Method and apparatus for establishing three-dimensional model, and radiotherapy system

Abstract

Provided in the present description are a method and apparatus for establishing a three-dimensional model, and a radiotherapy system. A target region is acquired on the basis of a medical image; the material of each voxel unit in the target region is defined on the basis of a target template in a material template library matched with the target region, wherein a candidate template available for matching in the material template library includes a homogeneous material template and a heterogeneous material template; and a three-dimensional model is established on the basis of the material of each voxel unit in each target region. In this solution, differences in image quality of medical images and variations in tissues and organs of individual patients are taken into consideration. By matching a target template, the material of each voxel unit in each target region is more specifically defined by using a different strategy, so as to establish a three-dimensional model using the more suitable material. The model more closely resembles actual characteristics of tissues and organs of patients, such that treatment plans can more accurately evaluate dose limits for normal tissues and prescribed doses for tumors.

Description

A method, apparatus, and radiotherapy system for creating a three-dimensional model Technical Field

[0001] This application relates to the field of radiotherapy technology, and in particular to a method, apparatus and radiotherapy system for creating a three-dimensional model. Background Technology

[0002] With the development of atomic science, radiotherapy methods such as cobalt-60, linear accelerators, and electron beams have become one of the main means of cancer treatment. However, traditional photon or electron therapy is limited by the physical conditions of radiation itself. While killing tumor cells, it also damages a large amount of normal tissue along the beam path. In addition, due to the different sensitivities of tumor cells to radiation, traditional radiotherapy is often ineffective in treating more radiation-resistant malignant tumors (such as glioblastoma multiforme and melanoma).

[0003] To reduce radiation damage to surrounding normal tissues, the concept of targeted therapy in chemotherapy has been applied to radiotherapy. For highly radiation-resistant tumor cells, radiation sources with high relative biological effectiveness (RBE) are being actively developed for radiotherapy, such as proton therapy, heavy ion therapy, and neutron capture therapy. Neutron capture therapy combines the above two concepts; for example, boron neutron capture therapy (BNCT) utilizes the specific accumulation of boron-containing drugs on tumor cells, combined with precise beam modulation, to provide a better cancer treatment option than traditional radiation.

[0004] Three-dimensional models are widely used in scientific experimental analysis and simulation. For example, in the field of nuclear radiation and protection, to simulate the absorbed dose of the human body under certain radiation conditions to help operators formulate treatment plans, it is often necessary to use computer technology to process medical image data to create accurate three-dimensional models required by Monte Carlo software, and then combine this with Monte Carlo software for simulation calculations. In the field of neutron capture therapy, when creating three-dimensional models required by Monte Carlo software based on medical image data and performing dose calculations and evaluations, it is necessary to define the material information reflected by each voxel unit in the model. The accuracy and precision of the materials determine the reliability of the dose calculation results. During the treatment plan formulation process, a significant amount of time is usually spent defining the materials in various regions. At the same time, the accuracy of medical image acquisition equipment and the quality of the obtained images are limited. The three-dimensional models converted from the parameters of medical image data are often not accurate enough, especially regarding the definition of materials. If there are deviations, it will affect the accuracy of the calculated dose distribution. Treatment plans based on these calculation results will fail to achieve the desired effects and may even cause unexpected damage to normal tissues. Summary of the Invention

[0005] This specification provides a method, apparatus, and radiotherapy system for creating a three-dimensional model, in order to solve the problem that existing methods for creating three-dimensional models are difficult to reflect the differences in target sites and surgical requirements.

[0006] To address the aforementioned technical problems, this specification provides a method for establishing a three-dimensional model, comprising: acquiring a target region based on medical images; defining the material of voxel units within the target region based on a target template in a material template library matched by the target region, wherein the candidate templates that can be matched in the material template library include homogeneous material templates and heterogeneous material templates; and establishing a three-dimensional model based on the material of voxel units within each target region.

[0007] In some embodiments, defining the material of voxel units within the target region based on a target template in a material template library matched with the target region includes: matching candidate templates in a material template library as target templates based on the tissue / organ type and / or image value distribution corresponding to the target region.

[0008] In some embodiments, defining the material of voxel units within the target region based on a target template in a material template library matched with the target region includes: obtaining the tissue / organ type corresponding to the target region; matching a candidate template corresponding to the tissue / organ type as the target template, wherein the candidate template corresponding to the tissue / organ type is a homogeneous material template or a heterogeneous material template.

[0009] In some embodiments, defining the material of the voxel unit in the target region based on the target template in the material template library matched with the target region includes: matching the candidate template corresponding to the first reference tissue / organ type in the material template library as the target template; wherein, the image value reference range of the first reference tissue / organ type and the image value distribution of the target region satisfy a first preset condition, and the candidate template corresponding to the first reference tissue / organ type is a homogeneous material template or a heterogeneous material template.

[0010] In some embodiments, defining the material of voxel units within the target region based on a target template in a material template library matched with the target region includes: matching a homogeneous material template or a heterogeneous material template as a target template according to the relationship between the image value distribution of the target region and the image value reference range of each tissue and organ type.

[0011] In some embodiments, defining the material of voxel units within the target region based on a target template in a material template library matched with the target region includes: matching a heterogeneous material template as a target template in response to a second preset condition that the image value distribution of the target region and the target reference range satisfy the second preset condition, wherein the target reference range is an image value reference range of the tissue / organ type corresponding to the target region or an image value reference range of a second reference tissue / organ type, the second reference tissue / organ type being determined based on the image value distribution of the target region and the image value reference range of each tissue / organ type; and matching a homogeneous material template as a target template in response to a second preset condition that the image value distribution of the target region and the target reference range do not satisfy the second preset condition.

[0012] In some embodiments, the second preset condition is determined based on the resolution of the medical image.

[0013] In some embodiments, the step of matching a heterogeneous material template as a target template in response to the image value distribution of the target region and the target reference range satisfying a second preset condition includes: matching a heterogeneous material template as a target template in response to the target sub-interval of the image value distribution interval of the target region being located within the target reference range, wherein the target sub-interval is a value interval containing the average image value of the target region.

[0014] In some embodiments, defining the material of the voxel unit within the target region based on the target template in the material template library matched with the target region includes: matching a homogeneous material template or a heterogeneous material template as the target template based on the OAR information of the target region.

[0015] In some embodiments, the OAR information of the target area includes whether the target area is an OAR, or whether the target area is a primary OAR; the matching of a homogeneous material template or a heterogeneous material template as a target template based on the OAR information of the target area includes: matching a heterogeneous material template as a target template in response to the OAR information of the target area being an OAR, or matching a heterogeneous material template as a target template in response to the OAR information of the target area being a primary OAR.

[0016] In some embodiments, defining the material of voxel units within the target region based on a target template in a material template library matched with the target region includes: obtaining the tissue / organ type corresponding to the target region; outputting a prompt message in response to the fact that the tissue / organ type corresponding to the target region is different from a third reference tissue / organ type; and ensuring that the image value reference range of the third reference tissue / organ type and the image value distribution of the target region satisfy a third preset condition.

[0017] In some embodiments, defining the material of voxel units within the target region based on a target template in a material template library matched with the target region includes: acquiring at least two tissue / organ types associated with the image value distribution of the target region; determining the most irradiated tissue / organ type among the at least two tissue / organ types; and matching a homogeneous material template of the most irradiated tissue / organ type as the target template.

[0018] In some embodiments, the method further includes constructing heterogeneous material templates in the material template library by at least one of the following methods: constructing a formula based on the relationship between the density of tissues and organs and image values; constructing a formula based on the relationship between the elemental composition of tissues and organs and image values; constructing a formula based on key parameters of image values ​​and corresponding densities; and constructing a formula based on key parameters of image values ​​and corresponding elemental compositions.

[0019] A second aspect of this specification provides a three-dimensional model building apparatus, comprising: a first acquisition unit for acquiring a target region based on medical images; a definition unit for defining the material of voxel units within the target region based on a target template in a material template library matched by the target region, wherein the candidate templates that can be matched in the material template library include homogeneous material templates and heterogeneous material templates; and a building unit for building a three-dimensional model based on the material of voxel units within each target region.

[0020] In some embodiments, the defining unit is further configured to match candidate templates in a material template library as target templates based on the tissue / organ type and / or image value distribution corresponding to the target region.

