Tumor irreversible electroporation ablation operation parameter optimization system

Abstract

The invention discloses a tumor irreversible electroporation ablation operation parameter optimization system which comprises a population initialization module, an individual grade division module, a first generation sub-population generation module, a new population determination module, a new parent population establishment module, a new parent population updating module, a merging module and a result selection module. According to the method, an NSGA-II algorithm in a genetic algorithm is adopted to optimize operation parameters, and threshold setting is added. Results obtained by the system can assist doctors in making an optimal operation parameter scheme, and the operation targets of killing tumor tissues to the maximum extent, minimizing damage to normal tissues and minimizing the number of pulses as much as possible are achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of machine learning, and in particular is a system for optimizing surgical parameters of irreversible electroporation ablation of tumors. Background Art

[0002] Malignant tumors are a major disease that seriously threatens human health and are the most serious public health challenge facing the world. Malignant tumors of the liver, bile duct and pancreas are the most common types of tumors in the digestive system. In the current clinical diagnosis and treatment of malignant tumors, the most commonly used method is surgical resection, which is a means of radical treatment of solid tumors. However, since malignant tumors are prone to local or distant invasion and metastasis, the best time for surgical resection is often missed. Physical tumor ablation is another way to treat pancreatic cancer, which uses external energy to act on solid tumors to eradicate or destroy tumor tissue. It mainly includes three types: thermal ablation, cryoablation and electrical ablation. However, thermal ablation has the disadvantage of not being suitable for areas near heat-sensitive organs, and cryoablation is prone to incomplete tumor ablation and tumor recurrence. Irreversible electroporation technology is an emerging tumor electrical ablation technology. Its principle is to use electric pulses to act on the surface of cancer cells, irreversibly damage the phospholipid bilayer, and form defects on the cell membrane surface. When the electric pulse reaches a certain level, the cell membrane defect cannot be reversed, and eventually induces target cell necrosis. It is a non-thermal ablation technology that can avoid irreversible damage to adjacent tissue structures and preserve intracellular macromolecules and tissue scaffolds. Existing clinical studies have also confirmed that irreversible electroporation technology has significant advantages in preserving organ function, and is particularly suitable for local ablation of tumors in small and complex organs such as the pancreas.

[0003] For high-voltage pulse ablation of tumors, how to achieve the surgical goal of "maximizing tumor tissue killing and minimizing normal tissue damage" has always been a hot topic. In order to achieve the best surgical effect, reasonable preoperative planning is crucial. However, at this stage, the preoperative planning of ablation surgery mainly relies on the doctor's clinical experience and heavily relies on the doctor's professional skills, which is not conducive to formulating the best ablation surgery plan, and it is easy to fail to completely kill cancer cells or "accidentally injure" normal tissues. During the irreversible electroporation ablation process, if the surgical parameters are set unreasonably, the ablation range may be too small, and the tumor tissue may not be completely killed, thereby increasing the risk of tumor recurrence. It may also cause the ablation range to be too large, hurting the normal tissue around the tumor, and increasing the risk of bleeding and infection, causing other complications. At the same time, according to research, too many pulses will aggravate cell electrolyte disorders, causing a greater burden on patients, and the more pulses there are, the longer the ablation treatment time will be. Long-term treatment may increase the risk of surgical trauma, infection and other complications. Summary of the invention

[0004] In view of the problem that the preoperative setting of surgical parameters still depends on the doctor's subjective experience, the purpose of the present invention is to provide a tumor irreversible electroporation ablation surgical parameter optimization system, which can assist doctors in formulating the best surgical parameter plan to achieve the surgical goals of maximizing the killing of tumor tissue, minimizing damage to normal tissue, and keeping the number of pulses as small as possible.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A tumor irreversible electroporation ablation surgery parameter optimization system, comprising the following modules:

[0007] (1) Initialization population module, which is used to construct the fitness function, determine the optimization range of surgical parameters, randomly initialize the population, and randomly generate a certain number of individuals as the initialization population.

[0008] (2) An individual level division module is used to divide the initialization population obtained by the initialization population module into multiple Pareto frontiers of different levels using a non-dominated sorting method, and to divide the individual levels in the initial population.

