Neutron transport simulation method, device, medium and equipment and BNCT treatment system

Abstract

The invention provides a neutron transport simulation method and device, a medium, equipment and a BNCT treatment system. The method comprises the following steps: setting a plurality of voxel grids; the voxel grids are various human tissue materials corresponding to different boron drug concentrations; performing off-line coding on each voxel grid to obtain a corresponding physical coding layer, wherein the physical coding layer is used for predicting a corresponding neutron outgoing flow and neutron flux distribution based on the neutron incoming flow of the voxel grid; acquiring a CT image of the target object, and matching the voxel grid according to the CT image and the boron medicine planned concentration to acquire a voxel model of the target object; the physical coding layers of all voxel grids in the voxel model are coupled online. According to the method, the physical coding layers of various voxel grids are generated in an off-line manner, and the corresponding neutron outgoing flow and neutron flux distribution can be quickly predicted based on the given neutron incoming flow; and coupling a plurality of voxel grids based on a patient individualized voxel model to quickly obtain neutron flux distribution in the voxel model.

Description

Technical Field

[0001] This application belongs to the technical field of tumor treatment, and relates to a neutron transport simulation method, device, medium, and equipment, as well as a neutron transport simulation method, device, medium, and equipment of a BNCT treatment system and a BNCT treatment system. Background Art

[0002] As a highly potential cancer treatment method, boron neutron capture therapy (BNCT) has attracted much attention in recent years with the development of accelerators. Its principle is based on the high capture cross-section characteristics of 10B for thermal neutrons. When a boron-containing compound (boron drug) is specifically taken up by tumor cells, a thermal neutron beam is introduced for irradiation. After 10B captures a thermal neutron, a nuclear reaction occurs, generating high-energy α particles and 7Li atomic nuclei. The ranges of these particles are extremely short, only within a few micrometers, which means they will almost only damage the tumor cells containing 10B and have minimal impact on surrounding normal tissues. This unique treatment mechanism provides a highly precise way for cancer treatment, and is expected to improve the tumor treatment effect while significantly reducing the side effects on normal tissues.

[0003] Currently, the treatment effect of BNCT largely depends on the accuracy of dose calculation. The treatment planning system (TPS) is a key link in the entire boron neutron capture therapy process. In BNCT-TPS, accurate dose calculation is the core link to ensure treatment effect and safety. For example, through accurate dose calculation, it can be ensured that the tumor tissue receives sufficient treatment dose to completely kill tumor cells, while controlling the dose of normal tissues within a safe range to avoid serious complications caused by excessive dose. However, the transport process of neutrons in the human body often affects the effect of the BNCT treatment plan. Therefore, how to simulate the neutron transport process and generate a BNCT treatment plan based on this is a technical problem that needs to be urgently solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a neutron transport simulation method, device, medium, and equipment, as well as a BNCT treatment system, for solving the technical problem of how to simulate the neutron transport process and generate a BNCT treatment plan based on this.

[0005] In a first aspect, this application provides a neutron transport simulation method, which is applicable to a BNCT treatment planning system. The method includes: Setting a plurality of voxel grids; the voxel grids are various human tissue materials corresponding to different boron drug concentrations; Offline encode each of the voxel grids to obtain a corresponding physical encoding layer, which is used to predict the corresponding neutron outgoing flux and neutron flux distribution based on the neutron incident flux of the voxel grid; Obtain the CT image of the target object, and match at least one of the voxel grids according to the CT image and the boron drug planned concentration to obtain the voxel model of the target object; Online couple the physical encoding layers of all the voxel grids in the voxel model to obtain the neutron flux distribution of the voxel model.

[0006] In some embodiments of the first aspect of the present application, offline encoding each of the voxel grids to obtain a corresponding physical encoding layer includes: Customize the neutron incident flux and the neutron outgoing flux in the eight-dimensional phase space; the eight-dimensional phase space includes a time phase space, an energy phase space, a three-dimensional angular phase space, and a three-dimensional spatial phase space; Give different neutron incident fluxes to each of the voxel grids to correspondingly obtain the neutron outgoing flux and the neutron flux distribution in the eight-dimensional phase space; Offline encode the physical encoding layer of each voxel grid based on the given neutron incident flux and the corresponding neutron outgoing flux and neutron flux distribution.

[0007] In some embodiments of the first aspect of the present application, giving different neutron incident fluxes to each of the voxel grids to correspondingly obtain the neutron outgoing flux and the neutron flux distribution in the eight-dimensional phase space includes: Simulate the neutron transport process in the voxel grid based on the eight-dimensional phase space distribution of the neutron incident flux to obtain the eight-dimensional phase space distribution of the neutron outgoing flux and the eight-dimensional phase space distribution of the neutron flux; wherein, the neutron transport process is characterized by the relationship characteristics between the neutron incident flux and the neutron outgoing flux.

[0008] In some embodiments of the first aspect of the present application, combining the definitions of the eight-dimensional phase space of the neutron incident flux and the neutron outgoing flux, perform spatial integration on the voxel grid to characterize the neutron transport process in the voxel grid as the relationship characteristics between the neutron incident flux and the neutron outgoing flux.

