Boron neutron capture therapy plan generation method and device, storage medium and terminal

Abstract

The invention provides a boron neutron capture therapy plan generation method and device, a storage medium and a terminal. The method comprises the following steps: acquiring a target region biological effect dose target function, a healthy tissue biological effect dose target function, a dose uniformity target function, a treatment time target function and a target region and healthy tissue biological effect dose ratio target function as a target function set; acquiring a treatment time parameter, a treatment angle parameter and a target region boron concentration parameter as an optimization parameter set; and based on the target function set and the optimization parameter set, performing iterative optimization by using a preset multi-target optimization algorithm to obtain a corresponding Pareto frontier as a treatment plan. According to the method, the optimal treatment plan under different boron neutron capture therapy optimization parameter configurations can be generated, and diversified treatment requirements of different patients can be met.

Description

Technical Field

[0001] The invention belongs to the field of medical technology and relates to a method for generating a boron neutron capture therapy plan, in particular to a method and device for generating a boron neutron capture therapy plan, a storage medium and a terminal. Background Art

[0002] Boron Neutron Capture Therapy (BNCT) is a binary targeted tumor radiotherapy method that mainly utilizes the combined effects of epithermal neutron beams and tumor-friendly boron-containing drugs. Figure 1 A schematic diagram of boron neutron capture therapy is shown in FIG. Figure 1 As shown in the figure, the basic principle is: introduce boron-containing compounds into the patient's body. Since these compounds can be absorbed by tumor cells, they will quickly accumulate in tumor cells once they enter the body, and are rarely distributed in other normal tissues; then, use neutron beams to irradiate the tumor site, so that neutrons and B accumulated in tumor cells undergo boron neutron capture reaction 10 B(n,α) 7 Li, when B captures a neutron, it will split into heavy particles 7 Li and alpha particles, which irradiate and destroy tumor cells, and 7 The path of Li and α particles can minimize the damage to surrounding healthy tissues. Therefore, BNCT has the advantages of less damage to normal cells, a wide range of tumor diseases, short and flexible treatment course, and relatively low treatment cost.

[0003] In the prior art, the BNCT (boron neutron capture therapy) radiotherapy planning system (TPS) usually uses the boron concentration information provided by PET images and the spatial structure of the tissue provided by CT images to formulate a treatment plan. The setting of planning parameters includes multiple irradiation fields, field directions and irradiation time. However, the treatment plan formulated in this way can only meet the basic treatment requirements, and it is difficult to achieve the best treatment effect. For example, if the shortest treatment time is guaranteed, it is difficult to ensure the optimal overall treatment effect. In addition, strategies such as setting double irradiation fields or reasonably allocating irradiation time lack constraint optimization for treatment effects during the formulation process, and cannot generate a treatment plan with practical results. At the same time, the methods of the prior art lack attention to the specific needs of different patients. For example, patients with larger target areas may be more suitable for optimization in the direction of increasing the minimum dose of the target area, while patients with irregular target shapes may be more inclined to optimize the target dose uniformity.

[0004] Therefore, the existing technology cannot meet the diverse practical treatment needs of boron neutron capture therapy. Summary of the invention

[0005] The object of the present invention is to provide a method and device for generating a boron neutron capture therapy plan, a storage medium and a terminal, so as to solve the technical problem that the prior art cannot meet the diversified actual treatment needs of boron neutron capture therapy.

[0006] In a first aspect, the present invention provides a method for generating a boron neutron capture therapy plan, comprising:

[0007] Obtaining a target area biological effect dose objective function, a healthy tissue biological effect dose objective function, a dose uniformity objective function, a treatment time objective function, and a target area and healthy tissue biological effect dose ratio objective function as an objective function set;

[0008] Obtaining treatment time parameters, treatment angle parameters and target area boron concentration parameters as an optimization parameter set;

[0009] Based on the objective function set and the optimization parameter set, using a preset multi-objective optimization algorithm to perform iterative optimization to obtain a corresponding Pareto front as a treatment plan;

[0010] Each solution in the Pareto front represents a treatment plan.

[0011] In one embodiment of the present invention, the target area and healthy tissue biological effect dose ratio objective function is: the ratio of the target area tissue boron concentration to the normal tissue biological effect dose.

[0012] In one embodiment of the present invention, the comprehensive objective function of the preset multi-objective optimization algorithm is the weighted sum of all the objective functions, the weight of each objective function is dynamically updated during iteration, and the sum of the weights of all the objective functions is constant to 1.