[0009] (3) The first-generation subpopulation generation module is used to perform selection, crossover and mutation operations on the individuals in the population obtained by the individual level division module to generate the first-generation subpopulation Q 0 .

[0010] (4) The new population determination module is used to convert the parent population P obtained by the first generation subpopulation generation module into 0 and the offspring population Q 0 Merge into a new population R 0 .

[0011] (5) A new parent population establishment module is used to use a fast non-dominated sorting method to transform the new population R obtained by the new population determination module 0 The individuals in the population are divided into Pareto frontiers of different orders, and the elite strategy is adopted to add individuals of each level to the new parent population in turn according to the order of dominance level, until all individuals of each level cannot be added to the new parent population;

[0012] (6) A new parent population update module, which is used to calculate the crowding distance of individuals in the level where all individuals obtained by the new parent population establishment module cannot be added to the new parent population, and add individuals with larger crowding distance to the new parent population until the specified number of population individuals is reached;

[0013] (7) The merging module is used to perform selection, crossover and mutation operations on the individuals of the new parent population obtained by the new parent population update module to generate a child population, merge the parent population and the child population as a new population to replace the new population R obtained by the new population determination module 0, enter the next cycle, and repeatedly execute the new parent population establishment module, the new parent population update module and the merging module until the number of iterations reaches the specified number;

[0014] (8) A result selection module is used to compare the Pareto fronts obtained from all iterations with the set threshold. The result selection module outputs the top 10 Pareto fronts that meet the threshold requirements as the optimization results of surgical parameter settings for doctors to choose based on clinical practice.

[0015] Furthermore, in the initialization population module, the fitness function is set according to the objectives of the irreversible electroporation ablation surgery for tumors, including maximizing the tumor ablation rate, minimizing normal tissue damage, and minimizing the number of pulses applied during the surgery;

[0016] The surgical parameters include electrode parameters and pulse parameters. The electrode parameters include electrode spacing and electrode length. The pulse parameters include voltage amplitude, pulse width and pulse number. The surgical parameter optimization range is represented by lb and rb. lb is a set of minimum values ​​of each surgical parameter value, and rb is a set of maximum values ​​of each surgical parameter value. The elements in the set are the minimum or maximum values ​​of electrode spacing, electrode length, voltage amplitude, pulse width and pulse number, respectively.

[0017] The initialization method adopts random initialization, specifically: construct a 200×5 matrix, in which the elements are randomly generated decimals between 0 and 1, multiply it with the two matrices obtained by subtracting rb and lb, and finally add it to lb to obtain an initialization population P with a population size of 200 0 ;

[0018] Furthermore, in the individual ranking module, the non-dominated ranking method needs to compare each individual with all other individuals in turn to determine the dominance relationship in the determination of the first layer of non-dominated frontier, and when determining the second layer of non-dominated frontier, the individuals other than the individuals belonging to the first layer of non-dominated frontier are again subjected to dominance relationship judgment, and the above process is repeated to obtain the rankings of all individuals;

[0019] Furthermore, in the first-generation subpopulation generation module, the selection operation uses a binary tournament selection method, randomly selecting two individuals from the population each time, and selecting the better individual from among them to enter the subpopulation according to the fitness value, and repeating this operation multiple times until the size of the subpopulation is the same as the parent population;

[0020] The mutation operation uses a polynomial mutation method, and the mutation form is:

[0021] x 1j (t) = x 1j (t)+Δ j

[0022] In the formula,

[0023] Among them, u is a random number between 0 and 1. is the distribution factor, set its value to 1;

[0024] The crossover operation uses a simulated binary crossover operator, assuming that the two parent individuals and Two offspring individuals generated using the crossover operator and It can be calculated by the following formula:

[0025]

[0026] In the formula,

[0027] Among them, u is a random number between 0 and 1. is the distribution factor, set its value to 1.