[0009] In some embodiments of the first aspect of the present application, simulating the neutron transport process in the voxel grid based on the eight-dimensional phase space distribution of the neutron incident flux to obtain the eight-dimensional phase space distribution of the neutron outgoing flux and the eight-dimensional phase space distribution of the neutron flux includes: In the time-phase space, based on the asymptotic state approximation, discrete calculations are performed on the time distribution of the neutron incident flux to correspondingly obtain the time distribution of the neutron outgoing flux and the time distribution of the neutron flux; In the energy-phase space, based on the group approximation, discrete calculations are performed on the energy distribution of the neutron incident flux to correspondingly obtain the energy distribution of the neutron outgoing flux and the energy distribution of the neutron flux; In the three-dimensional angular phase space, based on quadrature discretization, discrete calculations are performed on the three-dimensional angular distribution of the neutron incident flux to correspondingly obtain the three-dimensional angular distribution of the neutron outgoing flux and the three-dimensional angular distribution of the neutron flux; In the three-dimensional spatial phase space, based on quadrature discretization, discrete calculations are respectively performed on the three-dimensional spatial distribution of the neutron incident flux to correspondingly obtain the three-dimensional spatial distribution of the neutron outgoing flux and the three-dimensional spatial distribution of the neutron flux.

[0010] In some embodiments of the first aspect of the present application, the online coupling of the physical coding layers of all the voxel grids in the voxel model includes: Set a voxel grid in the voxel model as the initial grid, and give a neutron incident flux to the initial grid; Predict the corresponding neutron outgoing flux and neutron flux distribution based on the physical coding layer of the initial grid; Use the neutron outgoing flux of the initial grid as the neutron incident flux of the next voxel grid to predict the corresponding neutron outgoing flux and neutron flux distribution based on the physical coding layer of the next voxel grid until all the voxel grids are traversed.

[0011] In a second aspect, the present application provides a BNCT treatment planning system, which is used to generate a BNCT treatment plan based on the neutron flux distribution in the voxel model of the target object obtained by the method as described above.

[0012] In a third aspect, the present application provides a neutron transport simulation device, which includes: A setting module, configured to set a plurality of voxel grids; the voxel grids are various human tissue materials corresponding to different boron drug concentrations; A physical coding module, configured to perform offline coding on each of the voxel grids to obtain a corresponding physical coding layer, and the physical coding layer is used to predict the corresponding neutron outgoing flux and neutron flux distribution based on the neutron incident flux of the voxel grid; An acquisition module, configured to acquire a CT image of the target object, and match at least one of the voxel grids according to the CT image and the boron drug planned concentration to obtain the voxel model of the target object; A calculation module, configured to perform online coupling on the physical coding layers of all the voxel grids in the voxel model to obtain the neutron flux distribution of the voxel model.

[0013] In a fourth aspect, the present application provides a terminal, including: a processor and a memory, communicatively connected between the memory and the processor; the memory is configured to store a computer program, and the processor is configured to execute the computer program stored in the memory, so that the terminal executes the method as described above.

[0014] In a fifth aspect, the present application provides a computer storage medium, and the method as described above is implemented when the computer program is executed by a processor.

[0015] As described above, the neutron transport simulation method, device, medium, and equipment provided by the present application and the BNCT treatment system have the following beneficial effects: The present application innovatively proposes to offline generate the physical coding layers of various voxel grids, so as to quickly realize predicting the neutron outgoing flux and neutron flux distribution based on a given neutron incident flux, and online couple multiple voxel grids based on the patient's individualized voxel model, thereby quickly simulating the neutron flux distribution in different patients, and further building a personalized dose evaluation model for different patients, solving problems such as high cost, slow response, low generalization ability of artificial intelligence algorithms, and strong dependence on big data in the generation of traditional BNCT treatment plans, further improving the generalization ability and calculation accuracy of the algorithm, and meeting the requirements of clinical applications for calculation accuracy and time. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It shows a schematic flowchart of the neutron transport simulation method provided by an embodiment of the present application.

[0017] Figure 2 It shows a schematic flowchart of the neutron transport simulation method provided by an embodiment of the present application.

[0018] Figure 3 It shows a schematic diagram of the relationship between the three-dimensional angular phase space and the three-dimensional spatial phase space provided by an embodiment of the present application.

[0019] Figure 4 It shows a schematic diagram of the distribution of N classical quadrature points in the three-dimensional angular phase space provided by an embodiment of the present application.

[0020] Figure 5 It shows an application schematic diagram of the neutron transport simulation method provided by an embodiment of the present application.

[0021] Figure 6 It shows a schematic structural diagram of the neutron transport simulation device described in an embodiment of the present application.

[0022] Figure 7 Shown is a schematic diagram of the hardware structure of a terminal provided by an embodiment of the present application. Detailed implementation manners

[0023] 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 other 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.

[0024] 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 herein 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.

[0025] 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.