[0013] In one embodiment of the present invention, each update includes:

[0014] All weights are adjusted according to the ratio of the current value of each objective function to the ideal value, so that all weights in the comprehensive objective function meet the preset condition, and the preset condition is that the weight size is proportional to the preset indicator size;

[0015] The preset indicator is the absolute value of the difference between the current value of the objective function and the ideal value minus 1.

[0016] In one embodiment of the present invention, each weight is dynamically updated as follows:

[0017]

[0018] in, represents the preset weight, Indicates the preset weight value of the previous generation, represents the ideal value of the objective function, represents the current value of the objective function, and γ represents the preset adjustment factor.

[0019] In one embodiment of the present invention, the initial weight of the target area biological effect dose objective function is the inverse of the dose variation coefficient within the target area;

[0020] The initial weight of the healthy tissue biological effect dose objective function is the ratio of the actual biological effect dose of the healthy tissue to the safe dose limit of the healthy tissue;

[0021] The initial weight of the dose uniformity objective function is the ratio of the surface area of ​​the target area to the volume of the target area;

[0022] The initial weight of the treatment time objective function is the ratio of the estimated time of the current treatment plan to the maximum treatment time allowed by the patient or the device;

[0023] The initial weight of the target area and healthy tissue biological effect dose ratio objective function is the reciprocal of the difference between the target area biological effect dose and the healthy tissue biological effect dose.

[0024] In one embodiment of the present invention, the preset multi-objective optimization algorithm includes the NSGA-II algorithm and the MOEA / D algorithm.

[0025] In a second aspect, the present invention further provides a boron neutron capture therapy plan generating device, characterized in that it comprises:

[0026] An objective function establishment module is used to obtain a target area biological effect dose objective function, a healthy tissue biological effect dose objective function, a dose uniformity objective function, a treatment time objective function, and a target area and healthy tissue biological effect dose ratio objective function as an objective function set;

[0027] An optimization parameter acquisition module is used to acquire treatment time parameters, treatment angle parameters and target area boron concentration parameters as an optimization parameter set;

[0028] A multi-objective optimization module, used to perform iterative optimization based on the objective function set and the optimization parameter set using a preset multi-objective optimization algorithm to obtain a corresponding Pareto front as a treatment plan;

[0029] Each solution in the Pareto front represents a treatment plan.

[0030] In a third aspect, the present invention further provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for generating a boron neutron capture therapy plan as described above.

[0031] In a fourth aspect, the present invention further provides a terminal, comprising a processor and a memory, wherein the memory is communicatively connected to the processor;

[0032] 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 boron neutron capture therapy plan generation method as described above.

[0033] As described above, the boron neutron capture therapy plan generation method and device, storage medium and terminal of the present invention have the following beneficial effects:

[0034] The present invention constructs an objective function set and an optimization parameter set, generates a Pareto chart frontier based on a preset multi-objective optimization algorithm, and provides an optimal treatment plan under multiple different optimization parameter configurations. It has a balance between diverse optimization parameters and treatment plans, which is conducive to meeting the diverse treatment needs of different patients, and significantly improves the efficiency of generating boron neutron capture therapy treatment plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic diagram of boron neutron capture therapy is shown.

[0036] Figure 2 A schematic flow chart of a method for generating a boron neutron capture therapy plan according to an embodiment of the present invention is shown.

[0037] Figure 3 A schematic diagram of a process of obtaining a Pareto front by using the NSGA-II algorithm as a preset multi-objective optimization algorithm in the boron neutron capture therapy plan generation method described in an embodiment of the present invention is shown.

[0038] Figure 4 A schematic diagram of the Pareto front results obtained under single-field irradiation in the boron neutron capture therapy plan generation method described in an embodiment of the present invention is shown.

[0039] Figure 5 The schematic diagram of the structure of the boron neutron capture therapy plan generating device according to the embodiment of the present invention is shown.

[0040] Figure 6 A schematic diagram of the structure of a terminal according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0041] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0042] The principles and implementation methods of the boron neutron capture therapy plan generation method and device, storage medium and terminal of this embodiment will be explained in detail below, so that those skilled in the art can understand the boron neutron capture therapy plan generation method, device, storage medium and terminal of this embodiment without creative work.

[0043] In order to solve the above technical problems existing in the prior art, an embodiment of the present invention provides a method for generating a boron neutron capture therapy plan.