[0028] Furthermore, in the merging operation of the new population determination module, the number of individuals in the offspring population is consistent with that in the parent population, and the number of individuals in the new population is the sum of the offspring population and the parent population. The number of initialized populations is 200, so the number of individuals in the offspring population and the parent population is 200, and the number of individuals in the new population is 400;

[0029] Furthermore, in the new parent population establishment module, the fast non-dominated sorting method operates as follows:

[0030] First, calculate the new population R obtained by the new population determination module 0 The number of individuals dominated and dominated by each individual in the set is formed into a set. Next, a set of individuals that are not dominated by other individuals is established, and the level of this set is set to 1. After that, the remaining individuals dominated by the individual are formed into a set, and the number of individuals dominated by each individual is subtracted by 1. If the number of individuals dominated after subtraction is 0, it is used as one of the elements of the new set of individuals that are not dominated by other individuals. Finally, repeat the above operation to divide sets of different levels, that is, to divide into Pareto frontiers of different orders;

[0031] The elite strategy is to transform the parent population P t and the offspring population Q t Merge into a new population R t , then the parent population P t and the offspring population Q tThe individuals in the population compete together. The individuals in the population are quickly sorted and the crowding distance is calculated. The individuals with low rankings enter the next generation, and then the crowding distance is calculated. The individuals with larger crowding distances enter the next generation first, until the number of the next generation reaches a certain number;

[0032] Furthermore, the crowding distance in the new parent population update module is calculated by calculating the crowding distance between an individual and its two closest individuals at the same level, and then finding the optimal individual according to the crowding distance. The expression is as follows:

[0033]

[0034] In the formula, and are the target values ​​of the two adjacent individuals of individual i on the kth target, Indicates that individual i has the kth target f k The crowding distance on .

[0035] Furthermore, the maximum number of iterations specified in the merging module is 200.

[0036] Furthermore, the threshold conditions in the result selection module are that the thermal damage volume ratio must be less than 20%, and the tumor ablation rate must be greater than 75%. The first 10 Pareto solution sets that meet this threshold condition can be screened as output results.

[0037] Compared with the prior art, the present invention has the following technical effects:

[0038] The present invention aims at the problem that the current setting of surgical parameters before surgery still relies on the subjective experience of doctors. At the same time, in order to avoid the serious consequences of incomplete ablation or damage to adjacent tissues due to inappropriate surgical parameter settings, and to achieve the best effect of irreversible electroporation ablation surgery for tumors, it is necessary to formulate the best surgical plan before surgery, among which the most important thing is to determine the values ​​of each surgical parameter. The system of the present invention uses the NSGA-II algorithm in the genetic algorithm to optimize the surgical parameters, and adds threshold settings: the tumor ablation rate must be greater than 75%, and the thermal damage volume ratio must be less than 20%, so as to better screen out the parameter combination that best meets the surgical conditions. Doctors can select a group of parameter combinations from them as the surgical parameter values ​​for IRE ablation according to actual clinical conditions. In summary, the results obtained by the present invention can assist doctors in formulating the best surgical parameter plan to achieve the surgical goals of maximizing the killing of tumor tissue, minimizing damage to normal tissue, and minimizing the number of pulses. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A schematic diagram of the geometric structure of the electrode used in the experiment of the present invention;

[0040] Figure 2It is a visualization diagram of the parameter optimization results of the two-electrode sheet model in the experiment of the present invention;

[0041] Figure 3 It is a schematic diagram showing the visualization of the parameter optimization results of the three-electrode sheet model in the experiment of the present invention. DETAILED DESCRIPTION

[0042] The specific contents of the present invention are further explained in detail below in conjunction with embodiments.

[0043] Taking high-voltage pulse ablation surgery for pancreatic tumors as an example, how to achieve the surgical goal of "maximizing tumor tissue killing and minimizing normal tissue damage" has always been a hot issue. In order to achieve the best surgical effect, reasonable preoperative planning is crucial. However, the preoperative planning of ablation surgery at this stage mainly depends on the doctor's clinical experience and is heavily dependent on the doctor's professional skill level, which is not conducive to formulating the best ablation surgery plan, and it is easy to fail to completely kill cancer cells or "accidentally injure" normal tissues. Therefore, this embodiment uses the NSGA-II algorithm with a threshold to optimize the surgical parameters, obtain a parameter combination that meets the surgical effect requirements, assist clinicians in making decisions, and achieve precise ablation of pancreatic tumors.

[0044] A parameter optimization method for irreversible electroporation ablation of pancreatic tumors based on NSGA-II algorithm with added threshold, comprising the following steps:

[0045] (1) Initialization population module, which is used to construct the fitness function, determine the optimization range of surgical parameters, randomly initialize the population, and randomly generate a certain number of individuals as the initialization population.