[0026] Before further elaborating on the present application, the nouns and terms involved in the embodiments of the present application are described. The nouns and terms involved in the embodiments of the present application are applicable to the following explanations: BNCT: Boron Neutron Capture Therapy, which is a boron neutron capture therapy. When irradiating a patient with neutrons, only the 10 B enriched in cancer cells undergoes a neutron capture reaction, and the high-energy particles released only act within the cancer cells, thereby achieving precise killing of cancer cells.

[0027] BNCT-TPS: BNCT Treatment Planning System, which is used to obtain BNCT treatment plans, including radiation doses.

[0028] Neutron incident flux: In this application, it is represented as the neutron flux directed into the voxel grid from each surface, and its custom variables are in the eight-dimensional phase space.

[0029] Neutron outgoing flux: In this application, it is represented as the neutron flux directed out of the voxel grid from each surface, and its custom variables are in the eight-dimensional phase space.

[0030] Neutron flux: A scalar physical quantity, which is defined in this application as: at time t, in a unit volume at position r in space, within a unit energy interval of energy E, the total track length traveled by neutrons within a unit solid angle in the direction of motion Ω per unit time.

[0031] Neutron flux distribution: In this application, it is defined as the distribution of neutron flux in different phase space dimensions, representing the complete state of neutron flux in the phase space.

[0032] It should be noted that after the neutron incident flux enters the voxel grid, the transport process of the neutron beam within the voxel grid can be represented by the dynamic particle transport equation, as shown in Equation (1): (1)

[0033] Where t is time; r is three-dimensional space; Ω is three-dimensional angle; E is particle energy.

[0034] Among them, the first term on the left side of the equation represents the rate of change of neutron flux with time, divided by the neutron velocity v. It reflects the degree of change in the neutron fluence with a specific energy E and moving in the direction Ω per unit time and unit volume, taking into account the time accumulation effect caused by the neutrons moving at velocity v. Among them, is the neutron flux ; v is the neutron velocity .

[0035] Among them, the second term on the left side of the equation is the convection term, which represents the change of neutrons in space due to their own motion (moving in the direction Ω at velocity v), that is, the influence of the space transport process on the neutron flux. Among them, Ω is a three-dimensional angle, dimensionless; The dimension of is

[0036] Among them, the third term on the left side of the equation represents the collision removal term in is the total cross section , which represents the comprehensive probability that neutrons interact with the medium in various ways (such as absorption, scattering, etc.) and thus deviate from their original direction of motion and energy state. Multiplying it by the neutron flux represents the reduction in the number of neutrons with specific energy and direction per unit volume due to various interactions such as collisions.

[0037] Among them, the first term on the right side of the equation is the scattering production term, and its integral is carried out over all possible incident energies E' and incident directions Ω'. Among them, is the scattering cross-section , which represents that per unit volume, the probability that neutrons scatter from energy E' , direction Ω' to energy E , direction Ω. Multiplying it by the neutron flux and then integrating and summing can obtain the number of neutrons with the current energy and direction generated due to the scattering effect.

[0038] Among them, the second term on the right side of the equation is the external source term , that is, the BNCT neutron source, which represents the number of neutrons generated per unit volume at position r, energy E, direction Ω, and time t.

[0039] Actually, the first term on the left side of Equation (1) is the rate of change of neutrons with time, the second term is the neutron leakage rate, and the third term is the neutron absorption rate; the right side of the equation is the neutron production rate, where the first term is the neutron source; the second term is the scattering neutron production rate. That is, Equation (1) represents the dynamic transport process of neutrons. It can be seen from Equation (1) that the neutron flux is affected by neutron leakage, interaction losses (total cross-section and scattering cross-section), scattering, and source terms. Then, solving Equation (1) can simulate the transport process of neutrons and obtain the neutron flux distribution accordingly.

[0040] Currently, the therapeutic effect of BNCT depends to a large extent on the accuracy of BNCT dose calculation to ensure that patients can receive sufficient neutron radiation. In BNCT-TPS, the traditional method uses the Monte Carlo method to simulate the transport process of neutron beams in human tissues, including neutron scattering, absorption (interaction with boron), etc. This method solves Equation (1) through a large number of random samplings and statistical analyses, simulates the transport process of neutrons, thereby obtaining the neutron flux distribution in the human body, and then obtaining the BNCT dose. However, due to the complexity of the BNCT process, including neutron scattering and absorption in biological tissues, as well as the non-uniform distribution of boron in tumor cells and normal tissues, etc., the calculation of the neutron flux distribution will be affected by multiple factors.

[0041] For this reason, artificial intelligence algorithms have been widely studied and applied. It can solve Equation (1) by constructing a complex relationship model, simulate the dynamic transport process of neutrons, and then obtain the neutron flux distribution. However, the interpretability of the model is poor. An artificial intelligence model is often a complex black box. Although it can give accurate calculation results, it is difficult to explain its calculation process and decision-making basis, which to a certain extent limits doctors' understanding and trust in treatment plans. Secondly, artificial intelligence algorithms are highly dependent on data. A large amount of high-quality data is required for training to obtain an accurate model. If the data quality is not high, the data volume is insufficient, or the data distribution is uneven, it may lead to poor generalization ability of the model and inability to quickly obtain accurate results for different patients. Currently, Monte Carlo software is often used to perform a large number of pre-calculations on the sample set of cases to train the model with numerical simulation data. If the actual situation of the patient is very different from the sample set of cases, it is impossible to obtain more accurate results through the existing model, and pre-calculation and model training need to be carried out again, which goes against the original intention of introducing artificial intelligence methods for acceleration.