[0044] Figure 2 FIG. 1 is a flow chart showing a method for generating a boron neutron capture therapy plan according to an embodiment of the present invention, with reference to FIG. Figure 2 As shown, the method for generating a boron neutron capture therapy plan according to an embodiment of the present invention comprises the following steps:

[0045] Step S100: Obtaining a target area biological effect dose objective function, a healthy tissue biological effect dose objective function, a dose uniformity objective function, a treatment time objective function and a target area and healthy tissue biological effect dose ratio objective function as an objective function set.

[0046] The present invention generates a treatment plan for boron neutron capture therapy (BNCT) based on a multi-objective optimization algorithm by defining multiple objective functions and optimizing parameters. The objective function set includes multiple objective functions, each of which represents a different optimization goal.

[0047] Specifically, the target area biological effect dose objective function is:

[0048] Among them, fitness target It represents the target area biological effect dose objective function, Dose target Indicates the minimum biological effect dose expected in the target area, Dose min Indicates the minimum biological effect dose of the actual target area;

[0049] The objective function of the biological effect dose of healthy tissue is:

[0050] Among them, fitness organMaxDose represents the objective function of the biological effect dose of healthy tissue, Dose safeIndicates the maximum safe dose for healthy tissues, Dose organ,max Indicates the actual maximum dose to healthy tissue

[0051] The dose uniformity objective function is:

[0052] Among them, fitness DoseUniformity Indicates that Dose max Indicates the maximum dose of the target area, Dose min Indicates the minimum dose of the target area, Dose mean Represents the average dose in the target area.

[0053] The treatment time objective function is:

[0054] Among them, fitness TreatmentTime represents the treatment time objective function, Time actual Indicates the actual treatment time, Time max Indicates the maximum treatment time.

[0055] The objective function of the biological effect dose ratio between the target area and healthy tissue is:

[0056] Among them, fitness Boron represents the objective function of the biological effect dose ratio between the target area and healthy tissue, C target represents the mean biological effect dose of the target area, C organ Represents the mean biological effect dose of normal tissue.

[0057] In a specific embodiment of step S100, the objective function of the biological effect dose ratio of the target area to the healthy tissue is: the ratio of the boron concentration of the target area tissue to the biological effect dose of the normal tissue. By introducing this objective function and taking the ratio of the two as the objective function, the treatment plan can be effectively optimized to maximize the dose in the target area while limiting damage to healthy tissue. For example, increasing the biological effect dose of the target area may cause more radiation to healthy tissue, while reducing the dose to healthy tissue may reduce the therapeutic effect on the target area. The use of the dose ratio objective function can establish a balance between the two and give priority to optimizing the key indicators of the treatment plan, that is, the comprehensive embodiment of "treatment effect / safety". By optimizing the dose ratio between the target area and healthy tissue, it helps to generate a more reliable treatment plan and reduce the risk of complications.

[0058] Step S200: Acquire treatment time parameters, treatment angle parameters and target area boron concentration parameters as an optimization parameter set.

[0059] The parameters in the optimization parameter set are variables that need to be adjusted during the optimization process. The treatment time parameter represents the duration of treatment; the treatment angle parameter represents the irradiation angle of radiation during treatment. By adjusting the irradiation angle, the dose distribution of the target area and healthy tissue can be changed; the target area boron concentration parameter represents the distribution concentration of boron elements in the target area. The boron concentration affects the effect of boron neutron capture therapy. By adjusting the parameters in the optimization parameter set, a balance can be achieved between efficacy and safety to ensure the best treatment effect.

[0060] Step S300: Based on the objective function set and the optimization parameter set, an iterative optimization is performed using a preset multi-objective optimization algorithm to obtain a corresponding Pareto front as a treatment plan.

[0061] The embodiment of the present invention uses a multi-objective optimization algorithm to generate a treatment plan. The objective function set defines the optimization goal, that is, the multiple treatment effects that are expected to be achieved, and the optimization parameter set defines the variables to be adjusted. By adjusting these optimization parameters, a solution that meets the requirements of the objective function is found. The preset multi-objective optimization algorithm is a pre-selected multi-objective optimization algorithm. The multi-objective optimization algorithm is to solve a balanced solution under multiple optimization goals. The balanced solution can optimize other goals as much as possible on the basis of satisfying some optimization goals. The Pareto optimal solution is a solution that cannot be further optimized on one goal without making another goal worse, and the Pareto front is a set of all Pareto optimal solutions. Multi-objective optimization algorithms include weighted sum method, genetic algorithm, particle swarm optimization algorithm, simulated annealing algorithm, target programming algorithm, etc. The Pareto front obtained in step S300 is the treatment plan solution set generated by the embodiment of the present invention. Each solution in the Pareto front represents a treatment plan, which provides a scientific and flexible decision-making basis for treatment.