[0046] The fitness function is set according to the goals of irreversible electroporation ablation of pancreatic tumors, including maximizing the tumor ablation rate, minimizing normal tissue damage, and minimizing the number of pulses applied during the operation;

[0047] The surgical parameters include electrode parameters and pulse parameters. The electrode parameters include electrode spacing and electrode length. The pulse parameters include voltage amplitude, pulse width and pulse number. The surgical parameter optimization range is represented by lb and rb. lb is a set of minimum values ​​of each surgical parameter value, and rb is a set of maximum values ​​of each surgical parameter value. The set lb is [500, 8, 8, 25, 0], and the set rb is [2000, 10, 10, 100, 300]. The elements in the set are the minimum or maximum values ​​of electrode spacing, electrode length, voltage amplitude, pulse width and pulse number, respectively.

[0048] The initialization population adopts a random initialization method, specifically: construct a 200×5 matrix, in which the elements are randomly generated decimals between 0 and 1, multiply it with the two matrices obtained by subtracting rb and lb, and finally add it to lb to obtain an initialization population P with a population size of 200 0 ;

[0049] In this module, the setting of the fitness function makes the optimization goal clearer and more specific. The determination of surgery-related parameters and the optimization range can make the optimization object clearer. The use of random initialization method to generate the initialization population is an important beginning for subsequent population iterations and can help the algorithm perform better.

[0050] The individual level division module is used to divide the initialization population obtained by the initialization population module into multiple Pareto frontiers of different levels using the non-dominated sorting method, and divide the individual levels in the initialization population. The specific operations include the following:

[0051] When determining the first-layer non-dominated frontier, the non-dominated sorting method needs to compare each individual with all other individuals in turn to determine the dominance relationship; when determining the second-layer non-dominated frontier, the dominance relationship of the individuals other than the individuals belonging to the first-layer non-dominated frontier is judged again, and the above process is repeated to obtain the ranks of all individuals.

[0052] In this module, non-dominated sorting can obtain the Pareto frontier and corresponding level of each individual in the initial population, laying the foundation for generating the first generation of sub-populations.

[0053] The first generation subpopulation generation module is used to select, crossover and mutate individuals in the population obtained by the individual level division module to generate the first generation subpopulation Q 0 ;

[0054] The selection operation uses a binary tournament selection method, randomly selecting two individuals from the population each time, and selecting the better individual from among them to enter the sub-population according to the fitness value, and repeating this operation multiple times until the size of the offspring population is the same as the parent population.

[0055] The mutation operation uses a polynomial mutation method, and the mutation form is:

[0056] x 1j (t) = x 1j (t)+Δ j

[0057] In the formula,

[0058] Among them, u is a random number between 0 and 1. is the distribution factor, set to 1.

[0059] The crossover operation uses a simulated binary crossover operator, assuming that the two parent individuals and Two offspring individuals generated using the crossover operator and Calculated by the following formula:

[0060]

[0061] In the formula,

[0062] Among them, u is a random number between 0 and 1. is the distribution factor, set to 1.

[0063] In this module, based on the fitness evaluation, some individuals with high fitness are selected for the next generation of reproduction to produce new individuals. At the same time, crossover and mutation operations can increase the algorithm's ability to explore information, which is also the core operation of the genetic algorithm.

[0064] The new population determination module is used to convert the parent population P obtained by the first generation subpopulation generation module 0 and the offspring population Q 0 Merge into a new population R 0 .

[0065] In the merging operation, the number of individuals in the offspring population is consistent with that in the parent population, and the number of individuals in the new population is the sum of the offspring population and the parent population. The number of initialized populations is 200. Therefore, the number of individuals in the offspring population and the parent population is 200, and the number of individuals in the new population is 400.

[0066] In this module, by merging the parent population and the offspring population, the new generation of population is produced through joint competition, which is conducive to retaining excellent individuals to a greater extent and improving the performance of the surgical parameter optimization method;

[0067] The new parent population establishment module is used to use the fast non-dominated sorting method to divide the individuals in the new population determination module into Pareto frontiers of different orders, and adopt the elite strategy to add individuals of each level to the new parent population in turn according to the order of dominance level until all individuals of the level cannot be added to the new parent population.