[0042] To at least solve the above technical problems, the present application innovatively proposes a physical coding layer for offline generation of various voxel grids, and on-line coupling of typical voxel grid units to build a patient-specific voxel model, and then simulate the transport process of neutrons in different voxel models, obtain the neutron flux distribution in the voxel model, and finally achieve the purpose of improving the calculation efficiency, generalization ability and application range, which helps to quickly and accurately generate the BNCT treatment plan.

[0043] Figure 1 The flow schematic diagram of the neutron transport simulation method provided by the embodiment of the present application is shown. The neutron transport simulation method provided by the present application is applicable to the BNCT treatment planning system to quickly generate the BNCT treatment plan. As Figure 1 shown, the neutron transport simulation method includes steps S1 to S3.

[0044] S1. Set a plurality of voxel grids; the voxel grids are various human tissue materials corresponding to different boron drug concentrations.

[0045] In some embodiments, various human tissue materials are constructed using the human tissue material element-related database published by the International Commission on Radiation Units and Measurements (ICRU). For example, there are N types of human tissue material grids, and the boron concentration range in the clinical treatment process, such as (0-100GY), can be divided into 100 boron concentrations (1GY, 2GY, 3GY...100GY), and these 100 different boron concentrations are filled into the grids of N types of human tissue materials to construct a 100xN voxel grid. That is, for each type of human tissue material grid, the different boron concentrations in the material are fully considered, so as to construct multiple voxel grids to meet the needs of clinical generation of BNCT treatment plans.

[0046] S2. Offline encoding is performed on each voxel grid to obtain a corresponding physical coding layer, where the physical coding layer is used to predict the corresponding neutron outgoing flow and neutron flux distribution based on the neutron incident flow of the voxel grid.

[0047] Specifically, offline coding means that there is no need to know any information about the patient in advance, and a physical coding layer is directly constructed for each voxel grid through a large amount of universal data. Through the physical coding layer, when a neutron incident flow is given to the voxel grid, the neutron transport process in the voxel grid can be quickly simulated, so as to respond to and predict the corresponding neutron outflow flow and neutron flux distribution. Figure 2 FIG. 1 is a flow chart of a neutron transport simulation method provided in an embodiment of the present application. Figure 2 As shown, performing offline encoding on each voxel grid to obtain a corresponding physical coding layer includes steps S21 to S23.

[0048] S21. Customize the neutron incident flow and the neutron outgoing flow in an eight-dimensional phase space; the eight-dimensional phase space includes a time phase space, an energy phase space, a three-dimensional angle phase space, and a three-dimensional space phase space.

[0049] Specifically, the neutron incident flux is defined as , and the neutron incident flow is located on the voxel grid surface and the angle between the neutron motion direction and the normal direction is greater than 90 degrees, the neutron outflow is defined as , and the neutron outflow is located on the surface of the voxel grid and the angle between the neutron movement direction and the normal direction is less than 90 degrees. In addition, the internal area of ​​the voxel grid is recorded as D, and the surface is recorded as S.

[0050] Then, in combination with the eight-dimensional phase space definition of the neutron incident flow and the neutron outgoing flow, spatial integration is performed on the voxel grid to characterize the neutron transport process in the voxel grid as the characteristic relationship between the neutron incident flow and the neutron outgoing flow. That is, in combination with the above definition, spatial integration of formula (1) is performed in the internal area D of a certain voxel grid, and formula (2) can be obtained: (2)

[0051] That is, as can be seen from Equation (2), the dynamic transport process of neutrons in each voxel grid can actually be represented by the relationship characteristics between the neutron incident flux and the neutron outgoing flux in the voxel grid, that is .

[0052] Wherein Figure 3 It is shown as a schematic diagram of the relationship between the three-dimensional angular phase space and the three-dimensional spatial phase space provided by the embodiments of the present application. By simulating the neutron transport process in the voxel grid based on Equation (2) in the eight-dimensional phase space, it is possible to comprehensively consider the influence of various variables that the dynamic transport process of neutrons in the voxel grid will actually be subject to, so as to obtain the neutron flux distribution corresponding to the neutron incident flux and the neutron outgoing flux in the eight-dimensional phase space, enabling a more accurate simulation of the neutron transport process in the voxel grid.

[0053] S22. Given different neutron incident fluxes for each of the voxel grids to correspondingly obtain the neutron outgoing flux and the neutron flux distribution in the eight-dimensional phase space.

[0054] S23. Based on the given neutron incident flux, the corresponding neutron outgoing flux, and the neutron flux distribution, offline encode the physical encoding layer of each voxel grid.