[0021] In some embodiments, the defining unit is further configured to match a homogeneous material template or a heterogeneous material template as a target template based on the relationship between the image value distribution of the target region and the image value reference range of each tissue and organ type.

[0022] In some embodiments, the defining unit is further configured to match a homogeneous material template or a heterogeneous material template as a target template based on the OAR information of the target region.

[0023] In some embodiments, the apparatus further includes: a second acquisition unit, configured to acquire at least two tissues or organs associated with the image value distribution of the target region; a determination unit, configured to determine the most irradiated tissue or organ type among the at least two tissue or organ types; the definition unit is further configured to match a homogeneous material template of the most irradiated tissue or organ type as a target template.

[0024] In some embodiments, the apparatus further includes a construction unit for constructing heterogeneous material templates in the material template library by at least one of the following methods: construction based on the relationship between the density of tissues and organs and image values; construction based on the relationship between the elemental composition of tissues and organs and image values; construction based on key parameters of image values ​​and corresponding densities; and construction based on key parameters of image values ​​and corresponding elemental compositions.

[0025] A third aspect of this specification provides a radiotherapy system, comprising: a beam irradiation device for generating a radiation beam; a treatment planning module for generating a treatment plan; and a control module for controlling the beam irradiation device to generate the radiation beam according to the treatment plan. The treatment planning module includes a three-dimensional model building device, which comprises: a first acquisition unit for acquiring a target region based on medical images; a definition unit for defining the material of voxel units within the target region based on a target template in a material template library that matches the target region, wherein the matchable candidate templates in the material template library include homogeneous material templates and heterogeneous material templates; and a building unit for building a three-dimensional model based on the material of the voxel units within each target region.

[0026] A fourth aspect of this specification provides an electronic device including a processor and a memory for storing processor-executable instructions, wherein the processor, when executing the instructions, implements the steps of the method described in any of the first aspects.

[0027] A fifth aspect of this specification provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the method described in any of the first aspects.

[0028] The method, apparatus, and radiotherapy system for establishing the aforementioned three-dimensional model provided in this specification acquire the target region based on medical images, and define the material of voxel units within the target region based on target templates in a material template library that matches the target region. The material template library includes homogeneous material templates and heterogeneous material templates, and then establishes a three-dimensional model based on the material of voxel units within each target region. Using heterogeneous material templates allows for a more detailed and accurate representation of the material distribution in the target region based on medical images, more closely reflecting the patient's actual physical condition, which is beneficial for establishing a high-precision three-dimensional model. Using homogeneous material templates avoids the problem of image acquisition errors causing deviations in the image values ​​themselves, and then using overly detailed methods to define the material of voxel units based on these deviations, thus exacerbating the image acquisition error. Furthermore, using homogeneous material templates can reduce the computational load to some extent. This approach considers the varying image quality of medical images and the differences in tissues and organs among individual patients. By matching target templates and employing different strategies, it more effectively defines the material of each voxel unit within each target region. This allows for the creation of a more suitable 3D model, which more closely approximates the actual characteristics of the patient's tissues and organs. This enables treatment planning to more accurately assess dose limitations for normal tissues and prescribed doses for tumors, ensuring the safety of treatment plans and improving radiotherapy efficacy. Furthermore, it avoids unnecessary computational burden, facilitating widespread application. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0030] Figure 1 is a schematic diagram of a method for creating a three-dimensional model provided in this specification;

[0031] Figure 2 is another schematic diagram of the method for creating a three-dimensional model provided in this specification;

[0032] Figure 3 is another schematic diagram of the method for creating a three-dimensional model provided in this specification;

[0033] Figure 4 is a schematic diagram showing the relationship between the distribution of image values ​​in the target region and the reference range of brain tissue.

[0034] Figure 5 is a schematic diagram showing the relationship between the image value distribution of the target area and the reference range of the mucosal tissue.

[0035] Figure 6 is another schematic diagram of the method for creating a three-dimensional model provided in this specification;

[0036] Figure 7 is another schematic diagram of the method for creating a three-dimensional model provided in this specification;

[0037] Figure 8 is another schematic diagram of the method for creating a three-dimensional model provided in this specification;

[0038] Figure 9 is another schematic diagram of the method for creating a three-dimensional model provided in this specification;

[0039] Figure 10 is another schematic diagram of the method for creating a three-dimensional model provided in this specification;

[0040] Figure 11 is another schematic diagram of the method for creating a three-dimensional model provided in this specification;

[0041] Figure 12 is a schematic diagram of the device for creating a three-dimensional model provided in this specification;

[0042] Figure 13 is a schematic diagram of a radiotherapy system;

[0043] Figure 14 shows another schematic diagram of a radiotherapy system;

[0044] Figure 15 is a schematic diagram of the electronic device provided in this specification. Detailed Implementation

[0045] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application. The various embodiments in this specification, such as the examples provided "in some embodiments" or "in other embodiments," can each be used as an embodiment on their own, or they can be combined with at least one other embodiment to form some new embodiments, which can still be implemented and achieve corresponding beneficial effects.

[0046] This specification provides a method for creating a three-dimensional model. This method can be used to generate treatment plans for radiotherapy methods mentioned in the background section. Specifically, the radiotherapy method can be boron neutron capture therapy. Boron neutron capture therapy is merely an example; the method for creating the three-dimensional model provided in this specification can also be used for other radiotherapy methods, which will not be detailed here.

[0047] As shown in Figure 1, the method includes the following steps S10 to S30.

[0048] S10: Determine the target region based on medical images.

[0049] Medical images can include CT images, PET-CT images, MRI images, etc. They can also be other three-dimensional medical images reconstructed from planar images. In practice, images that can be processed to obtain CT values ​​or pseudo-CT values ​​can be used to determine the target area.

[0050] S10 can be determined by the operator (such as a doctor) based on experience; it can also be determined automatically based on a preset method. The automatic determination method can adopt existing technology, and the operator can confirm or modify it after automatic determination.

[0051] In some embodiments, the target region can be a region of interest (ROI) determined based on medical images. The target region can be a critical organ, such as the eye or liver; an important tissue, such as bone or brain tissue; a tumor, such as a brain tumor; or a special medium that needs to be defined when building a 3D model, such as air. The electronic device automatically defines or reads the image values ​​from the medical images, or the operator manually defines the boundaries of each target region (e.g., the boundaries of each ROI). By defining the boundaries of all target regions, the 3D model of the area to be irradiated is divided into different target regions, that is, each voxel unit (i.e., voxel cell) in the 3D model is assigned to a corresponding target region. In the 3D model corresponding to the area to be irradiated, at least one target region can be operated using S20-S30, while other target regions can be operated using existing methods; alternatively, all target regions can be operated using S20-S30.

[0052] S20: Define the material of the voxel unit in the target region based on the target template in the material template library that matches the target region. The candidate templates that can be matched in the material template library include homogeneous material templates and heterogeneous material templates.

[0053] Image values ​​are parameters of medical images. For example, they can be the CT value or HU value of each pixel unit in medical images such as CT and PET-CT, or pseudo-CT values ​​or pseudo-HU values ​​converted from other images to CT images, or image parameters from other medical images. In some embodiments, image values ​​can be used to characterize the density of tissues and organs.

[0054] Defining the material of a voxel unit may include defining at least one of density, elemental composition, one or more atomic numbers, one or more mass numbers, and average ionization energy. It is understood that if one of these factors is used to define the material of each voxel unit, the results of the other factors can be derived by calculation based on the conversion relationships between the aforementioned factors disclosed in the art. For example, the density can be calculated by defining the elemental composition of the voxel unit as its material definition. In some embodiments, the elemental composition may include each element and its corresponding mass percentage, such as the mass percentages of elements like hydrogen (H), carbon (C), nitrogen (N), oxygen (O), sodium (Na), magnesium (Mg), phosphorus (P), sulfur (S), chlorine (Cl), argon (Ar), potassium (K), calcium (Ca), and iron (Fe).