[0068] The fast non-dominated sorting method first calculates the number of individuals dominated and dominated by each individual and forms a set. Then, a set of individuals that are not dominated by other individuals is established, and the level of this set is set to 1. After that, the remaining individuals dominated by the individual are formed into a set, and the number of individuals dominated by each individual is subtracted by 1. If the number of individuals dominated after subtraction is 0, it is used as one of the elements of the new set of individuals that are not dominated by other individuals. Finally, the above operations are repeated to divide the Pareto frontiers into different orders as the processing objects of the elite strategy;

[0069] The above elite strategy is to obtain the parent population P t and the offspring population Q t Merge into a new population R t , then the parent population P t and the offspring population Q t The individuals in the group compete together. Perform non-dominated sorting calculation and crowding distance calculation to obtain individuals with low rankings to enter the next generation, and then calculate the crowding distance. Individuals with larger rankings will enter the next generation first, until the number of the next generation reaches a certain number.

[0070] In this step, fast non-dominated sorting greatly reduces the computational complexity. At the same time, the corresponding levels of individuals in the population can be obtained. The use of the elite strategy can speed up the execution of the surgical parameter optimization method and help retain better individuals.

[0071] Step 6: Calculate the crowding distance of individuals in the level where all individuals obtained in step 5 cannot be added to the new parent population, and add individuals with larger crowding distance to the new parent population until the specified number of population individuals is reached.

[0072] The crowding distance is calculated by calculating the crowding distance between two adjacent individuals at the same level, and then finding the optimal individual according to the crowding distance. The expression is as follows:

[0073]

[0074] In the formula, and are the target values ​​of the two adjacent individuals of individual i on the kth target, Indicates that individual i has the kth target f k The crowding distance on .

[0075] In this step, the calculation of the crowding distance is a way to compare individuals in the population, so that the solutions can be more evenly distributed on the Pareto front, which helps to maintain the diversity of the population.

[0076] Step 7: Select, crossover and mutate the individuals of the new parent population obtained in step 6 to generate a child population, merge the parent population and the child population to enter the next cycle until the number of iterations reaches the specified number.

[0077] The prescribed number of times is 200 times.

[0078] Step 8: Compare the Pareto front obtained from all iterations with the set threshold, and take the top 10 solution sets that meet the threshold as the optimization results of surgical parameter settings. The optimization results can be selected by doctors according to clinical practice.

[0079] The thresholds are set as thermal damage volume ratio less than 20% and tumor ablation rate greater than 75%, and the first 10 Pareto solution sets that meet the threshold conditions can be screened as output results.

[0080] This embodiment is implemented using Pycharm software and is performed under two irreversible electroporation ablation catheter electrode structures, including a two-electrode sheet structure model and a three-electrode sheet structure model. Figure 1 As shown, the pancreatic duct tissue is a horizontally set tube with an inner diameter of 0.2 cm, an outer diameter of 0.25 cm, and a length of 10 cm. The tumor tissue is an ellipsoid with semi-axis a, semi-axis b, and semi-axis c values ​​of 5 mm, 6 mm, and 6 mm, respectively. The red cylinder represents the positive electrode, and the yellow cylinder represents the negative electrode. As can be seen from the figure, both are catheter electrodes, and the electrodes are set parallel to the luminal tissue, simulating the situation where the catheter electrode enters the pancreatic duct through the natural cavity of the human body for ablation. The difference is that the two-electrode sheet type includes a positive electrode and a negative electrode, and the two electrodes are symmetrically placed about the ellipsoid representing the tumor, while the three-electrode sheet type includes a positive electrode and two negative electrodes, the positive electrode is located in the middle of the simulation model, and the two negative electrodes are symmetrically placed on both sides of the positive electrode.

[0081] For the two-electrode sheet model, the optimization results of the irreversible electroporation ablation surgical parameters are shown in Table 1. In this table, the surgical parameters include electrode spacing, electrode length, voltage amplitude, pulse width and number of pulses. The thermal damage volume and tumor ablation rate are the ratio of the damaged volume of the luminal tissue and the ablated volume of the tumor during the irreversible electroporation process corresponding to the parameters on the left. At the same time, the output results are visualized, and the Pareto front solution is relatively scattered, and more optimal solutions can be found in the parameter space, rather than concentrated in a certain area. The results are shown in Figure 2 shown.