[0055] Specifically, given different neutron incident fluxes for each of the voxel grids to correspondingly obtain the neutron outgoing flux and the neutron flux distribution in the eight-dimensional phase space includes: simulating the neutron transport process in the voxel grid based on the eight-dimensional phase space distribution of the neutron incident flux to obtain the eight-dimensional phase space distribution of the neutron outgoing flux and the eight-dimensional phase space distribution of the neutron flux; wherein, the neutron transport process is characterized by the relationship characteristics between the neutron incident flux and the neutron outgoing flux.

[0056] As described above, it can be seen from Equation (2) that the neutron dynamic transport process within the voxel grid can be characterized by the relationship between the neutron incident flux and the corresponding neutron outgoing flux. At this time, the neutron transport process within the voxel grid can be simulated based on the neutron incident flux in the eight-dimensional phase space by the Monte Carlo method, so as to obtain the eight-dimensional phase space distribution of the neutron outgoing flux and the eight-dimensional phase space distribution of the neutron flux. And based on Equation (2), it can be known that the neutron transport process can be characterized by the relationship between the neutron incident flux and the neutron outgoing flux. At this time, a large number of neutron outgoing fluxes and neutron flux distributions can be obtained by using the Monte Carlo method, and the input-output prediction model of the voxel grid can be trained by using the given neutron incident flux and the corresponding neutron outgoing flux and neutron flux distribution, that is, a physical encoding layer is constructed, so as to enable the voxel grid to quickly respond to the given neutron incident flux to predict the corresponding neutron outgoing flux and neutron flux distribution. And the construction process of the physical encoding layer, that is, the neutron transport process within each voxel grid can be explained based on Equation (2), and the interpretability is relatively strong.

[0057] Among them, the input-output prediction model is a neural network model, and this application does not impose any restrictions on the model type.

[0058] In some embodiments, in the time phase space, the time distribution of the neutron incident flux is discretely calculated based on the asymptotic state approximation to correspondingly obtain the time distribution of the neutron outgoing flux and the time distribution of the neutron flux. That is to say, in the time dimension, the time derivative term of the equation is processed by the asymptotic state approximation to obtain the change of the neutron flux with time in the time phase space, that is, to obtain the time distribution of the neutron flux and the time distribution of the neutron outgoing flux. Among them, the processing of the time derivative term of the equation by the asymptotic state approximation usually refers to the processing method of the time derivative part when discretizing or approximating a time-dependent differential equation. In some embodiments, usually with 1% of the treatment time as the time interval, the time variable of the neutron incident flux is discretely calculated. For example, the forward Euler format can be used for discrete calculation.

[0059] In some embodiments, in the energy phase space, the energy distribution of the neutron incident flux is discretely calculated based on the group approximation to correspondingly obtain the energy distribution of the neutron outgoing flux and the energy distribution of the neutron flux. That is to say, in the energy dimension, the "group approximation" method is used for discrete processing, that is, it is considered that the neutron flux does not change within a certain energy range, and then the energy distribution of the neutron outgoing flux will not change either.

[0060] In some implementations, in the three-dimensional angular phase space, the three-dimensional angular distribution of the neutron incident flux is discretely calculated based on quadrature discretization to correspondingly obtain the three-dimensional angular distribution of the neutron outgoing flux and the three-dimensional angular distribution of the neutron flux.

[0061] In some embodiments, in a three-dimensional spatial phase space, the three-dimensional spatial distribution of the neutron incident flux is discretely calculated based on quadrature discretization to correspondingly obtain the three-dimensional spatial distribution of the neutron outgoing flux and the three-dimensional spatial distribution of the neutron flux.

[0062] Among them, for the three-dimensional angular phase space, S N quadrature points can be used for discrete calculation to obtain the integral value of the three-dimensional angular phase space, that is, the three-dimensional angular distribution of the neutron flux within the voxel grid, and the Monte Carlo method is used to simulate the three-dimensional angular distribution of the neutron outgoing flux. In some embodiments, as Figure 4 shown, N classical quadrature points are selected in the three-dimensional angular phase space to obtain the distribution function values of neutrons in each discrete angular direction, and the integral value of the three-dimensional angular phase space, that is, the angular distribution of the neutron flux, can be approximately obtained by integrating the distribution function values in all discrete angular directions using Equation (3).

[0063] (3)

[0064] Among them, for the three-dimensional spatial phase space, discrete calculation is performed according to the quadrature points of the classical multi-dimensional plane, and the integral value of the three-dimensional spatial phase space, that is, the three-dimensional spatial distribution of the neutron flux, can be approximately obtained by solving Equation (3) in the form of weighted summation, and the Monte Carlo method is used to simulate the three-dimensional spatial distribution of the neutron outgoing flux.

[0065] That is, when the neutron incident flux is given , the Monte Carlo method is used to simulate the neutron transport process within the voxel grid of the neutron incident flux, that is, the relationship characteristics between the neutron outgoing flux and the neutron incident flux are obtained based on Equation (2), so as to obtain the neutron flux distribution and the neutron outgoing flux of the voxel grid in the eight-dimensional phase space , and an input-output prediction model of the voxel grid is trained based on this, that is, a physical coding layer of the voxel grid is constructed, enhancing the interpretability of the physical coding layer.