[0055] The above-described S20 can determine a target template matching the target region based on the type of tissue or organ corresponding to the target region. For example, a material template library pre-sets candidate templates corresponding to various tissue and organ types, such as eyes, brain, mucous membranes, and glands. In some embodiments, after delineating the target region in a patient's medical image, the operator (e.g., a doctor) determines the type of tissue or organ corresponding to the target region and inputs this type into an electronic device. The electronic device automatically determines a matching target template based on the input tissue or organ type. In other embodiments, the type of tissue or organ corresponding to the target region can also be automatically determined based on a preset method, which can employ existing technology. After automatically determining the tissue or organ type, it can be confirmed or modified by the operator.

[0056] In addition, the operator can select whether the target area uses a homogeneous material template or a heterogeneous material template, and the electronic equipment will automatically determine the matching target template.

[0057] The aforementioned S20 can also determine a target template matching the target region based on the image value distribution corresponding to the target region. Different tissue and organ types typically exhibit different image value distributions in medical images; for example, the range of image values ​​differs, as do the uniformity or dispersion of the image value distribution. Therefore, a target template can be automatically matched based on the image value distribution of the target region in the medical image. For instance, a material template library may pre-set candidate templates corresponding to various tissue and organ types, such as the eye, brain, mucous membrane, and glands. It can also pre-determine the image value distribution data corresponding to each tissue and organ type. After delineating the target region in the patient's medical image, the medical image with the delineated target region is input into an electronic device. The electronic device compares the medical image distribution of the target region with the image value distribution data corresponding to each tissue and organ type to determine the matching target tissue and organ type, and thus determine the matching target template.

[0058] In some embodiments, S20 can further combine the organ type and image value distribution corresponding to the target area to determine the target template that matches the target area. For example, the material template library pre-sets candidate templates corresponding to various organ types such as eyes, brain, mucous membranes, and glands. Each organ type (eyes, brain, glands) has multiple candidate templates. After delineating the target area in the patient's medical image, the medical image with the delineated target area is input into the electronic device, along with the organ type corresponding to the target area. The electronic device compares the medical image distribution of the target area with the image value distribution data corresponding to each organ type, and further combines the organ type to determine the matching target template. Specifically, the electronic device can first determine several candidate target templates based on the organ type, and then further determine the target template from these candidate templates based on the image value distribution; or it can first determine several candidate target templates based on the image value distribution, and then further determine the target template from these candidate templates based on the organ type.

[0059] The material template library may include only homogeneous material templates of various tissues and organs, or it may include only heterogeneous material templates of various tissues and organs. In some embodiments, at least some tissues and organs in the material template library have both homogeneous and heterogeneous material templates.

[0060] In a homogeneous material template, the types and proportions of the constituent elements corresponding to various image values ​​are the same, and the density is uniform. That is, a homogeneous material template is designed to use the same material definition for all voxel units within the target area. In a heterogeneous material template, the types and / or proportions of the constituent elements corresponding to (at least two) ranges of image values ​​are different. Consequently, the density within the target area may not be uniform. That is, a heterogeneous material template is designed to divide voxel units within the target area into different material definitions based on image values. A specific manifestation of a heterogeneous material template can be: a heterogeneous material template includes multiple sub-ranges of image values ​​and the proportions of each constituent element corresponding to each sub-range; the proportions of the same constituent element corresponding to at least two sub-ranges are not completely identical.

[0061] There can be only one or more heterogeneous material templates corresponding to a single tissue or organ type. The number of heterogeneous material templates corresponding to a single tissue or organ type can be determined based on the different levels of coarseness in the division of image value sub-intervals.

[0062] In some embodiments, the material template library may include only templates for one type of tissue or organ, and the material template library may include multiple different candidate templates corresponding to that tissue or organ type. For example, the material template library may include only homogeneous material templates and heterogeneous material templates corresponding to the eye.

[0063] In some embodiments, the material template library may include candidate templates corresponding to various tissue and organ types, wherein one tissue or organ type may correspond to one or more candidate templates. One tissue or organ may correspond to multiple candidate templates; specifically, one tissue or organ may correspond to both homogeneous material templates and heterogeneous material templates.

[0064] In some embodiments, a candidate template corresponds to a type of tissue or organ. S20 can first obtain the target tissue or organ type that matches the medical image of the target region, and then further define the material of the voxel unit in the target region based on the candidate template corresponding to the target tissue or organ type.

[0065] In some embodiments, a type of tissue or organ corresponds to multiple candidate templates, as shown in Figure 2. In step S20, the type of tissue or organ corresponding to the target region can be first obtained, and then the candidate template corresponding to that type of tissue or organ can be matched as the target template. The candidate template corresponding to the type of tissue or organ can be a homogeneous material template or a heterogeneous material template. Specifically, the homogeneous material template corresponding to the type of tissue or organ can be used as the target template, or the heterogeneous material template corresponding to the type of tissue or organ can be used as the target template.

[0066] In some embodiments, a type of tissue or organ may correspond to multiple candidate templates, and S20 may also match a candidate template from the candidate template library as the target template based on the distribution of image values.

[0067] In some embodiments, the image value is a unit for measuring tissue density. It is calculated based on the attenuation coefficient of different tissues and organs after absorbing X-rays. Generally, the higher the density of the area, the higher the image value.

[0068] The distribution of image values ​​in the target region refers to the statistical results of the image values ​​of each voxel in the target region of a medical image. For example, the image values ​​may include how many voxels in the target region correspond to each range of image values ​​(i.e., for each range, how many voxels in the target region have image values ​​within that range), or the frequency or percentage of each voxel image value in the target region's medical image in each range, or the mean and / or mode of the image values ​​of each voxel in the target region.

[0069] In some embodiments, S20 matches a candidate template from a candidate template library as the target template based on the image value distribution, as shown in Figure 3. Specifically, it can be: matching a candidate template corresponding to a first reference tissue / organ type in a material template library as the target template; wherein, the image value reference range of the first reference tissue / organ type and the image value distribution of the target region satisfy a first preset condition, and the candidate template corresponding to the first reference tissue / organ type is a homogeneous material template or a heterogeneous material template. Specifically, a homogeneous material template corresponding to the first reference tissue / organ type can be used as the target template, or a heterogeneous material template corresponding to the first reference tissue / organ type can be used as the target template.

[0070] In some embodiments, the first preset condition may be that the mean and mode of the image values ​​of the target region meet the requirements of the image value reference range of the first reference tissue / organ type. For example, a material template library may pre-store multiple candidate templates, each with its corresponding reference tissue / organ type. Based on the reference tissue / organ type, a reference range for the mean image value and / or the reference range for the mode image value can be provided. Then, the mean and / or mode of the image values ​​of all voxels in the medical image of the target region can be statistically analyzed first. It can be determined which reference tissue / organ type's image value mean reference range the mean image value in the statistical results meets, and / or which reference tissue / organ type's image value mode reference range the mode image value in the statistical results meets. If the mean and / or mode image values ​​in the statistical results meet the reference range of a certain reference tissue / organ type A, the candidate template corresponding to the reference tissue / organ type A (i.e., the first reference tissue / organ type) can be used as the target template matched to the target region.

[0071] In other embodiments, when the image value distribution of the target region is such that each value interval corresponds to a certain number of voxels within the target region, the inventors have found that the image values ​​of each voxel within the target region generally conform to a normal distribution. Based on this, the aforementioned first preset condition can be determined based on the normal distribution law. For example, the electronic device pre-stores image value reference ranges corresponding to each tissue / organ type. If the confidence level that the image value distribution of the medical image in the target region matches the image value reference range of tissue / organ type B reaches a first matching confidence threshold, then the candidate template corresponding to tissue / organ type B (i.e., the first reference tissue / organ type) can be used as the target template. Specifically, a homogeneous material template corresponding to tissue / organ type B can be used as the target template, or a heterogeneous material template corresponding to tissue / organ type B can be used as the target template.

[0072] In some embodiments, the confidence level of the match can be determined according to the law of normal distribution. Specifically, the confidence level can be the confidence level corresponding to a target sub-interval in the distribution range of image values ​​of the target region, wherein the target sub-interval is a sub-interval contained within the reference range of image values ​​for organ type C. For example, first calculate the average value μ and standard deviation σ of each voxel image value in the medical image of the target region. If the interval [μ-2σ, μ+2σ] is contained within the reference range of image values ​​for organ type B, then the confidence level of the match reaches 95%; if the interval [μ-3σ, μ+3σ] is contained within the reference range of image values ​​for organ type C, then the confidence level of the match reaches 98% (this correspondence between the interval and the confidence level is based on the fundamental knowledge of Gaussian distribution in mathematics). The above-mentioned first matching confidence threshold can be 95%, 98%, or other predetermined values.