[0082] Table 1 Parameter optimization results of the two-electrode sheet model

[0083]

[0084] As shown in Table 1, the surgical parameter combination obtained by the NSGA-II algorithm with added threshold meets the threshold requirements, but there is no parameter combination in which the tumor ablation rate reaches 100% and the thermal damage volume ratio is 0%. Doctors can choose the best surgical parameter combination based on clinical practice. For example, if the ablation surgery goal is to minimize the damage to normal tissue, the doctor can choose the parameter combination with serial number 10 in the table; if the ablation surgery goal is to achieve complete killing of tumor tissue, the doctor can choose the parameter combination with serial number 1 in the table; if the ablation surgery goal is to achieve a smaller number of pulses, the doctor can choose the parameter combination with serial number 10 in the table.

[0085] For the three-electrode structure model, the parameter optimization method of irreversible electroporation ablation of pancreatic tumors using the NSGA-II algorithm with a threshold is used. The algorithm output results are shown in Table 2, and the corresponding Pareto solution set and frontier visualization results are shown in Figure 3 shown.

[0086] Table 2 Parameter optimization results of the three-electrode sheet model

[0087]

[0088] It can be seen from Table 2 that the surgical parameter combination obtained by the parameter optimization method of irreversible electroporation ablation of pancreatic tumors using the NSGA-II algorithm with a threshold meets the threshold requirements. There is no parameter combination in which the tumor ablation rate reaches 100% and the thermal damage volume ratio reaches 0%. However, the parameter combination with the highest tumor ablation rate and the lowest thermal damage volume ratio is No. 4 in the table. Therefore, if the doctor is more concerned about the electrothermal damage, the parameter combination with No. 4 can be selected; if the doctor prefers to achieve a smaller number of pulses, the surgical parameter combination with No. 9 or 10 in the table can be selected.

Claims

1. A tumor irreversible electroporation ablation surgery parameter optimization system, characterized in that: Includes the following modules: The population initialization module is used to construct a fitness function, determine the optimization range of surgical parameters, randomly initialize the population, and randomly generate a certain number of individuals as the initialization population; An individual level division module is used to divide the initialization population obtained by the initialization population module into multiple Pareto frontiers of different levels using a non-dominated sorting method, and to divide the individual levels in the initial population; The first-generation subpopulation generation module is used to perform selection, crossover and mutation operations on the individuals in the population obtained by the individual level division module to generate the first-generation subpopulation Q0; A new population determination module is used to merge the parent population P0 and the child population Q0 obtained by the first-generation child population generation module into a new population R0; The new parent population establishment module is used to use the fast non-dominated sorting method to divide the individuals in the new population R0 obtained by the new population determination module into Pareto frontiers of different orders, and adopt the elite strategy to add individuals of each level to the new parent population in turn according to the order of dominance level until all individuals of each level cannot be added to the new parent population; The new parent population update module is used to calculate the crowding distance of individuals in the level where all individuals obtained by the new parent population establishment module cannot be added to the new parent population, and add individuals with larger crowding distance to the new parent population until the specified number of population individuals is reached; The merging module is used to perform selection, crossover and mutation operations on the individuals of the new parent population obtained by the new parent population updating module to generate a child population, merge the parent population and the child population as a new population to replace the new population R0 obtained by the new population determination module, enter the next cycle, and repeatedly execute the new parent population establishment module, the new parent population update module and the merging module until the number of iterations reaches the specified number; The result selection module is used to compare the Pareto fronts obtained from all iterations with the set thresholds, and output the first several Pareto fronts that meet the threshold requirements as the optimization results of the surgical parameter settings.