[0066] S3. Obtain the CT image of the target object, and match at least one of the voxel grids according to the CT image and the boron drug planned concentration to obtain the voxel model of the target object; S4. Perform online coupling on the physical coding layers of all the voxel grids in the voxel model to obtain the neutron flux distribution of the voxel model.

[0067] Specifically, when a physical coding layer is constructed for each voxel grid, the voxel model of the target object can be obtained based on the specific information of the target object, and online coupling is performed on the physical coding layers of all the voxel grids in the voxel model to obtain the neutron flux distribution of the voxel model.

[0068] In some embodiments, obtaining a CT image of a target object and matching at least one of the voxel grids according to the CT image and the planned boron drug concentration to obtain a voxel model of the target object includes: obtaining a CT image of the target object, determining the cancerous region in the CT image, and combining the boron drug concentration (planned boron drug concentration) in a pre-determined clinical treatment plan to match the corresponding at least one voxel grid, thereby obtaining a voxel model of the target object through the matched voxel grids.

[0069] Furthermore, the cancerous region can actually be composed of multiple voxel grids, that is, the voxel grid is a more microscopic division unit. For example, according to the CT image, the cancerous region can be determined to be the brain of the target object, and there are multiple tissue materials in the brain, then this region matches multiple voxel grids. In some other embodiments, a certain cancerous region may be precise enough, and in this case, one voxel grid is matched. In addition, the voxel model of the target object can also include the voxel grids corresponding to the cancerous region and the voxel grids of the normal region. That is, all voxel grids are completely matched according to various human tissue materials of the target object to form a voxel model.

[0070] For example, if step S1 sets N voxel grids and the voxel model of a certain target object is composed of M voxel grids (M is not greater than N), then the physical coding layers of the M voxel grids can be coupled, and the neutron efflux and neutron flux distribution corresponding to each can be predicted through the physical coding layers of the M voxel grids. Then, based on the neutron flux distribution within all voxel grids, the neutron flux distribution in the voxel model of the target object can actually be obtained.

[0071] Specifically, online coupling the physical coding layers of all the voxel grids in the voxel model includes steps S41 to S43.

[0072] S41. Set a voxel grid in the voxel model as the initial grid and give a neutron incident flux to the initial grid.

[0073] S42. Predict the corresponding neutron efflux and neutron flux distribution based on the physical coding layer of the initial grid.

[0074] S43. Use the neutron efflux of the initial grid as the neutron incident flux of the next voxel grid to predict the corresponding neutron efflux and neutron flux distribution based on the physical coding layer of the next voxel grid until all the voxel grids are traversed.

[0075] For example, the voxel model of a certain target object is composed of M voxel grids, and a neutron incident flux J + set in the given treatment plan is given, and a voxel grid M1 is set as the initial grid. Input J + into M1 to predict the corresponding neutron efflux J1 through the physical coding layer of M1- and the neutron flux distribution within M1, and take the neutron efflux J1 of M1 - as the neutron influx of the next voxel grid M2, so as to predict the corresponding neutron efflux J2 through the physical encoding layer of M2 - and the neutron flux distribution within M2, and so on until all the voxel grids are traversed, obtaining the neutron flux distributions within all the voxel grids, thereby realizing the neutron flux distribution within the voxel model.

[0076] Figure 5 The figure shows an application schematic diagram of the neutron transport simulation method provided by an embodiment of the present application. That is, a variety of classical voxel grids are formed by various tissue materials of the human body and different concentrations of boron drugs. And based on the given neutron influx, the corresponding neutron efflux and neutron flux distribution of each voxel grid, train the input-output prediction model of the voxel grid, thereby offline generating the physical encoding layer of each voxel grid, so as to represent the dynamic transport process of neutrons in various voxel grids based on the physical encoding layer, that is, it can quickly predict the corresponding neutron efflux and the neutron flux distribution within the voxel grid for the given neutron influx in the voxel grid. Then, according to the actual problems of the patient, online couple each voxel grid to form a voxel model targeted at the patient, and thereby based on the neutron flux distributions of all the voxel grids in the voxel model, that is, quickly obtain the neutron flux distribution within the voxel model.

[0077] Therefore, the present application can offline generate the physical encoding layer of the voxel grid without the need to know any information about the patient in advance. And different from the very complex material distribution of the voxel model, the geometric structure of the voxel grid is very simple, so that it only takes less computational cost to solve Equation (2) to obtain the neutron flux distribution and neutron efflux. In addition, different from the idea of directly solving the particle transport equation in the continuous infinite-dimensional phase space by the traditional Monte Carlo method, online coupling each voxel grid only needs to perform calculations on millions of voxel grids in the eight-dimensional phase space, limiting the dimension of the problem within the range of millions, improving the computational efficiency and computational accuracy. Finally, the online coupling method ensures that the differences between different voxel models can be considered, ensuring the strong generalization ability of the present method.

[0078] The protection scope of the neutron transport simulation method described in the embodiments of the present application is not limited to the execution order of the steps listed in this embodiment. Any solution realized by adding or subtracting steps of the prior art and replacing steps according to the principle of the present application is included in the protection scope of the present application.