[0062] In a specific implementation of step S300, the present invention performs multi-objective optimization based on weight updating. In the prior art of multi-objective optimization, multiple objective functions are combined into a comprehensive objective function for optimization, wherein each objective function has a specific weight, but the prior art usually optimizes with a fixed weight coefficient. In this embodiment, the comprehensive objective function of the preset multi-objective optimization algorithm is the weighted sum of all objective functions, which is shown in the following formula:

[0063] fitness all =a1fitness TargetMinDose +a2fitness OrganMaxDose +a3fintness DoseUniformity +a4fitness TreatmentTime +a5fitness Boron

[0064] Among them, fitnessall represents the comprehensive objective function of the preset multi-objective optimization algorithm, and a1-a5 represent the weights of the objective functions. These weights are dynamically updated in iterations, and all weights are normalized after each update so that the sum of the weights of all objective functions is constant to 1. Specifically, the normalization method is:

[0065]

[0066] Among them, a new represents the updated weight, a normalized Represents the normalized weight.

[0067] Those skilled in the art can flexibly set different weight update methods based on the embodiments of the present invention. It should be noted that as the value of the objective function changes during the optimization process, the method of dynamically adjusting the weight is conducive to improving the ability to explore the search space to improve the effect of the Pareto frontier, and the output treatment plan quality is better.

[0068] Optionally, the present invention adjusts the weight based on the deviation of the current value of the objective function from the ideal value. Specifically, each update includes: obtaining the current value and ideal value of all objective functions, and adjusting all weights according to the ratio of the current value of each objective function to the ideal value, so that all weights in the comprehensive objective function meet the preset conditions. The preset condition is: the size of the weight is proportional to the size of the preset indicator, and the preset indicator is the absolute value of the difference between the ratio of the current value of the objective function and the ideal value minus 1. That is, taking the ratio of the current value to the ideal value as the deviation, the larger the deviation from 1 (that is, the larger the absolute value of the difference between the ratio minus 1), the higher the priority of the objective function in the iteration. Take A and B as two examples to illustrate, if the deviation of objective function A is greater than that of objective function B, then increase the weight of objective function A and reduce the weight of objective function B. The updated weight must meet the preset indicator. The higher the preset indicator, the higher the weight.

[0069] The purpose of the above settings is to release more optimization space for objective functions with larger deviations during the iteration process of the multi-objective optimization algorithm. Objective functions with larger deviations are given priority during the optimization process, making the iterative optimization more directional and helping to quickly approach the Pareto frontier.

[0070] In a specific dynamic update method, the present invention updates the weights with reference to the following formula:

[0071]

[0072] in, represents the preset weight, Indicates the preset weight value of the previous generation, represents the ideal value of the objective function, Represents the current value of the objective function, γ represents the preset adjustment factor, and the preset weight is any weight in the comprehensive objective function. It should be noted that the above formula updates the weight accordingly through the actual difference between the current value of the objective function and the ideal value, ensuring that the iterative optimization focuses more on the optimization of the objective function that is far from the ideal target value, and ensures the overall reliability of the treatment plan. The updated value of the weight is more accurate. At the same time, the preset adjustment factor is used to control the amplitude of the weight adjustment. By setting unrecognized preset adjustment factors, the adaptability and flexibility of the boron neutron capture therapy plan generation method of the embodiment of the present invention can be improved to adapt to the needs of different optimization problems. Optionally, the preset adjustment factor is any value in the interval [0.1, 0.2].

[0073] In a specific embodiment, the present invention sets the initial weights of different objective functions in the following manner. Specifically, the initial weight of the target area biological effect dose objective function is the inverse of the dose variation coefficient within the target area, that is:

[0074]

[0075] Where a1 represents the initial weight of the target area biological effect dose objective function, CV 靶区剂量 Indicates the coefficient of variation of the dose within the target area. The coefficient of variation of the dose within the target area is the ratio of the dose standard deviation to the dose mean. The higher the coefficient of variation, the more uneven the dose distribution in the target area, and the higher a1. When the coefficient of variation approaches 0, the dose coverage is uniform.

[0076] The initial weight of the healthy tissue biological effect dose objective function is the ratio of the actual biological effect dose of healthy tissue to the safe dose limit of healthy tissue, that is:

[0077]

[0078] Among them, BED 健康组织 The actual biological effect dose of healthy tissue, BED 限制 Indicates the safe dose limit for healthy tissues. If BED 健康组织 The closer to BED 限制 , the larger the weight a2 is, the purpose is to give priority to protecting healthy tissues in the next iteration.

