A fdfd nonlinear conformal filtering method for suppressing step error of specific absorption rate

Abstract

The application discloses a kind of FDTD nonlinear conformal filter methods of inhibiting step error of specific absorption rate, comprising: obtaining grid model;Determine the eighth cube in each grid as material domain;The 3x3x3 three-dimensional grid matrix of using preset is scanned to the grid, determine the step grid on the two sides of human tissue / air interface;Optimize the conductivity and dielectric constant of the material domain of step grid, after all step grid is completed optimization, the first electric field component of all grid is calculated;The first electric field component of each material domain vertex of same tissue is linearly interpolated \ extrapolated, obtain the second electric field component of the center of material domain, and calculate local SAR;After obtaining the local SAR of all material domains of same tissue, the tissue average SAR of this tissue is calculated, after obtaining the local SAR of all material domains of different tissues, the 10-g SAR containing different tissues is calculated.The application can effectively inhibit the step error of the local SAR of the complex structure surface of human body, improve the accuracy of SAR statistical result.

Description

Technical Field

[0001] This invention belongs to the field of electromagnetic field technology, and specifically relates to an FDTD nonlinear conformal filtering method for suppressing the step error of the specific absorption rate. Background Technology

[0002] The FDTD (Finite-Difference Time-Domain) algorithm is widely used to calculate electromagnetic radiation doses, such as SAR (Specific Absorption Rate), which is the electromagnetic power absorbed per unit mass of human tissue, based on anatomical 3D mesh models of the human body. Mesh models approximate curved interfaces between different materials using stepped meshes, which result in significant calculation errors for local SAR in these stepped meshes. Inside the human body, the differences in electromagnetic parameters between adjacent tissues are relatively small, and with sufficiently high mesh resolution, the stepped error in local SAR of the body mesh is not significant. However, on the body surface, the differences in electromagnetic parameters across the tissue / air interface are substantial, leading to significant stepped errors in local SAR of the surface mesh.

[0003] The current international standard for safe electromagnetic radiation doses (IEEE Standard for Safety Levels with Respect to Human Exposure to Electric, Magnetic, and Electromagnetic Fields, 0Hz to 300GHz, IEEE Standard C95.1) is relevant. TM According to the ICNIRP guidelines (2019 and ICNIRP, “ICNIRP guidelines for limiting exposure to electromagnetic fields (100kHz to 300GHz),” Health Phys., vol. 118, no. 5, pp. 483–524, 2020), the radiation dose of electromagnetic waves with frequencies from 100kHz to 10GHz is quantified using the average SAR (10g-SAR) in a 10g cube of human tissue. The proportion of the stepped grid on the body surface in a 10g cube is small, and the step error of local SAR has little impact on the statistical results of SAR. When studying the microscopic effects of electromagnetic radiation on the complex structures of the human body, SAR distribution data in tissues are required. For example, the damaging effect of electromagnetic radiation on corneal cells needs to be correlated with the effect of each grid on the corneal surface and the local SAR of that grid. This requires considering the step error of local SAR. When calculating the tissue-averaged SAR of the entire cornea based on the SAR distribution, since the stepped grid on the corneal surface accounts for a large proportion of the entire cornea, the statistical analysis must also consider the step error of local SAR.

[0004] Currently, the FDTD algorithm can suppress step errors in electric and magnetic fields through material conformal techniques. However, these techniques, based on the varying proportions of different materials in each mesh, are not suitable for anatomical mesh models where only one type of tissue is filled within the mesh. For anatomical mesh models, existing techniques eliminate local SAR with step errors in the body surface mesh during analysis. The IEC / IEEE international standard (“Determining the peak spatial-average specific absorption rate (SAR) in the human body from wireless communications devices, 30MHz to 6GHz - Part 1: General requirements for using the finite-difference time-domain (FDTD) method for SAR calculations,” in IEC / IEEE 62704-1:2017, pp.1–86, Oct. 2017) designs an algorithm that analyzes the body surface mesh grid by grid. If a body surface mesh is found to be within a step structure, it is considered an invalid mesh, and its local SAR is not included in the statistical analysis.