[0073] As shown in Figure 4, the horizontal axis represents image values. The bold black line segment on the horizontal axis represents the reference range of image values ​​for brain tissue. The dotted area represents the actual distribution of image values ​​in the target region's medical image (the vertical axis can represent the number of voxels corresponding to the image values, or the frequency or percentage of voxel occurrences, etc.). The projection of the bidirectional arrow in the dotted area onto the horizontal axis represents the actual image value distribution interval corresponding to a 95% confidence level. As can be seen from Figure 4, the actual image value distribution interval corresponding to a 95% confidence level is within the reference range of image values ​​for brain tissue. In this embodiment, it is assumed that the confidence level of the target region's medical image matching the reference range of brain tissue image values ​​reaches 95%, and a homogeneous or heterogeneous material template corresponding to the brain tissue can be used as the target template.

[0074] As shown in Figure 5, the horizontal axis represents the image values. The bold black line segment on the horizontal axis represents the reference range of image values ​​for the mucosal tissue. The dotted area represents the actual image value distribution of the medical image in the target area (the vertical axis can represent the number of voxels corresponding to the image values, or the frequency or percentage of voxel occurrence, etc.). The projection of the bidirectional arrow in the dotted area onto the horizontal axis represents the actual image value distribution interval corresponding to 95% confidence. As can be seen from Figure 5, a portion of the actual image value distribution interval corresponding to 95% confidence is outside the reference range of image values ​​for the mucosal tissue. That is, the confidence level of the target area medical image matching the reference range of image values ​​for the mucosal tissue does not reach 95%. In this embodiment, it is considered that the target area medical image does not match the mucosal tissue, and the homogeneous material template or heterogeneous material template corresponding to the mucosal tissue will not be used as the target template.

[0075] In some embodiments, a type of tissue or organ corresponds to a homogeneous material template and a heterogeneous material template. As shown in Figure 6, S20 can match a homogeneous material template or a heterogeneous material template as a target template based on the relationship between the image value distribution of the target region and the image value reference range of each type of tissue or organ.

[0076] In some embodiments, reference ranges for image values ​​of various tissue and organ types can be preset in the electronic device based on common knowledge in the art, or they can be determined manually. The relationship between the distribution of image values ​​in the target region and the reference ranges for image values ​​of various tissue and organ types can include a degree of matching or a confidence level of matching.

[0077] Matching degree refers to the ratio of the size of the overlapping interval to the size of the reference range of image values ​​for the tissue / organ type. The overlapping interval refers to the range of image values ​​in the target region that overlaps with the reference range of image values ​​for the tissue / organ type. A matching degree threshold can be preset. If the matching degree between the target region and a tissue / organ type reaches the preset threshold, a heterogeneous material template can be matched as the target template.

[0078] Match confidence refers to the confidence level that a target region matches a tissue or organ type. Match confidence can be obtained through any of the following methods:

[0079] 1. Randomly select image values ​​of N voxels within the target region, and use the ratio of the number of image values ​​that are included in the reference range of tissue / organ type D to N as the confidence score of the match between the target region and tissue / organ type D.

[0080] 2. Based on the normal distribution, the matching confidence level is determined as the confidence level corresponding to the target sub-interval within the image value distribution range of the target region. The target sub-interval is a sub-interval contained within the reference range of organ type D. For example, first calculate the average value μ and standard deviation σ of each voxel image value within the medical image of the target region. If the interval [μ-2σ, μ+2σ] is contained within the reference range of organ type C, the matching confidence level reaches 95%; if the interval [μ-3σ, μ+3σ] is contained within the reference range of organ type C, the matching confidence level reaches 98%. A second matching confidence threshold can be preset. If the matching confidence level between the target region and organ type D reaches the second matching confidence threshold, a heterogeneous material template can be matched as the target template. The second matching confidence threshold can be equal to or different from the first matching confidence threshold. In some embodiments, the second matching confidence threshold can be set to be greater than or equal to the first matching confidence threshold.

[0081] In some embodiments, a tissue or organ type may have a homogeneous material template and a heterogeneous material template, as shown in FIG7. S20 may include the following S21 and S22.

[0082] S21: In response to the image value distribution of the target region and the target reference range satisfying the second preset condition, a heterogeneous material template is matched as the target template, wherein the target reference range is the image value reference range of the tissue organ type corresponding to the target region or the image value reference range of the second reference tissue organ type, and the second reference tissue organ type is determined based on the image value distribution of the target region and the image value reference range of each tissue organ type.

[0083] In some embodiments, S21 may be: first determine or obtain the type of tissue or organ corresponding to the target region, and then, if the image value distribution of the target region and the image value reference range of the type of tissue or organ corresponding to the target region meet the second preset condition, then match the heterogeneous material template as the target template.

[0084] In some embodiments, S21 may be as follows: without considering the organ type corresponding to the target region, the image value distribution of the target region is compared with the image value reference range of each organ type, and the organ type that best matches is the second reference organ type. If the image value distribution of the target region and the image value reference range of the second reference organ type meet a second preset condition, then a heterogeneous material template is matched as the target template. Which specific reference organ type's heterogeneous material template is matched can be determined later. The method for determining the reference organ type can refer to the content of the other embodiments described above, or can be determined using existing technology.

[0085] The second preset condition mentioned above may be that the matching degree reaches a preset matching degree threshold, and / or that the matching confidence degree reaches a preset matching confidence degree.

[0086] In some embodiments, S21 includes: matching a heterogeneous material template as a target template in response to the target sub-interval of the image value distribution range of the target region being located within the target reference range. The target sub-interval is a value range containing the average image value of the target region.

[0087] Once the medical image of the target region is determined, the range of image values ​​for that region is also determined. Typically, the reference ranges for image values ​​of various tissue and organ types stored in electronic devices have clearly defined upper and lower limits. However, due to differences in medical imaging conditions and individual patient variations, the maximum and minimum values ​​of the image values ​​in the target region may exceed the reference range for the corresponding tissue and organ type. Consequently, the distribution range of image values ​​in the target region may partially exceed the reference range for the corresponding tissue and organ type. In this case, to determine the target template based on comparing image value ranges, a sub-range containing the average image value of the target region can be extracted from the distribution range of image values ​​in the target region. This sub-range is then compared with the reference ranges for image values ​​corresponding to each tissue and organ type. If the target sub-range falls within the reference range of the pre-determined target tissue / organ type or a second reference tissue / organ type, a heterogeneous material template is matched as the target template.

[0088] S22: In response to the fact that the image value distribution of the target area does not meet the second preset condition with respect to the target reference range, a homogeneous material template is matched as the target template.

[0089] The second preset condition can be the same as or different from the first preset condition.

[0090] When the second preset condition differs from the first preset condition, the types of preset conditions may be different. For example, the first preset condition is the range of the mean and / or mode of the image values ​​of each voxel in the target area, while the second preset condition is determined based on the normal distribution law.

[0091] Even when the second preset condition differs from the first preset condition, the preset conditions can be of the same type but with different value ranges. For example, the first preset condition might be that the confidence level of the image value distribution of the target region's medical image matching the candidate template reaches a first matching confidence threshold (e.g., 95%), while the second preset condition might be that the confidence level of the image value distribution of the target region's medical image matching the candidate template reaches a second matching confidence threshold (e.g., 98%). In some cases, the second preset condition can be set to a more stringent value range than the first preset condition; the example above illustrates this.

[0092] In some embodiments, the second preset condition is determined based on the resolution of the medical image. Specifically, if the resolution of the medical image reaches a preset resolution threshold, the value in the second preset condition can be determined to be a first value; if the resolution of the medical image does not reach the preset resolution threshold, the value in the second preset condition can be determined to be a second value. The second preset condition determined by the first value is more stringent and harder to achieve than the second preset condition determined by the second value. Similarly, the resolution of the medical image can be divided into three or more value intervals, and different values ​​for the second preset condition can be set for each interval.