[0079] The initial weight of the dose uniformity objective function is the ratio of the surface area of ​​the target area to the volume of the target area, that is:

[0080]

[0081] Where a3 represents the dose uniformity objective function, A 靶区 represents the surface area of ​​the target area, V 靶区Represents the volume of the target area. The more complex (more irregular) the target area shape is, the greater the weight a3 is, and its purpose is to give priority to optimizing dose uniformity in the next iteration.

[0082] The initial weight of the treatment time objective function is the ratio of the estimated time of the current treatment plan to the maximum treatment time allowed by the patient or device, that is:

[0083]

[0084] Among them, T 治疗 represents the estimated time of the current treatment plan, T 最大可耐受 Indicates the maximum treatment time, which has different time values ​​depending on the patient or device. The closer the estimated time of the current treatment plan is to the maximum treatment time, the greater the weight a4 is.

[0085] The initial weight of the target area and healthy tissue biological effect dose ratio objective function is the inverse of the difference between the target area biological effect dose and the healthy tissue biological effect dose, that is:

[0086]

[0087] Among them, C B,靶区 represents the target area biological effect dose, C B,健康组织 It represents the biological effect dose of healthy tissue. The smaller the difference between the biological effect dose of target area and healthy tissue, the greater the weight a5. Its purpose is to prioritize the optimization of biological effect dose distribution.

[0088] The above initial weights are initial weight values ​​generated based on actual dose data along the optimization target direction. This generation method is conducive to guiding the weights of the next iteration. The overall iteration direction will be more focused. Therefore, the convergence speed of iteratively obtaining the optimal value will be faster.

[0089] Taking specific data as an example, the above initial weights and weight update methods are described as follows:

[0090] Assumptions: CV target dose is 0.2; BED of healthy tissue is 7 Gy-eq, and the limit is 8 Gy-eq; the target surface area is 500 cm 2 , volume is 200cm 3 ; The treatment time is 25 minutes, and the upper tolerance limit is 30 minutes; the target area biological effect dose is 25Gy-eq, and the healthy tissue biological effect dose is 5Gy-eq.

[0091] The initial weight coefficients are calculated as a1=5, a2=0.875, a3=2.5, a4=0.833, and a5=0.05. After normalization, a1=0.54, a2=0.095, a3=0.27, a4=0.090, and a5=0.005. In the first generation of optimization, the following mean objective function values ​​of the current population are calculated (all objective values ​​are 1): f1=0.85, f2=0.8, f3=0.9, f4=0.85, and f5=0.95.

[0092] The process of updating weights is as follows:

[0093] Target dose coverage weight (a1):

[0094] Healthy tissue protection weight (a2):

[0095] Dose uniformity weight (a3):

[0096] Treatment time weight (a4):

[0097] Boron concentration ratio weight (a5):

[0098] Re-normalization calculation, for example, a1 = (0.5481 / (0.5481 + 0.0969 + 0.2727 + 0.09135 + 0.005025)) = 0.540, the normalization method of other updated weights is the same as a1. After that, the weight coefficient is updated in each iteration, and the iteration direction of the comprehensive objective function value is close to 1, so as to achieve personalized treatment plan optimization for the complex treatment characteristics of BNCT.

[0099] Figure 4 FIG. 4 is a schematic diagram showing the Pareto front results obtained in the case of single-field irradiation in the method for generating a boron neutron capture therapy plan according to an embodiment of the present invention, with reference to FIG. Figure 4 As shown, Objective1 is the treatment time, Objective2 is the treatment angle, and Objective3 is the boron concentration in the target area. The optimization result contains multiple points, each of which represents an optimal solution, and each point is the optimal treatment plan under the current number of iterations.

[0100] Optionally, the preset multi-objective optimization algorithms include the NSGA-II algorithm and the MOEA / D algorithm. The NSGA-II algorithm is a multi-objective optimization algorithm based on a genetic algorithm, which efficiently searches for a uniformly distributed Pareto front through non-dominated sorting, crowding calculation, selection and evolution. Its advantage in the scenario of the present invention is that it can simultaneously process multiple objective functions in the embodiment of the present invention to generate corresponding treatment plans; MOEA / D (multi-objective evolutionary algorithm based on decomposition) is an algorithm that optimizes by decomposing a multi-objective problem into multiple single-objective sub-problems, and obtains the Pareto front based on target decomposition, neighborhood search and co-evolution. Its advantage in the scenario of the present invention is that it has a high convergence speed. According to actual needs, those skilled in the art may also select other suitable multi-objective optimization algorithms as the preset multi-objective optimization algorithms.