[0005] However, due to the skin effect of biological tissues, the local SAR values ​​generated by electromagnetic radiation on the body surface are relatively large. This technique of removing local SAR on the body surface not only loses the anatomical structure of the body surface and the local SAR, but also has a significant impact on the SAR statistical results such as tissue-average SAR and 10g-SAR when the grid resolution is limited. Summary of the Invention

[0006] To address the aforementioned problems in the existing technology, this invention provides an FDTD nonlinear conformal filtering method for suppressing the step error of the specific absorption rate. This invention solves the above problems through the following technical solution:

[0007] This invention provides an FDTD nonlinear conformal filtering method for suppressing the step error of specific absorption rate, comprising:

[0008] Obtain a mesh model that has been pre-divided uniformly at the human tissue / air interface;

[0009] Based on the location of the material's conductivity and dielectric constant on the grid in the FDTD algorithm, one-eighth of the cube in each grid is determined as the material domain;

[0010] Each grid is used as the center grid in sequence, and a preset 3×3×3 three-dimensional grid matrix is ​​used for scanning to determine the stepped grid on both sides of the human tissue / air interface.

[0011] Based on the volume proportion of different material domain types in the three-dimensional mesh matrix with the stepped mesh as the center, the conductivity and dielectric constant of the material domains of the stepped mesh are optimized. After all the stepped meshes are optimized, the first electric field component of all meshes is calculated using the FDTD algorithm.

[0012] Linearly interpolate or linearly extrapolate the first electric field component at the apex of each material domain of the same organization to obtain the second electric field component at the center of the material domain, and use the second electric field component to calculate the local SAR.

[0013] After obtaining the local SAR of all material domains of the same tissue, the tissue-average SAR of the tissue is statistically calculated. After obtaining the local SAR of all material domains of different tissues, the 10-g SAR of different tissues is statistically calculated.

[0014] In one embodiment of the present invention, the material domain type includes an organization domain and an air domain;

[0015] The step of sequentially using each grid as a central grid and scanning with a preset 3×3×3 three-dimensional grid matrix to determine the stepped grids on both sides of the human tissue / air interface includes:

[0016] Each grid is used as the center grid in sequence, and a pre-set 3×3×3 three-dimensional grid matrix is ​​used for scanning;

[0017] When the central grid contains an organization domain and the remaining 26 adjacent grids in the three-dimensional grid matrix contain at least one air domain, or when the central grid contains an air domain and the remaining 26 adjacent grids in the three-dimensional grid matrix contain at least one organization domain, determine whether the central grid is a stepped grid.

[0018] After scanning the mesh model of the human tissue / air interface, the stepped meshes on both sides of the human tissue / air interface are obtained.

[0019] In one embodiment of the present invention, the preset 3×3×3 three-dimensional mesh matrix includes a first vertex;

[0020] The step of determining whether the central grid is a stepped grid includes:

[0021] An xyz coordinate system is established with the first vertex of the three-dimensional mesh matrix as the origin. The three-dimensional mesh matrix is ​​divided into 9 mesh groups, each mesh group including 3 meshes. The orthographic projections of the 3 meshes in the x, y or z directions coincide.

[0022] In the x, y, or z directions, when the material domain types of three grids in at least two of the nine grid groups are not completely ordered, the central grid is a stepped grid.

[0023] In one embodiment of the present invention, the step of optimizing the conductivity and dielectric constant of the material domains of the stepped mesh based on the volume proportion of meshes of different material domain types contained in the three-dimensional mesh matrix centered on the stepped mesh, and calculating the first electric field component of all meshes using the FDTD algorithm after all stepped meshes have been optimized, includes:

[0024] Determine the volume percentage of meshes containing different material domain types in a three-dimensional mesh matrix centered on a stepped mesh;

[0025] The effective conductivity and effective dielectric constant are obtained by nonlinearly weighting the conductivity and dielectric constant of different material domain types using the volume ratio.

[0026] Replace the conductivity of the material domain of the stepped mesh with the effective conductivity, and replace the dielectric constant with the effective dielectric constant;

[0027] After the conductivity and dielectric constant of the material domains of all stepped meshes have been replaced, the first electric field component of all meshes is calculated using the FDTD algorithm.

[0028] In one embodiment of the present invention, the effective conductivity is determined according to the following formula:

[0029]

[0030] In the formula, N represents the total number of material domain types in the three-dimensional mesh matrix centered on the stepped mesh, n i σ represents the volume percentage of the mesh containing the i-th material domain type. i Let σ represent the conductivity of the i-th material domain type, k represent the preset weight, and σ represent the conductivity of the i-th material domain type. eff The effective conductivity is calculated.

[0031] In one embodiment of the present invention, the effective dielectric constant is determined according to the following formula:

[0032]

[0033] In the formula, ε i ε represents the dielectric constant of the i-th material domain type. eff The effective dielectric constant is calculated.