[0093] For example, if the resolution of the medical image reaches a preset resolution threshold, the second preset condition is that the confidence level of the image value distribution of the medical image in the target area matching the candidate template reaches 98% (i.e., the first value is 98%); if the resolution of the medical image does not reach the preset resolution threshold, the second preset condition is that the confidence level of the image value distribution of the medical image in the target area matching the candidate template reaches 95% (i.e., the second value is 95%).

[0094] In other words, the higher the resolution of a patient's medical image and the more accurate the range of image values, the more stringent the conditions for using a heterogeneous material template become. At lower resolutions, medical images are limited by conditions and cannot present more details of tissues and organs, which can easily lead to higher dispersion in image values. The above embodiment determines the second preset condition based on the resolution of the medical image, taking into account the actual conditions of medical image acquisition. This allows for a more accurate determination of whether to use a heterogeneous material template to improve the accuracy of dose assessment in the target area, and also minimizes unnecessary use of heterogeneous material templates to reduce the computational load during dose assessment. Specifically, using a heterogeneous material template allows for a more detailed and accurate characterization of the material distribution in the target area based on the medical image, more closely reflecting the patient's actual physical condition, which is beneficial for establishing a high-precision 3D model. Using a homogeneous material template avoids the problem of image acquisition errors causing deviations in the image values ​​themselves, and then using overly detailed methods to define the material of voxel units based on these deviations, thus exacerbating the image acquisition error. Furthermore, using a homogeneous material template can reduce the computational load to some extent.

[0095] In some embodiments, as shown in FIG8, S20 includes the following S23 and S24.

[0096] S23: Obtain the tissue / organ type corresponding to the target region.

[0097] S24: In response to the fact that the type of organ corresponding to the target region is different from the type of the third reference organ, output a prompt message; the image value reference range of the third reference organ type and the image value distribution of the target region meet the third preset condition.

[0098] The third reference organ type is determined based on the distribution of image values ​​in the target region. The third reference organ type can be determined using the exact same method as either the first or second reference organ type, or using the same method but with different preset threshold values, or even using completely different methods. The method for determining the third reference organ type can be understood by referring to the methods for determining the first and second reference organ types, and will not be detailed here.

[0099] The type of tissue or organ corresponding to the target area in step S23 can be input by the operator into the electronic device; or the type of tissue or organ corresponding to the target area can be automatically determined based on a preset method, and can also be confirmed or modified by the operator after automatic determination. That is to say, S23 and S24 above use different methods to determine the type of tissue or organ, and if the types of tissue or organ determined by the two methods are inconsistent, a prompt message is output.

[0100] Normally, the tissue and organ type obtained in step S23 should be consistent with the tissue and organ type determined by the distribution of image values ​​in the target area. If they are inconsistent, it may be due to an abnormal distribution of image values ​​in the target area. In this case, the operator needs to intervene in time and use human experience to confirm which tissue and organ type to use to match the candidate template.

[0101] For example, the prompt information may include prompting the operator to input a specified tissue or organ type through the human-computer interaction interface. The prompt information may include the tissue or organ type corresponding to the target region and the matching information of the third reference tissue or organ type with the medical image of the target region (the matching information may include at least one of the following: the medical image of the target region, the range of image values ​​of the medical image of the target region, the reference range of image values ​​of the tissue or organ type corresponding to the target region, the reference range of image values ​​of the third reference tissue or organ type, the matching confidence of the tissue or organ type corresponding to the target region, and the matching confidence of the third reference tissue or organ type).

[0102] After outputting the prompt information, the system can also accept the type of tissue or organ input by the operator through the human-computer interaction interface, and select a candidate template as the target template based on the input tissue or organ type, for example, using a homogeneous material template corresponding to the input tissue or organ type as the target template.

[0103] For example, the prompt information may include prompting the operator to input a specified candidate template through the human-computer interaction interface. In response to the input specified candidate template, the electronic device presents the matching information between the specified candidate template and the medical image of the target area (the matching information may include, for example, the reference range of image values ​​of the tissue and organ type corresponding to the medical image of the target area and the reference range of image values ​​of the tissue and organ type corresponding to the specified candidate template, and / or the matching confidence of the medical image of the target area and the specified candidate template, etc.), and accepts the candidate template selected by the operator through the human-computer interaction interface as the target template.

[0104] For example, the prompt information may include displaying the overlapping intervals between the image value reference ranges of each tissue / organ type and the image value ranges of the target region's medical image, and / or the degree of overlap (the degree of overlap can be expressed as the ratio of the size of the overlapping interval to the size of the image value range of the target region's medical image, or the degree of overlap can be expressed as the ratio of the size of the overlapping interval to the size of the image value reference range of the tissue / organ type) that reach a predetermined degree of overlap threshold, and / or their corresponding candidate templates. The operator is then prompted to select from these tissue / organ types or candidate templates through a human-computer interaction interface to determine the target template. Alternatively, the prompt information may also include displaying the most radiation-sensitive tissue / organ type among the tissue / organ types corresponding to each associated candidate template, and prompting the operator to select one from the candidate templates of the most sensitive tissue / organ type as the target template. Here, the associated candidate template can be a candidate template whose degree of overlap with the image value range of the target region's medical image reaches a predetermined degree of overlap threshold.

[0105] In some embodiments, the material template library includes candidate templates corresponding to various tissue and organ types. As shown in Figure 9, S20 includes the following S25, S26, and S27.

[0106] S25: Obtain at least two tissue / organ types associated with the image value distribution of the target region.

[0107] The at least two tissue and organ types associated with the image value distribution of the medical image in the target area can refer to a predetermined number of tissue and organ types that have the highest degree of conformity with the image value distribution of the medical image in the target area (e.g., the highest confidence level of the match), or several tissue and organ types whose matching confidence level reaches a preset confidence threshold (the preset matching threshold can be the same as the confidence level value in the first preset condition and the second preset condition mentioned above, or it can be a lower value, such as 50%, 20%, etc.); or it can be the associated tissue and organ type and some tissue and organ types that are adjacent to the associated tissue and organ type, wherein the associated tissue and organ type is the tissue and organ type with the highest confidence level of matching with the image value distribution of the medical image in the target area.

[0108] "The at least two tissue / organ types associated with the image value distribution of the medical image of the target area" can be two tissue / organ types, three tissue / organ types, or all tissue / organ types involved in the material template library.

[0109] In some embodiments, steps S25 to S27 described above may be performed after obtaining the tissue / organ type corresponding to the target region, in response to a discrepancy between the tissue / organ type corresponding to the target region and the fourth reference tissue / organ type. The fourth reference tissue / organ type can be determined using the exact same method as any of the first, second, and third reference tissue / organ types, or using the same method but with different preset threshold values, or using entirely different methods. The method for determining the fourth reference tissue / organ type can be understood by referring to the methods for determining the first, second, and third reference tissue / organ types, and will not be detailed here.

[0110] In some embodiments, the number of associated tissue and organ types corresponding to the target area can be received by the operator through a human-computer interaction interface, and then S25 to S27 can be executed in response to the input number of associated tissue and organ types.

[0111] S26: Determine the most irradiated tissue or organ type among the at least two tissue or organ types.

[0112] S27: Match the homogeneous material template of the tissue or organ type most sensitive to radiation as the target template.

[0113] In some cases, due to limitations in medical imaging equipment, certain areas of tissue may not be clearly displayed in the images. This results in one sub-region of the target area matching the image values ​​of the first tissue / organ type E, while another sub-region matches the image values ​​of the second tissue / organ type F. Considering the tolerance of tissues / organs and the conservatism of dose assessment, to improve irradiation safety and avoid vascular wall rupture caused by irradiation in cases of misjudgment, a candidate template corresponding to the more irradiation-sensitive tissue / organ type between tissue / organ type E and tissue / organ type F can be used as the target template to establish a three-dimensional model of the target area.

[0114] Background dose is an important indicator for calculating the total dose of BNCT. Background dose mainly comes from the dose generated by the interaction of the radiation source with hydrogen and nitrogen elements. The higher the proportion of these two elements, the more sensitive the tissue or organ. Those skilled in the art can use this principle to determine the most sensitive tissue or organ among multiple tissue / organ types. In some embodiments, tissue / organ types that are more sensitive to radiation typically have lower densities. That is, when the image value of a sub-region of the medical imaging target area matches the first type of tissue / organ, and the image value of another sub-region matches the second type of tissue / organ, in steps S25 to S27, when performing three-dimensional modeling of the sub-region with the higher image value in the target area, this sub-region is converted into a low-density material with similar properties to the tissue for dose assessment.