[0079] The present application also provides a BNCT treatment planning system, which is used to obtain the neutron flux distribution in the voxel model of the target object based on the method described above to generate a BNCT treatment plan. That is, in some embodiments, the BNCT-TPS can use the neutron transport simulation method provided in the embodiments of the present application to quickly obtain the neutron flux distribution in the body of the target patient, so as to calculate the BNCT dose and generate a BNCT treatment plan.

[0080] The embodiments of the present application also provide a neutron transport simulation device, which can implement the neutron transport simulation method described in the present application. However, the implementation devices of the neutron transport simulation device described in the present application include, but are not limited to, the structures of the neutron transport simulation devices 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 protection scope of the present application.

[0081] Figure 6 The structural schematic diagram of the neutron transport simulation device described in the embodiments of the present application is shown, as Figure 7 shown, the neutron transport simulation device includes: A setting module 41, configured to set a plurality of voxel grids; the voxel grids are various human tissue materials corresponding to different boron drug concentrations; A physical coding module 42, configured to perform offline coding on each of the voxel grids to obtain a corresponding physical coding layer, and the physical coding layer is used to predict the corresponding neutron outgoing flux and neutron flux distribution based on the neutron incident flux of the voxel grid; An acquisition module 43, configured to acquire the CT image of the target object, and match at least one of the voxel grids according to the CT image and the boron drug planned concentration to obtain the voxel model of the target object; A calculation module 44, configured to perform online coupling on the physical coding layers of all the voxel grids in the voxel model to obtain the neutron flux distribution of the voxel model.

[0082] It should be noted that the structures and principles of the setting module, the physical coding module, the acquisition module, and the calculation module, as well as the beneficial effects obtained by the device, are the same as those in the above embodiments, and will not be elaborated herein.

[0083] Based on the same technical concept, the neutron transport simulation method provided in the embodiments of the present application can be implemented on the terminal side or the server side.

[0084] Please refer to Figure 7, which is an optional hardware structure diagram of a terminal provided by an embodiment of the present application. The terminal may be a mobile phone, a computer device, a tablet device, a personal digital processing device, a factory background processing device, etc. The terminal includes: at least one processor 61, a memory 62, at least one network interface 64, and a user interface 63. Each component in the device is coupled together through a bus system 65. It can be understood that the bus system 65 is used to realize the connection and communication between these components. In addition to the data bus, the bus system also includes a power bus, a control bus, and a status signal bus.

[0085] Among them, the user interface 63 may include a display, a keyboard, a mouse, a trackball, a click gun, a key, a button, a touchpad, or a touch screen, etc.

[0086] It can be understood that the memory 62 may be a volatile memory or a non-volatile memory, and may also include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM, Read Only Memory), a programmable read-only memory (PROM, Programmable Read-Only Memory), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM, StaticRandom Access Memory), synchronous static random access memory (SSRAM, Synchronous StaticRandom Access Memory). The memory characterized in the embodiments of the present application is intended to include but not be limited to these and any other suitable categories of memories.

[0087] The memory 62 in the embodiments of the present application is used to store various categories of data to support the operation of the terminal. Examples of these data include: any executable program for operating on the terminal 60, such as an operating system 621 and an application program 622; the operating system 621 contains various system programs, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks. The application program 622 may include various application programs, such as a media player (MediaPlayer), a browser (Browser), etc., for implementing various application services. Implementing the neutron transport simulation method provided by the embodiments of the present application may be included in the application program 622.

[0088] The method disclosed in the embodiments of the present application can be applied to or implemented by the processor 61. The processor 61 may be an integrated circuit chip with signal processing capabilities. In the implementation process, the steps of the above method can be completed by the integrated logic circuit in the hardware of the processor 61 or instructions in the form of software. The above processor may 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 61 can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor 61 may be a microprocessor or any conventional processor, etc. Combining the steps of the accessory optimization method provided in the embodiments of the present application can be directly embodied as being executed and completed by the hardware decoding processor, or executed and completed by a combination of the hardware and software modules in the decoding processor. The software module may be located in the storage medium, and this storage medium is located in the memory. The processor reads the information in the memory and combines its hardware to complete the steps of the foregoing method.

[0089] In an exemplary embodiment, the terminal 60 may be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), or complex programmable logic devices (CPLDs) for executing the foregoing method.

[0090] The embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored, and when the program is called by a processor, it implements the neutron transport simulation method provided by the present application.

[0091] Among them, the computer-readable storage medium may be a tangible device that can hold and store instructions used by the instruction execution device. The computer-readable storage medium may be, for example (but not limited to), an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: portable computer disks, hard disks, random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), static random access memories (SRAMs), portable compact disc read-only memories (CD-ROMs), digital versatile discs (DVDs), memory sticks, floppy disks, and mechanical encoding devices.

[0092] The computer-readable programs represented herein can be downloaded from a computer-readable storage medium to various computing / processing devices, or downloaded to an external computer or an external storage device through a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network adapter or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0093] The above embodiments are only illustrative of the principles and effects of the present application and are not intended to limit the present application. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed in the present application should still be covered by the claims of the present application.