[0034] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0035] This invention provides an FDTD nonlinear conformal filtering method for suppressing step error in specific absorptivity. First, a three-dimensional mesh matrix is ​​used to scan the human tissue / air interface grid-by-grid to determine the step mesh. Then, the effective conductivity and effective dielectric constant of the material domains surrounding the step mesh are optimized to obtain the effective conductivity and effective dielectric constant of the material domains within the step mesh. Next, the first electric field component of all grids is obtained using the FDTD algorithm. Using the first electric field component at the vertices of each material domain within the same tissue, the second electric field component at the center of each material domain is obtained through linear interpolation or linear extrapolation, and the local SAR at the center of the material domain is calculated. Then, the tissue-averaged SAR of the tissue is calculated. After obtaining the local SAR at the centers of all material domains in different tissues, the 10-g SAR is calculated. This invention can effectively suppress the step error of local SAR on the human body surface and is suitable for calculating the SAR distribution and SAR statistics of complex human body surface structures, providing accurate local SAR, tissue-averaged SAR, and 10-g SAR.

[0036] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0037] Figure 1 This is a flowchart of an FDTD nonlinear conformal filtering method for suppressing the step error of the specific absorption rate provided in an embodiment of the present invention;

[0038] Figure 2 This is a schematic diagram of a mesh model and material domain provided in an embodiment of the present invention;

[0039] Figure 3 This is a schematic diagram of a method for determining whether a central grid is a stepped grid, provided by an embodiment of the present invention.

[0040] Figure 4 This is a schematic diagram of calculating local SAR provided in an embodiment of the present invention;

[0041] Figure 5 This is a schematic diagram of a uniform sphere model of the cornea or skin with a resolution of 0.25 mm provided in an embodiment of the present invention;

[0042] Figure 6 This is a schematic diagram illustrating the variation of the average SAR step error of the corneal and skin sphere models provided in this embodiment of the invention with a preset weight k;

[0043] Figure 7 This is a schematic diagram of the SAR step error distribution of the corneal and skin sphere model cross-section provided in the embodiment of the present invention;

[0044] Figure 8 This is a schematic diagram of a two-dimensional anatomical model of the eyeball with resolutions of 0.025mm and 0.25mm provided in an embodiment of the present invention;

[0045] Figure 9 This is a schematic diagram of replacing the material domain of the body surface mesh with the local SAR of the material domain of the body mesh according to an embodiment of the present invention;

[0046] Figure 10 This is a schematic diagram of the precise SAR distribution in the eye grid models with resolutions of 0.025mm and 0.25mm provided in the embodiments of the present invention;

[0047] Figure 11a This is a schematic diagram of the SAR distribution with stepped error in the surface region of a 0.25mm resolution eye mesh model provided in an embodiment of the present invention;

[0048] Figure 11b This is a schematic diagram of the step error of local SAR in the surface region of an eye-shaped mesh model with a resolution of 0.25 mm provided in an embodiment of the present invention. Detailed Implementation

[0049] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0050] Figure 1 This is a flowchart of an FDTD nonlinear conformal filtering method for suppressing the step error of the specific absorption rate, provided in an embodiment of the present invention. Figure 1 As shown, this embodiment of the invention provides an FDTD nonlinear conformal filtering method for suppressing the step error of the specific absorption rate, comprising:

[0051] S1. Obtain a mesh model that has been pre-divided uniformly at the human tissue / air interface;

[0052] S2. Based on the position of the material's conductivity and dielectric constant on the grid in the FDTD algorithm, determine one-eighth of the cube in each grid as the material domain;

[0053] S3. Using each grid as the center grid in sequence, scan using the preset 3×3×3 three-dimensional grid matrix to determine the stepped grid on both sides of the human tissue / air interface.

[0054] S4. Based on the volume ratio of different material domain types in the three-dimensional mesh matrix with the stepped mesh as the center mesh, optimize the conductivity and dielectric constant of the material domain of the stepped mesh. After all the stepped meshes are optimized, use the FDTD algorithm to calculate the first electric field component of all meshes.

[0055] S5. Linearly interpolate or linearly extrapolate the first electric field component at the apex of each material domain of the same organization to obtain the second electric field component at the center of the material domain, and use the second electric field component to calculate the local SAR.

[0056] S6. After obtaining the local SAR of all material domains of the same tissue, the tissue-average SAR of the tissue is statistically calculated. After obtaining the local SAR of all material domains of different tissues, the 10-g SAR of different tissues is statistically calculated.

[0057] Specifically, a mesh model is obtained by uniformly dividing the human tissue / air interface in advance. In each mesh, the FDTD algorithm assigns the electrical conductivity and dielectric constant of the material to an eighth of a cube to determine it as a material domain. The material domain is filled with a material. The space between adjacent material domains does not participate in the FDTD algorithm and does not affect the calculation results.