[0115] For example, when the image values ​​of one sub-region of a target area match bone tissue and the image values ​​of another sub-region match mucosal tissue, the sub-region matching bone tissue can be represented using the material of mucosal tissue. Hydrogen and nitrogen are more abundant in mucosal tissue than in bone tissue; therefore, representing the target area using the material of mucosal tissue is safer when developing a treatment plan, considering tolerability.

[0116] In some other embodiments, as shown in FIG10, the method for establishing a three-dimensional model includes the following steps S1010 to S1030.

[0117] S1010: Obtain the target region based on medical images.

[0118] S1020: Define the material of the voxel unit in the target region based on the target template in the material template library that matches the target region. The candidate templates that can be matched in the material template library include homogeneous material templates.

[0119] S1030: Establish a three-dimensional model based on the material of voxel units in each target region.

[0120] In S1010 to S1030 above, the target template for matching in S1020 can be obtained by using S25, S26 and S27 above. For other descriptions, please refer to other parts of this specification.

[0121] In some embodiments, a type of tissue or organ corresponds to a homogeneous material template and a heterogeneous material template, as shown in Figure 11. S20 can match a homogeneous material template or a heterogeneous material template as a target template based on the OAR information of the target region.

[0122] The OAR (Organ at risk) information for a target area can include whether the target area is an OAR. Alternatively, the OAR information for a target area can include whether the target area is a primary OAR. The OAR of a target area can also be a quantified value of the severity of the target area.

[0123] The OAR information of the target area can be determined based on factors such as the type of radiotherapy, the affected area of ​​the patient's body, and different treatment sites. This OAR information can be determined manually or by a computer program. Different types of radiotherapy, different areas of the patient's body covered by the irradiation range, and different treatment sites correspond to different major organs at risk. Methods for obtaining the OAR information of the target area are existing technologies and will not be detailed in this specification.

[0124] The IAEA (International Atomic Energy Agency) Integrated Testing Center's guidelines for reporting dose and volume in BNCT state that for each OAR (Organizational Area of ​​Reduction), the maximum dose should be reported when the radiation received by part or the entire organ of that OAR exceeds the acceptable tolerable level. The volume received exceeding the acceptable tolerable dose should be assessed from a dose-volume histogram. Therefore, the accuracy of OAR dose is crucial for BNCT. Consequently, when the target area is an OAR, a more accurate and detailed characterization of the target area's material is needed to obtain a more accurate three-dimensional model, thereby leading to more accurate subsequent dose calculations.

[0125] In some embodiments, the OAR information of the target region includes whether the target region is an OAR. Accordingly, S20 may include: in response to the OAR information of the target region being an OAR, matching a heterogeneous material template as the target template.

[0126] In some embodiments, the OAR information of the target area includes whether the target area is a primary OAR. Accordingly, S20 may include: in response to the OAR information of the target area being a primary OAR, matching a heterogeneous material template as a target template. Here, the primary OAR refers to the most important OAR among OARs, such as OARs adjacent to the tumor or adjacent to the radiation field. It is understood that the primary OAR is determined by the type and location of radiotherapy; different types of radiotherapy and different treatment sites correspond to different primary OARs. The primary OAR corresponding to different tumor sites can be determined by referring to publicly available literature in the art or by experience. The primary OAR can be determined manually or by a computer program according to preset rules, such as based on the distance from the tumor or radiation field. The primary OARs corresponding to each tumor site determined manually or by a computer program can be stored. Subsequently, based on the stored content, the corresponding primary OAR for the current tumor site can be obtained, thus determining whether the target template is a primary OAR.

[0127] The two embodiments described above determine the use of heterogeneous material templates based on whether the target area is the OAR or the primary OAR. This allows the material distribution of the three-dimensional model established for the patient's OAR or primary OAR to be more detailed and closer to the patient's actual condition, thereby improving the accuracy of dose assessment for the patient's OAR.

[0128] If the target area is not an OAR or not the primary OAR, the target template type can be either a homogeneous material template or a heterogeneous material template. In some embodiments, other conditions can be considered to determine whether a heterogeneous material template is necessary for the target area. For example, these other conditions may be determined based on the image value distribution of the target area. If, based on other conditions, it is determined that a heterogeneous material template is unnecessary for the target area, a homogeneous material template can be matched as the target template.

[0129] In some embodiments, when the target area is not an OAR or not the primary OAR, the radiation dose received by the target area is low, thus having little impact on the dose distribution of radiotherapy. There is no need to describe the material heterogeneity of the target area in too much detail. In this case, using a homogeneous material template to build a three-dimensional model of the target area can reduce the amount of computation during three-dimensional modeling and dose assessment. It also avoids the problem that the image values ​​themselves are biased due to image acquisition errors, and that the material of the voxel unit is defined in too much detail based on the biased image values, which would aggravate the image acquisition error and lead to inaccurate dose calculations in the future.

[0130] S30: Establish a three-dimensional model based on the material of voxel units in each target region.

[0131] During neutron capture therapy, the background dose is closely related to the hydrogen and nitrogen elements in human tissues. When establishing a three-dimensional model, the material of each voxel unit can be determined, and the hydrogen and nitrogen content in each voxel unit can be determined. This makes it easier to accurately determine the background dose of normal tissues and organs based on the hydrogen and nitrogen content in each voxel unit when determining the treatment plan, and to ensure the safety of normal tissues when determining the total radiation dose.

[0132] The method for establishing the 3D model provided in this specification, after determining the target region in the medical image, matches the target template from a material template library containing homogeneous and heterogeneous material templates, and defines the material of each voxel unit based on the target template, thereby establishing the 3D model. This approach considers the different image qualities of medical images and the differences in various tissues and organs of individual patients. By matching the target template, different strategies are used to more specifically define the material of each voxel unit in each target region, thus establishing a 3D model with more suitable materials. This model more closely resembles the actual tissue and organ characteristics of the patient, enabling treatment planning to more accurately assess the dose limitations of normal tissues and the prescribed dose for tumors, ensuring the safety of the treatment plan and improving the effectiveness of radiotherapy. Furthermore, it does not increase unnecessary computational load, which is beneficial for widespread application.

[0133] In some embodiments, after determining the target template based on any of the above methods, before building the 3D model using the target template, the determined target template can be presented through a human-computer interaction interface for operator confirmation. In other embodiments, after determining the target template based on any of the above methods, the 3D model can also be built, thereby improving the automation level of 3D model building and reducing the operator's workload.

[0134] In some embodiments, heterogeneous material templates in the material template library can be constructed using at least one of the following methods: construction based on the relationship between the density of tissues / organs and image values; construction based on the relationship between the elemental composition of tissues / organs and image values; construction based on key parameters of image values ​​and corresponding densities; and construction based on key parameters of image values ​​and corresponding elemental compositions. Those skilled in the art can also construct heterogeneous template libraries based on the relationship or key parameters of known image values ​​with other characterizing materials, such as at least one of the aforementioned atomic numbers, one or more mass numbers, and average ionization energies.

[0135] For example, existing technologies disclose the following relationships: the HU value (x) and mass density (y) of air and lungs are expressed as: y = 0.001x + 1.0050; the HU value (x) and mass density (y) of soft tissue are expressed as: y = 0.0010x + 1.0042; and the HU value (x) and mass density (y) of bone tissue are expressed as: y = 0.006x + 1.0152. These relationships can be used to construct heterogeneous material templates for various tissue and organ types.

[0136] In some embodiments, a heterogeneous material template of a target tissue or organ type in the material template library can be constructed by the following steps S41 to S43.

[0137] S41: Based on a preset step size, the reference range of image values ​​corresponding to the target tissue / organ type is divided into multiple sub-intervals of image values.

[0138] S42: Based on a predetermined formula relating tissue / organ density to image values, determine the density value corresponding to each sub-interval of image values; and / or, based on a predetermined key pair of tissue / organ density to image values, use interpolation to determine the density value corresponding to each sub-interval of image values.