2. The irreversible electroporation ablation parameter optimization system for tumors according to claim 1, characterized in that: In the initialization population module, the fitness function is set according to the goals of the irreversible electroporation ablation surgery for tumors, including maximizing the tumor ablation rate, minimizing normal tissue damage, and minimizing the number of pulses applied during the surgery; The surgical parameters include electrode parameters and pulse parameters. The electrode parameters include electrode spacing and electrode length. The pulse parameters include voltage amplitude, pulse width and pulse number. The optimization range of surgical parameters is represented by lb and rb, where lb is the set of minimum values ​​of each surgical parameter, and rb is the set of maximum values ​​of each surgical parameter. The elements in the set are the minimum or maximum values ​​of electrode spacing, electrode length, voltage amplitude, pulse width, and pulse number, respectively. The initialization method adopts random initialization, specifically: construct a 200×5 matrix, in which the elements are randomly generated decimals between 0 and 1, multiply it with the two matrices obtained by subtracting rb and lb respectively, and finally add it to lb to obtain the initialized population P0 with a population size of 200.

3. The irreversible electroporation ablation parameter optimization system for tumors according to claim 1, characterized in that: In the individual ranking module, the non-dominated sorting method needs to compare each individual with all other individuals in turn to judge the dominance relationship when determining the first layer of non-dominated frontier. When determining the second layer of non-dominated frontier, the dominance relationship of individuals other than those belonging to the first layer of non-dominated frontier is judged again. The above process is repeated to obtain the ranks of all individuals.

4. The irreversible electroporation ablation parameter optimization system for tumors according to claim 1, characterized in that: In the first-generation subpopulation generation module, the selection operation uses a binary tournament selection method, randomly selecting two individuals from the population each time, and selecting the better individual from among them to enter the subpopulation according to the fitness value, and repeating this operation multiple times until the size of the subpopulation is the same as the parent population; The mutation operation uses a polynomial mutation method, and the mutation form is: x 1j (t)=x 1j (t)+Δ j In the formula, Among them, u is a random number between 0 and 1. is the distribution factor, set its value to 1; The crossover operation uses a simulated binary crossover operator, assuming that the two parent individuals and Two offspring individuals generated using the crossover operator and It can be calculated by the following formula: In the formula, Among them, u is a random number between 0 and 1. is the distribution factor, set its value to 1.

5. The irreversible electroporation ablation parameter optimization system for tumors according to claim 1, characterized in that: In the merging operation of the new population determination module, the number of individuals in the offspring population is consistent with that in the parent population in the population merging, and the number of individuals in the new population is the sum of the offspring population and the parent population.

6. The irreversible electroporation ablation parameter optimization system for tumors according to claim 1, characterized in that: In the new parent population establishment module, the fast non-dominated sorting method operates as follows: First, the number of dominated and dominated individuals of each individual in the new population R0 obtained by the new population determination module is calculated and formed into a set; then, a set of individuals that are not dominated by other individuals is established, and the level of this set is set to 1; then, the remaining individuals dominated by the individual are formed into a set, and the number of dominated individuals of each individual is subtracted by 1. If the number of dominated individuals after subtraction is 0, it is used as one of the elements of the new set of individuals that are not dominated by other individuals; finally, the above operations are repeated to divide sets of different levels, that is, to divide into Pareto frontiers of different orders; The elite strategy is to transform the parent population P t With the offspring population Q t Merge into a new population R t , then the parent population P t and the offspring population Q t The individuals in the population participate in the competition together; the individuals in the population are quickly non-dominated sorted and the crowding distance is calculated, and the individuals with low ranks enter the next generation, and then the crowding distance is calculated, and the larger ones enter the next generation first, until the number of the next generation reaches a certain number.

7. The irreversible electroporation ablation parameter optimization system for tumors according to claim 1, characterized in that: The crowding distance in the new parent population update module is calculated by calculating the crowding distance between an individual and its two closest individuals at the same level, and then finding the optimal individual according to the crowding distance. The expression is as follows: In the formula, and are the target values ​​of the two adjacent individuals of individual i on the kth target, Indicates that individual i has the kth target f k The crowding distance on .

8. The irreversible electroporation ablation parameter optimization system for tumors according to claim 1, characterized in that: The maximum number of iterations specified in the merging module is 200.

9. The tumor irreversible electroporation ablation parameter optimization system according to claim 1, characterized in that: The threshold conditions in the result selection module are that the thermal damage volume ratio must be less than 20%, and the tumor ablation rate must be greater than 75%. The first 10 Pareto solution sets that meet the threshold conditions can be screened as output results.

Images