is not specifically limited. The memory 702 in the embodiment of the present application is used to store various categories of data to support the operation of the terminal 700. Examples of these data include: any executable program for operating on the terminal 700, such as an operating system 7021 and an application 7022; the operating system 7021 includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic businesses and processing hardware-based tasks. The application 7022 can include various applications, such as a media player (MediaPlayer), a browser (Browser), etc. The implementation of the BNCT treatment plan generation method based on the barrier algorithm provided in the embodiment of the present application can be included in the application 7022.

[0145] The barrier algorithm-based BNCT treatment plan generation method disclosed in the above embodiment of the present application can be applied to the processor 701 or implemented by the processor 701. The processor 701 may be an integrated circuit chip with signal processing capabilities. During implementation, the steps of the above method can be completed by hardware integrated logic circuits in the processor 701 or instructions in software form. The above processor 701 can be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 701 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor 701 can be a microprocessor or any conventional processor, etc.

[0146] In an exemplary embodiment, the terminal 700 may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), and complex programmable logic devices (CPLDs) to execute the aforementioned methods.

[0147] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with a computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0148] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

[0149] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical concepts disclosed in this application shall be covered by the claims of this application.

Claims

1. A BNCT treatment plan generation method based on a barrier algorithm, characterized in that: The method comprises: Acquire multiple irradiation fields; respectively calculating the unit time dose distribution of each irradiation field; The irradiation time combination of the multiple irradiation fields is used as a generation target, and an evaluation index and constraint conditions of the generation target are determined; the evaluation index is set as the radiation dose variance of each voxel grid in the tumor target area; obtaining the generation target of the minimum value of the evaluation index under the constraint condition based on the barrier algorithm to obtain the optimal irradiation time combination of the multiple irradiation fields; The irradiation time of each irradiation field is determined based on the optimal irradiation time combination, so as to generate a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution.

2. The BNCT treatment plan generation method based on the barrier algorithm according to claim 1, characterized in that: The constraint conditions include tumor target area radiation constraint, organ at risk radiation constraint and irradiation time constraint: wherein the tumor target area radiation constraint and the organ at risk radiation constraint are determined by an average radiation dose.

3. The BNCT treatment plan generation method based on the barrier algorithm according to claim 1, characterized in that: The generation goal of obtaining the minimum evaluation index value under the constraints of the constraint conditions based on the barrier algorithm includes: Acquire a generation model based on a barrier algorithm, wherein the generation model is used to obtain a generation target of a minimum value of the evaluation index that satisfies a constraint condition; Iteratively solving the generation model until an approximate error of the solution is not less than a preset threshold, thereby obtaining an optimal irradiation time combination for the multiple irradiation fields; The preset threshold is the result of comparing the number of the constraint conditions with the hyperparameters of the barrier algorithm.

4. The BNCT treatment plan generation method based on the barrier algorithm according to claim 3, characterized in that: The method further comprises: When the solution approximation error is less than a preset threshold, the barrier algorithm hyperparameters are updated to update the preset threshold, and the generation model is updated based on the updated barrier algorithm hyperparameters to re-iterate the solution; wherein, the comparison result of the barrier algorithm hyperparameters and the updated parameters is used as the updated barrier algorithm hyperparameters.

5. The BNCT treatment plan generation method based on the barrier algorithm according to claim 3, characterized in that: The method comprises: If the generative model has no feasible solution, the constraint conditions are adjusted to re-obtain the generative model and continue to iterate the solution.

6. The BNCT treatment plan generation method based on the barrier algorithm according to claim 3, characterized in that: Generating a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution includes: Obtaining a minimum value of radiation to the tumor target area and a maximum value of radiation to the organ at risk based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution; Performing a generation and determination operation based on the minimum value of the tumor target area radiation and the maximum value of the organ-at-risk radiation; When the generation judgment operation is passed, a BNCT treatment plan is generated based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution; When the generation judgment operation fails, the constraint condition is adjusted to re-acquire the generation model and continue to iterate and solve.

7. The BNCT treatment plan generation method based on the barrier algorithm according to claim 6, characterized in that: The generation judgment operation includes: Determine whether the maximum radiation value of the organ at risk is less than the maximum tolerated dose; if not, the generation determination operation fails; if so, continue to determine whether the minimum radiation value of the tumor target area is not less than the prescribed dose; If so, the generation judgment operation is passed; if not, the generation judgment operation is failed.

8. A BNCT treatment plan generation device based on a barrier algorithm, characterized in that: The device comprises: An acquisition module, used for acquiring multiple irradiation fields; A calculation module, used for respectively calculating the unit time dose distribution of each irradiation field; a target module, configured to combine the irradiation times of the multiple irradiation fields as a generated target, and determine evaluation indicators and constraints for the generated target; the evaluation indicator is set as the radiation dose variance of each voxel grid within the tumor target area; A search module, configured to search for the generation target of the minimum value of the evaluation index under the constraint of the constraint condition based on a barrier algorithm, so as to obtain an optimal irradiation time combination for the multiple irradiation fields; A generating module is configured to determine the irradiation time of each irradiation field based on the optimal irradiation time combination, so as to generate a BNCT treatment plan based on the irradiation time of all irradiation fields and the corresponding unit time dose distribution.

9. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the BNCT treatment plan generation method based on the barrier algorithm according to any one of claims 1 to 7 is implemented.

10. A terminal, characterized in that: The terminal comprises a processor and a memory, wherein the memory is communicatively connected to the processor; the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the terminal performs the BNCT treatment plan generation method based on the barrier algorithm as described in any one of claims 1 to 7.

Images