Claims

1. A neutron transport simulation method, characterized in that: The method is applicable to a BNCT treatment planning system, and the method comprises: Setting a plurality of voxel grids; the voxel grids are various types of human tissue materials corresponding to different boron drug concentrations; Offline encoding is performed on each voxel grid to obtain a corresponding physical coding layer, wherein the physical coding layer is used to predict a corresponding neutron outflow and neutron flux distribution based on a neutron inflow of the voxel grid; Acquire a CT image of the target object, and match at least one of the voxel grids according to the CT image and the planned concentration of the boron drug to acquire a voxel model of the target object; The physical coding layers of all the voxel grids in the voxel model are coupled online to obtain the neutron flux distribution of the voxel model.

2. The neutron transport simulation method according to claim 1, characterized in that: Offline encoding each of the voxel grids to obtain a corresponding physical coding layer includes: Customizing the neutron incident flow and the neutron outgoing flow in an eight-dimensional phase space; the eight-dimensional phase space includes a time phase space, an energy phase space, a three-dimensional angle phase space, and a three-dimensional space phase space; Giving different neutron incident flows to each voxel grid to correspondingly obtain the neutron outgoing flow and the neutron flux distribution in the eight-dimensional phase space; The physical coding layer of each voxel grid is offline encoded based on the given neutron incident flux and the corresponding neutron outgoing flux and the neutron flux distribution.

3. The neutron transport simulation method according to claim 2, characterized in that: Giving different neutron incident flows to each voxel grid to correspondingly obtain the neutron outgoing flow and the neutron flux distribution in the eight-dimensional phase space includes: The neutron transport process in the voxel grid is simulated based on the eight-dimensional phase space distribution of the neutron incident flow to obtain the eight-dimensional phase space distribution of the neutron outgoing flow and the eight-dimensional phase space distribution of the neutron flux; wherein the neutron transport process is characterized by the characteristic relationship between the neutron incident flow and the neutron outgoing flow.

4. The neutron transport simulation method according to claim 2, characterized in that: The method comprises: In combination with the eight-dimensional phase space definition of the neutron incident flow and the neutron outgoing flow, spatial integration is performed on the voxel grid to characterize the neutron transport process within the voxel grid as the relationship characteristic between the neutron incident flow and the neutron outgoing flow.

5. The neutron transport simulation method according to claim 3, characterized in that: Simulating the neutron transport process in the voxel grid based on the eight-dimensional phase space distribution of the neutron incident flow to obtain the eight-dimensional phase space distribution of the neutron outflow and the eight-dimensional phase space distribution of the neutron flux includes: In the time phase space, based on asymptotic state approximation, discrete calculation is performed on the time distribution of the neutron incident flow, so as to correspondingly obtain the time distribution of the neutron outgoing flow and the time distribution of the neutron flux; In the energy phase space, based on the clustering approximation, a discrete calculation is performed on the energy distribution of the neutron incident flow to correspondingly obtain the energy distribution of the neutron outgoing flow and the energy distribution of the neutron flux; In the three-dimensional angular phase space, based on the quadrature discretization, a discrete calculation is performed on the three-dimensional angular distribution of the neutron incident flow, so as to correspondingly obtain the three-dimensional angular distribution of the neutron outgoing flow and the three-dimensional angular distribution of the neutron flux; In the three-dimensional phase space, the three-dimensional spatial distribution of the neutron incident flow is discretely calculated based on the quadrature discretization to obtain the three-dimensional spatial distribution of the neutron outgoing flow and the three-dimensional spatial distribution of the neutron flux.

6. The neutron transport simulation method according to claim 1, characterized in that: Online coupling of the physical coding layers of all the voxel grids in the voxel model comprises: Setting a voxel grid in the voxel model as an initial grid, and giving a neutron incident flow to the initial grid; Predicting the corresponding neutron emission flow and neutron flux distribution based on the physical coding layer of the initial grid; The neutron outflow of the initial grid is used as the neutron incident flow of the next voxel grid, so as to predict the corresponding neutron outflow and neutron flux distribution based on the physical coding layer of the next voxel grid until all the voxel grids are traversed.

7. A BNCT treatment planning system, characterized in that: The system is used for acquiring neutron flux distribution in a voxel model of a target object based on the method according to any one of claims 1 to 6 to generate a BNCT treatment plan.

8. A neutron transport simulation device, characterized in that: The device comprises: A setting module, used to set a plurality of voxel grids; the voxel grids are various types of human tissue materials corresponding to different boron drug concentrations; A physical coding module, used for performing offline coding on each voxel grid to obtain a corresponding physical coding layer, wherein the physical coding layer is used for predicting a corresponding neutron outflow and neutron flux distribution based on a neutron inflow of the voxel grid; An acquisition module, used for acquiring a CT image of a target object, and matching at least one of the voxel grids according to the CT image and the planned concentration of the boron drug to acquire a voxel model of the target object; A calculation module is used for online coupling of the physical coding layers of all the voxel grids in the voxel model to obtain the neutron flux distribution of the voxel model.

9. A terminal, characterized in that: include: 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 method according to any one of claims 1 to 6.

10. A computer storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images