[0139] The pre-selected formula for the relationship between tissue / organ density and image values ​​can be any of the formulas for the relationship between HU value and mass density mentioned above. The formulas differ for different tissue / organ types. The HU value and mass density formulas provided above are merely examples of three tissue / organ types already disclosed in the prior art; in reality, there are many other formulas corresponding to various tissue / organ types. Furthermore, with technological advancements, new formulas may emerge for the same tissue / organ type in the future.

[0140] The key pair of tissue / organ density and image value refers to the value of a tissue / organ density (or elemental composition) and its corresponding image value, which are determined in advance through experimental research. For example, existing technology has determined multiple image values ​​for ocular medical images of adults aged 40-50 and the corresponding tissue / organ density values ​​for each image value through experimental research. This experimental research data can be used to construct a heterogeneous material template library.

[0141] The density value (or element composition) corresponding to the image value sub-interval between two key value pairs can be determined by interpolation between two close key values.

[0142] A heterogeneous material template library can be constructed using either of these two methods: a predetermined formula relating tissue / organ density to image values, and a predetermined key value pair relating tissue / organ density to image values. In some embodiments, these two methods can be combined to construct the heterogeneous material template library; for example, unknown values ​​can be determined using formulas based on known key values. Table 1 below shows an example of a heterogeneous material template for an eye constructed by combining the two methods described above.

[0143] Table 1 shows an example of a heterogeneous material template for an eye constructed using both methods.

[0144] Among them, HU L This represents the minimum HU value actually collected in the target area where the template was applied. U This indicates the maximum HU value actually acquired in the target area where the template was applied. The HU value is a density unit in CT imaging, used to represent the density differences between different tissues or substances. Values ​​with an "a" in the upper right corner indicate that they are derived from eye data on tissue physical properties from a comprehensive reference book, while values ​​with a "b" in the upper right corner indicate that they are derived from mixed data (adults aged 40-50 years) from ICRU-46.

[0145] The density values ​​and elemental composition of the HU values ​​in Table 1 above, from the sub-interval [25, 30] to the sub-interval [75, 80], were determined by interpolation.

[0146] Similarly, Table 2 below is an example of a heterogeneous material template for bone tissue constructed by combining the two methods described above.

[0147] Table 2 shows an example of a heterogeneous material template for bone tissue constructed using both methods.

[0148] Among them, HU L This represents the minimum HU value actually collected in the target area where the template was applied. U This indicates the maximum HU value actually collected in the target area where the template was applied. Values ​​with "a" in the upper right corner of the density value represent known skeletal muscle-red bone marrow data (40% fat, 40% water, and 20% protein). Values ​​with "b" in the upper right corner of the density value represent known bone sponge data (33% cortical bone + 67% bone marrow). Values ​​with "c" in the upper right corner of the density value represent known bone-vertebral (C4) data. Values ​​with "d" in the upper right corner of the density value represent known skeleton-skull data. Values ​​with "e" in the upper right corner of the density value represent known skeleton-mandible data. Values ​​with "f" in the upper right corner of the density value represent bone-cortical bone data derived from ICRU-46.

[0149] It is important to note that in Tables 1 and 2 above, the starting value in the first row is written as min (minimum value), and the ending value in the last row is written as max (maximum value). This is because even if most voxels in the medical image of the target area conform to the image value reference range of that tissue or organ, there will still be a few voxels that do not fall within this reference range. If only the image value reference range of the reference tissue or organ is used, some voxels in the target area of ​​the medical image will not have corresponding density values ​​and / or elemental compositions, making it impossible to determine the material. Therefore, setting min (minimum value) and max (maximum value) allows these special voxels to also be set to reasonable materials and densities within the image value reference range, so as not to affect the calculation. Even if the values ​​in the tables are not used, this issue should still be considered when constructing a heterogeneous material library according to the relationships. That is, the maximum and minimum image values ​​of the heterogeneous template library constructed based on the aforementioned various relationships or parameters are the maximum and minimum image values ​​actually acquired in the target area using the template.

[0150] S43: Construct a heterogeneous material template corresponding to the target tissue / organ type based on the density value corresponding to each image value sub-interval.

[0151] For example, a heterogeneous material template can be constructed based on the density values ​​and elemental composition corresponding to each image value sub-interval in Table 1 above. Then, based on the heterogeneous material template, the constituent elements and / or density of each voxel unit in the target region can be determined, that is, the material of the target region can be determined.

[0152] The heterogeneous material template can be in the form of tables as shown in Tables 1 and 2 above, or it can be in the form of a matrix or expression. For example, the expression form can be obtained by fitting the data in Tables 1 and 2 to obtain the expression or curve in the heterogeneous material template. It should be understood that S41-S43 use density as an example for illustration; the method for determining the elemental composition can be analogous to the method for determining density.

[0153] This specification provides a three-dimensional model creation apparatus, which can be used to implement any of the three-dimensional model creation methods described above. As shown in Figure 12, the apparatus includes a first acquisition unit 10, a definition unit 20, and a creation unit 30.

[0154] The first acquisition unit 10 is used to acquire the target region based on medical images.

[0155] The definition unit 20 is used to define the material of the voxel unit in the target region based on the target template in the material template library that matches the target region. The candidate templates that can be matched in the material template library include homogeneous material templates and heterogeneous material templates.

[0156] The establishment unit 30 is used to establish a three-dimensional model based on the material of the voxel units in each target region.

[0157] In some embodiments, the defining unit 20 is further configured to match candidate templates in a material template library as target templates based on the tissue / organ type and / or image value distribution corresponding to the target region.

[0158] In some embodiments, the defining unit 20 is further configured to match a homogeneous material template or a heterogeneous material template as a target template based on the relationship between the image value distribution of the target region and the image value reference range of each tissue and organ type.

[0159] In some embodiments, the defining unit 20 is further configured to match a homogeneous material template or a heterogeneous material template as a target template based on the OAR information of the target region.

[0160] In some embodiments, the apparatus further includes a second acquisition unit and a determination unit.

[0161] The second acquisition unit is used to acquire at least two tissues and organs associated with the image value distribution of the target region.

[0162] The determining unit is used to determine the most irradiated tissue or organ type among the at least two tissue or organ types.

[0163] The defining unit is also used to match the homogeneous material template of the tissue or organ type most sensitive to irradiation as the target template.

[0164] In some embodiments, the apparatus further includes a construction unit for constructing heterogeneous material templates in the material template library by at least one of the following methods: construction based on the relationship between the density of tissues and organs and image values; construction based on the relationship between the elemental composition of tissues and organs and image values; construction based on key parameters of image values ​​and corresponding densities; and construction based on key parameters of image values ​​and corresponding elemental compositions.

[0165] In some embodiments, the treatment planning module further includes a reference unit that can store and retrieve reference ranges of image values ​​corresponding to various tissue and organ types.

[0166] In some embodiments, the first acquisition unit 10, the definition unit 20, the establishment unit 30, the second acquisition unit, the determination unit, the construction unit, and the reference unit can each be an independent execution unit, or they can share an execution unit with one or more modules. The execution unit may include one or more processors, or it may include one or more processors and one or more memories, wherein the memory stores instructions that can be executed by the processor.

[0167] This specification provides a radiotherapy system, as shown in Figure 13, which includes a beam irradiation device, a treatment planning module, and a control module.

[0168] The beam irradiation device is used to generate a radiation beam. The treatment planning module is used to generate a treatment plan. The control module is used to control the beam irradiation device to generate a radiation beam according to the treatment plan.

[0169] The treatment plan module includes a three-dimensional model creation device, which includes a first acquisition unit 10, a definition unit 20, and a creation unit 30.

[0170] The first acquisition unit 10 is used to acquire the target region based on medical images.

[0171] The definition unit 20 is used to define the material of the voxel unit in the target region based on the target template in the material template library that matches the target region. The candidate templates that can be matched in the material template library include homogeneous material templates and heterogeneous material templates.

[0172] The establishment unit 30 is used to establish a three-dimensional model based on the material of the voxel units in each target region.

[0173] In addition, the treatment plan module may also include a second acquisition unit, a determination unit, and a construction unit as described in the aforementioned three-dimensional model creation device. The descriptions and functions of the above devices can be understood by referring to the content of the three-dimensional model creation method section, and will not be repeated here.

[0174] In some embodiments, the radiotherapy system shown in FIG14 includes a reference unit in the treatment planning module, which can store and retrieve the reference range of image values ​​corresponding to each type of tissue and organ.