[0101] In a specific implementation, the NSGA-II algorithm is used as a preset multi-objective optimization algorithm to obtain the Pareto frontier. Figure 3 FIG. 1 shows a schematic diagram of a process of obtaining the Pareto frontier using the NSGA-II algorithm as a preset multi-objective optimization algorithm in the boron neutron capture therapy plan generation method according to an embodiment of the present invention, referring to FIG. Figure 3 As shown, the specific steps include:

[0102] Step S301: Obtain an initial population.

[0103] Specifically, at least one individual is generated based on the treatment parameters (beam angle, beam intensity, irradiation time, target area boron concentration) of the initial treatment plan as an initial population. For example, a treatment plan formulated by a doctor is used as an individual in the initial population to complete population initialization, and the initial population is the result of population initialization.

[0104] Optionally, the method of obtaining the initial population includes a random generation method and a heuristic strategy method. The random generation method is to establish a field angle range according to the concentration distribution of the PET image 10B and / or the geometric center of the CT image, and randomly generate the irradiation field direction within the field angle range. The random generation method can traverse all possibilities; the heuristic strategy is to set the value of the treatment parameter (such as the beam angle) to a preset interval based on historical treatment data. This method can calculate a higher quality solution more quickly through the empirical knowledge of historical data, but it may limit the search space, so that the global optimum cannot be achieved in some cases.

[0105] After the initial population is generated, the Monte Carlo (MCX) simulation is performed on the beam angle, beam intensity, irradiation time, target area boron concentration and other parameters of each individual to obtain the dose results. The dose results are then statistically analyzed and calculated to obtain the first-generation objective function values ​​corresponding to all objective functions in the objective function set.

[0106] Step S302: Perform fast non-dominated sorting, selection, crossover and mutation based on the initial population to obtain a child population, and merge the child population with the parent population to form a new population.

[0107] All objective functions in the objective function set are used as fitness functions, the fitness of individuals in the initialization population is calculated, and all individuals in the initialization population are non-dominated sorted according to the fitness. Based on the function value calculated by the above fitness function, two random individuals in the initialization population are compared to obtain the comparison result, and the comparison result is analyzed by the non-dominated sorting method to screen the target individual. Based on the target individual, the corresponding offspring population can be obtained by performing selection, crossover and mutation, and the offspring population is merged with its parent population (the parent population of the first generation offspring population is also the initial population) as a new population. This method provides doctors with optimization solutions with different focuses, which is conducive to outputting the best solution that meets actual needs, for example, outputting individuals (that is, treatment plans) that focus more on target dose coverage, or individuals that focus more on healthy tissue protection.

[0108] Step S303: Perform fast non-dominated sorting and crowding calculation on the new population in sequence to obtain a new parent population, and obtain a new child population corresponding to the new parent population through crossover and mutation.

[0109] For all individuals in the new population, the above objective function is used as the fitness function, and the corresponding objective function value is calculated. The individuals are sorted from high to low according to the objective function value to obtain the non-dominated level of each individual. If two individuals are in the same non-dominated level, the crowding distances of the two are compared to obtain the individual with the larger crowding distance as the selected individual. Repeat the above process until the scale of the selected individuals reaches the preset scale, and the set of individuals of the preset scale is used as the new parent population. Specifically, the crowding calculation refers to the following formula:

[0110]

[0111] Among them, D i represents the crowding distance of the i-th individual, M represents the number of objective functions, represents the objective function value of the i+1th individual on the mth objective function, represents the objective function value of the i-1th individual on the mth objective function, represents the maximum objective function value of the mth objective function, Represents the minimum objective function value of the mth objective function.

[0112] After obtaining the new parent population, the next generation population of the new parent population is used as the new child population through crossover and mutation operations. The crossover operation simulates binary crossover (SBX) to perform crossover operations on individuals in the new parent population to generate new individuals and form offspring, combining the advantages of the parent solution to generate offspring and expand the search space of the solution; the mutation operation performs a small range mutation on the genes (i.e., treatment parameters) of the new individuals generated above to generate new individuals to increase the diversity of the population.

[0113] Optionally, the method of individual selection is tournament selection, which ensures that high-quality solutions (individuals with low non-dominated levels and high crowding distances) have a higher probability of entering the next generation, simplifies the complexity of individual selection, and effectively maintains the diversity of solutions in combination with the crowding distance.

[0114] In a specific implementation of step S303, an adaptive crossover rate is used for crossover operation, and an adaptive mutation rate is used for mutation operation to adapt to the requirements of different search stages, better balance global search and local search, and improve the efficiency of the algorithm and the quality of the solution.

[0115] Step S304: determine whether the new offspring population meets the iteration termination condition. If so, output the new offspring population as the Pareto frontier; otherwise, merge the new offspring population with the new parent population as a new population and iterate using step S303.

[0116] Specifically, the iteration termination condition is that the iteration reaches the maximum number of iterations or the convergence of the solution meets the termination condition. For example, when the iteration reaches 100 times, or when the Pareto front changes less than a preset threshold in two iterations, it can be determined that the convergence of the solution meets the termination condition. If the iteration termination condition is not met, the new offspring population is merged with the new parent population and substituted as a new population into step S303 for repeated calculation until the iteration termination condition is met, and the final new offspring population is output as the Pareto front.

[0117] It should be noted that, based on the introduction of the steps of obtaining the Pareto frontier using the NSGA-II algorithm as the preset multi-objective optimization algorithm in this embodiment, those skilled in the art can select other suitable multi-objective optimization algorithms as the preset multi-objective optimization algorithm according to actual needs, and the present invention will not go into details here.

[0118] The protection scope of the boron neutron capture therapy plan generation method of the embodiment of the present invention is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present invention are included in the protection scope of the present invention.

[0119] The boron neutron capture therapy plan generation method of the embodiment of the present invention generates a Pareto chart frontier based on a preset multi-objective optimization algorithm by constructing an objective function set and an optimization parameter set, and provides an optimal treatment plan under multiple different optimization parameter configurations. At the same time, by introducing the target area and healthy tissue biological effect dose ratio objective function, it is possible to maximize the dose in the target area while limiting the damage to healthy tissue. In addition, the present invention proposes a method for dynamically updating weights during the optimization process of the boron neutron capture therapy plan, giving priority to the objective function with a large deviation during the optimization process, so that the iterative optimization is more directional and helps to quickly approach the Pareto frontier.

[0120] In order to solve the above technical problems existing in the prior art, an embodiment of the present invention further provides a boron neutron capture therapy plan generating device.

[0121] Figure 5 FIG. 1 shows a schematic diagram of the structure of a boron neutron capture therapy plan generating device according to an embodiment of the present invention, referring to FIG. Figure 5 As shown, the boron neutron capture therapy plan generating device according to the embodiment of the present invention comprises:

[0122] An objective function establishment module is used to obtain a target area biological effect dose objective function, a healthy tissue biological effect dose objective function, a dose uniformity objective function, a treatment time objective function, and a target area and healthy tissue biological effect dose ratio objective function as an objective function set;

[0123] An optimization parameter acquisition module is used to acquire treatment time parameters, treatment angle parameters and target area boron concentration parameters as an optimization parameter set;

[0124] A multi-objective optimization module, used to perform iterative optimization based on the objective function set and the optimization parameter set using a preset multi-objective optimization algorithm to obtain a corresponding Pareto front as a treatment plan;

[0125] Each solution in the Pareto front represents a treatment plan.

[0126] The boron neutron capture therapy plan generation device of the embodiment of the present invention constructs an objective function set and an optimization parameter set, generates a Pareto chart frontier based on a preset multi-objective optimization algorithm, and provides optimal treatment plans under multiple different optimization parameter configurations.

[0127] In order to solve the above technical problems existing in the prior art, an embodiment of the present invention further provides a storage medium on which a computer program is stored, characterized in that when the program is executed by a processor, all steps of the boron neutron capture therapy plan generation method of the embodiment are implemented.

[0128] The specific steps of the method for generating a boron neutron capture therapy plan and the beneficial effects obtained by applying the readable storage medium provided in the embodiment of the present invention are the same as those in the above embodiment, and will not be described in detail here.

[0129] A person of ordinary skill in the art can understand that all or part of the steps in the method for implementing the above-mentioned embodiment can be completed by instructing a processor through a program, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state hard disk, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above-mentioned storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a digital video disc (DVD)), or a semiconductor medium (for example, a solid-state disk (SSD)), etc.