[0175] In some embodiments, the first acquisition unit 10, the definition unit 20, the establishment unit 30, the second acquisition unit, the determination unit, the construction unit, and the reference unit may each be an independent execution unit, or they may share an execution unit with one or more modules. The execution unit may include one or more processors, or it may include one or more processors and one or more memories, wherein the memory stores instructions that can be executed by the processor.

[0176] In some embodiments, a radiotherapy system is a BNCT system, including a TPS (treatment planning system); in some embodiments, the radiotherapy system further includes a robotic arm control system; in some embodiments, the radiotherapy system further includes a data management system for interacting with and processing data from other systems and modules. Accordingly, one or more of the aforementioned treatment planning module, control module, first acquisition unit 10, definition unit 20, establishment unit 30, second acquisition unit, determination unit, and construction unit can be located in the TPS, the robotic arm control system, the data management system, or a separately located data processing device. The specific configuration can be set according to actual needs and circumstances, and is not limited here.

[0177] This invention also provides an electronic device, as shown in FIG15. The electronic device may include a processor 1501 and a memory 1502, wherein the processor 1501 and the memory 1502 may be connected by a bus or other means. FIG15 shows an example of connection by a bus.

[0178] Processor 1501 may be a central processing unit (CPU). Processor 1501 may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations thereof.

[0179] The memory 1502, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the three-dimensional model creation method in this embodiment of the invention (e.g., the first acquisition unit 10, definition unit 20, and creation unit 30 shown in FIG. 12). The processor 1501 executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions, and modules stored in the memory 1502, thereby realizing the three-dimensional model creation method in the above method embodiment.

[0180] Memory 1502 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by processor 1501, etc. Furthermore, memory 1502 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory 1502 may optionally include memory remotely located relative to processor 1501, and these remote memories may be connected to processor 1501 via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0181] The one or more modules are stored in the memory 1502. When executed by the processor 1501, they execute the method for establishing a three-dimensional model as shown in the embodiment of FIG1.

[0182] The specific details of the above-mentioned electronic device can be understood by referring to the relevant descriptions and effects in the method embodiments, and will not be repeated here.

[0183] This specification also provides a computer storage medium storing computer program instructions, which, when executed, implement the steps of the above-described method for establishing a three-dimensional model.

[0184] This specification also provides a computer program product comprising a computer program that, when executed by a processor, implements the steps of the above-described method for establishing a three-dimensional model.

[0185] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0186] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, please refer to each other. The focus of each embodiment is to describe the differences from other embodiments.

[0187] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions.

[0188] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0189] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute certain parts of the methods of various embodiments of this application.

[0190] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc.

[0191] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0192] Although this application has been described through embodiments, those skilled in the art will know that this application has many modifications and variations without departing from the spirit of this application, and it is intended that the appended claims cover such modifications and variations without departing from the spirit of this application.

Claims

1. A method for establishing a three-dimensional model, characterized in that, include: Target region is obtained based on medical images; The material of the voxel unit in the target region is defined based on the target template in the material template library that matches the target region. The candidate templates that can be matched in the material template library include homogeneous material templates and heterogeneous material templates. A three-dimensional model is built based on the material of the voxel units in each target region.

2. The method according to claim 1, characterized in that, Based on the target template in the material template library that matches the target region, define the material of the voxel unit within the target region, including: The candidate templates in the material template library are used as the target templates based on the tissue and organ types and / or image value distribution corresponding to the target region.

3. The method according to claim 2, characterized in that, The target template is selected from candidate templates in a material template library based on the image value distribution corresponding to the target region, including: The candidate template corresponding to the first reference tissue / organ type in the matching material template library is used as the target template; wherein, the reference range of the image value of the first reference tissue / organ type and the distribution of the image value of the target region satisfy a first preset condition, and the candidate template corresponding to the first reference tissue / organ type is a homogeneous material template or a heterogeneous material template.

4. The method according to claim 1, characterized in that, Based on the target template in the material template library that matches the target region, define the material of the voxel unit within the target region, including: Based on the relationship between the image value distribution of the target region and the image value reference range of each tissue and organ type, a homogeneous material template or a heterogeneous material template is matched as the target template, or... Based on the OAR information of the target region, a homogeneous material template or a heterogeneous material template is matched as the target template.

5. The method according to claim 4, characterized in that, Based on the target template in the material template library that matches the target region, define the material of the voxel unit within the target region, including: In response to the image value distribution of the target region and the target reference range satisfying the second preset condition, a heterogeneous material template is matched as the target template, wherein the target reference range is the image value reference range of the tissue organ type corresponding to the target region or the image value reference range of the second reference tissue organ type, and the second reference tissue organ type is determined based on the image value distribution of the target region and the image value reference range of each tissue organ type; In response to the fact that the image value distribution of the target area does not meet the second preset condition with respect to the target reference range, a homogeneous material template is matched as the target template.

6. The method according to claim 4, characterized in that, The OAR information of the target area includes whether the target area is an OAR, or whether the target area is a primary OAR; The step of matching a homogeneous material template or a heterogeneous material template as the target template based on the OAR information of the target region includes: In response to the OAR information of the target area, a heterogeneous material template is matched as the target template, or... In response to the OAR information of the target area as the primary OAR, a heterogeneous material template is matched as the target template.

7. The method according to claim 1, characterized in that, Based on the target template in the material template library that matches the target region, define the material of the voxel unit within the target region, including: Obtain at least two tissue / organ types associated with the image value distribution of the target region; Identify the most radiation-sensitive tissue / organ type among the at least two tissue / organ types; A homogeneous material template matching the type of tissue or organ most sensitive to radiation is used as the target template.

8. The method according to claim 1, characterized in that, The method further includes constructing heterogeneous material templates in the material template library by at least one of the following methods: Constructed based on the relationship between the density of tissues and organs and imaging values; Constructing a formula based on the relationship between the elemental composition of tissues and organs and image values; Constructed based on key parameters of image values ​​and corresponding densities; Constructed based on key parameters composed of image values ​​and corresponding elements.

9. A device for creating a three-dimensional model, characterized in that, include: The first acquisition unit is used to acquire the target region based on medical images; A definition unit is used to define the material of the voxel unit in the target region based on the target template in the material template library that matches the target region. The candidate templates that can be matched in the material template library include homogeneous material templates and heterogeneous material templates. Establish units, used to build three-dimensional models based on the materials of voxel units in each target region.

10. The apparatus according to claim 9, characterized in that, The defining unit is also used to match candidate templates in the material template library as target templates based on the tissue and organ types and / or image value distribution corresponding to the target region.

11. The apparatus according to claim 9, characterized in that, The defining unit is further configured to match a homogeneous material template or a heterogeneous material template as a target template based on the relationship between the image value distribution of the target region and the image value reference range of each tissue and organ type; or, the defining unit is further configured to match a homogeneous material template or a heterogeneous material template as a target template based on the OAR information of the target region.

12. The apparatus according to claim 9, characterized in that, The device further includes: The second acquisition unit is used to acquire at least two tissues and organs associated with the image value distribution of the target region; A determining unit is used to determine the most irradiated tissue or organ type among the at least two tissue or organ types; The defining unit is also used to match the homogeneous material template of the tissue or organ type most sensitive to irradiation as the target template.

13. The apparatus according to claim 9, characterized in that, It also includes building units for constructing heterogeneous material templates in the material template library by at least one of the following methods: Constructed based on the relationship between the density of tissues and organs and imaging values; Constructing a formula based on the relationship between the elemental composition of tissues and organs and image values; Constructed based on key parameters of image values ​​and corresponding densities; Constructed based on key parameters composed of image values ​​and corresponding elements.

14. A radiotherapy system, characterized in that, include: A beam irradiation device used to generate a radiation beam; The treatment plan module is used to generate treatment plans; The control module is used to control the beam irradiation device to generate a radiation beam according to the treatment plan; The treatment planning module includes a three-dimensional model creation device, which comprises: The first acquisition unit is used to acquire the target region based on medical images; A definition unit is used to define the material of the voxel unit in the target region based on the target template in the material template library that matches the target region. The candidate templates that can be matched in the material template library include homogeneous material templates and heterogeneous material templates. Establish units, used to build three-dimensional models based on the materials of voxel units in each target region.

15. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 8.

Images