[0130] In order to solve the above technical problems existing in the prior art, an embodiment of the present invention further provides a terminal. Figure 6 A schematic diagram of the structure of the terminal according to an embodiment of the present invention is shown, referring to Figure 6 As shown, the terminal of the embodiment of the present invention includes a processor and a memory, and the memory and the processor are communicatively connected; the memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory, so that the terminal executes all steps of the boron neutron capture therapy plan generation method of the above embodiment.

[0131] The specific steps of the method for generating a boron neutron capture therapy plan and the beneficial effects obtained by applying the terminal provided by the embodiment of the present invention are the same as those in the above embodiment, and will not be described in detail here.

[0132] It should be noted that the memory may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage. Similarly, the processor may also be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0133] Although the embodiments disclosed in the present invention are as above, the contents described are only embodiments adopted for facilitating the understanding of the present invention and are not intended to limit the present invention. Any technician in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present invention, but the protection scope of the present invention shall still be subject to the scope defined in the attached claims.

Claims

1. A method for generating a boron neutron capture therapy plan, comprising: Obtaining a target area biological effect dose objective function, a healthy tissue biological effect dose objective function, a dose uniformity objective function, a treatment time objective function, and a target area and healthy tissue biological effect dose ratio objective function as an objective function set; Obtaining treatment time parameters, treatment angle parameters and target area boron concentration parameters as an optimization parameter set; Based on the objective function set and the optimization parameter set, using a preset multi-objective optimization algorithm to perform iterative optimization to obtain a corresponding Pareto front as a treatment plan; Each solution in the Pareto front represents a treatment plan.

2. The method for generating a boron neutron capture therapy plan according to claim 1, characterized in that: The target area and healthy tissue biological effect dose ratio objective function is: the ratio of the target area tissue boron concentration to the normal tissue biological effect dose.

3. The method for generating a boron neutron capture therapy plan according to claim 1, characterized in that: The comprehensive objective function of the preset multi-objective optimization algorithm is the weighted sum of all the objective functions, the weight of each objective function is dynamically updated during iteration, and the sum of the weights of all the objective functions is constant to 1.

4. The method for generating a boron neutron capture therapy plan according to claim 3, characterized in that: Each update includes: All weights are adjusted according to the ratio of the current value of each objective function to the ideal value, so that all weights in the comprehensive objective function meet the preset condition, and the preset condition is that the weight size is proportional to the preset indicator size; The preset indicator is the absolute value of the difference between the current value of the objective function and the ideal value minus 1.

5. The method for generating a boron neutron capture therapy plan according to claim 4, characterized in that: The dynamic update method of each weight is: in, represents the preset weight, Indicates the preset weight value of the previous generation, represents the ideal value of the objective function, represents the current value of the objective function, and γ represents the preset adjustment factor.

6. The method for generating a boron neutron capture therapy plan according to claim 1, characterized in that: The initial weight of the target area biological effect dose objective function is the inverse of the dose variation coefficient within the target area; The initial weight of the healthy tissue biological effect dose objective function is the ratio of the actual biological effect dose of the healthy tissue to the safe dose limit of the healthy tissue; The initial weight of the dose uniformity objective function is the ratio of the surface area of ​​the target area to the volume of the target area; The initial weight of the treatment time objective function is the ratio of the estimated time of the current treatment plan to the maximum treatment time allowed by the patient or the device; The initial weight of the target area and healthy tissue biological effect dose ratio objective function is the reciprocal of the difference between the target area biological effect dose and the healthy tissue biological effect dose.

7. The method for generating a boron neutron capture therapy plan according to claim 1, characterized in that: The preset multi-objective optimization algorithms include NSGA-II algorithm and MOEA / D algorithm.

8. A boron neutron capture therapy plan generation device, characterized in that: include: An objective function establishment module is used to obtain a target area biological effect dose objective function, a healthy tissue biological effect dose objective function, a dose uniformity objective function, a treatment time objective function, and a target area and healthy tissue biological effect dose ratio objective function as an objective function set; An optimization parameter acquisition module is used to acquire treatment time parameters, treatment angle parameters and target area boron concentration parameters as an optimization parameter set; A multi-objective optimization module, used to perform iterative optimization based on the objective function set and the optimization parameter set using a preset multi-objective optimization algorithm to obtain a corresponding Pareto front as a treatment plan; Each solution in the Pareto front represents a treatment plan.

9. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for generating a boron neutron capture therapy plan according to any one of claims 1 to 7 is implemented.

10. A terminal, characterized in that: It 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 executes the boron neutron capture therapy plan generation method according to any one of claims 1 to 7.

Images