[0058] Figure 2 This is a schematic diagram of a mesh model and material domain provided in an embodiment of the present invention. Figure 2 As shown, each material domain representing materials 1, 2, and 3 coincides with the vertex near the origin of its mesh. Each material domain has three vertices located in the middle of the three edges of the mesh, which are the assignment positions for the electrical conductivity σ1, σ2, σ3 and the dielectric constant ε1, ε2, ε3 of materials 1, 2, and 3.

[0059] Using each grid as the central grid, and combining it with 26 adjacent grids (6 faces, 12 edges, and 8 vertices), a 3×3×3 three-dimensional grid matrix is ​​formed to determine the various stepped grids on both sides of the human tissue / air interface. In this embodiment, the three-dimensional grid matrix includes grids of different material domain types. For example, the material domain types include air and human tissues such as the cornea and iris. In step S4, in the three-dimensional grid matrix with the stepped grid as the central grid, the effective conductivity and effective dielectric constant are optimized and calculated based on the volume ratio, conductivity, and dielectric constant of the grids of different material domain types, and assigned to the material domains of the stepped grids. Then, the first electric field component of all grids is calculated using the FDTD algorithm.

[0060] Further, in steps S4-S5, the first electric field component is located at the three vertices of the material domain; then, using the first electric field components at the vertices of each material domain of the same structure, linear interpolation or extrapolation is performed based on spatial distance to obtain the second electric field component, which is located at the center of the material domain, and is then used to calculate the local SAR at that location. In step S6, after obtaining the local SAR of all material domains of the same structure, they are directly statistically averaged to calculate the structure-averaged SAR of that structure. After obtaining the local SAR of all material domains of different structures, they are statistically averaged within each 10-g mass cube to calculate the 10-g SAR.

[0061] Step S3, which involves sequentially using each grid as the central grid and scanning it with a preset 3×3×3 three-dimensional grid matrix to determine the stepped grids on both sides of the human tissue / air interface, includes:

[0062] Each grid is used as the center grid in sequence, and a pre-set 3×3×3 three-dimensional grid matrix is ​​used for scanning;

[0063] When the central grid contains an organization domain and the remaining 26 adjacent grids in the three-dimensional grid matrix contain at least one air domain, or when the central grid contains an air domain and the remaining 26 adjacent grids in the three-dimensional grid matrix contain at least one organization domain, determine whether the central grid is a stepped grid.

[0064] After scanning the mesh model of the human tissue / air interface, the stepped meshes on both sides of the human tissue / air interface are obtained.

[0065] Optionally, the preset 3×3×3 three-dimensional mesh matrix includes the first vertex;

[0066] The steps for determining whether the central grid is a stepped grid include:

[0067] An xyz coordinate system is established with the first vertex of the three-dimensional mesh matrix as the origin. The three-dimensional mesh matrix is ​​divided into 9 mesh groups, each containing 3 meshes. The orthographic projections of the 3 meshes in the x, y, or z directions coincide.

[0068] In the x, y, or z directions, when the material domain types of three meshes in at least two of the nine mesh groups are not ordered in the same way, the central mesh is a stepped mesh.

[0069] Figure 3 This is a schematic diagram illustrating how to determine whether a central grid is a stepped grid, as provided in an embodiment of the present invention. Figure 3 As shown, for ease of understanding, rays can be drawn in this embodiment to determine whether the central grid is a stepped grid. Specifically, nine rays are drawn in the x, y, or z directions. These nine rays correspond to the nine grids that pass through the three-dimensional grid matrix. In this embodiment, only rays that pass through grids containing different material domain types are considered. When there are more than two such rays among the nine rays, and the order of the material domain types of the three grids they pass through is not exactly the same, the central grid is a stepped grid.

[0070] It should be noted that as long as any ray in any of the x, y, or z directions satisfies the above conditions, the central grid can be determined as a stepped grid.

[0071] Optionally, in step S4 above, the step of optimizing the conductivity and dielectric constant of the material domains of the stepped mesh based on the volume proportion of different material domain types contained in the three-dimensional mesh matrix centered on the stepped mesh, and then calculating the first electric field component of all meshes using the FDTD algorithm after all stepped meshes have been optimized, includes:

[0072] S401. Determine the volume percentage of meshes containing different material domain types in a three-dimensional mesh matrix centered on a stepped mesh.

[0073] S402. The effective conductivity and effective dielectric constant are obtained by nonlinear weighted averaging of the conductivity and dielectric constant of different material domain types using the volume ratio.

[0074] S403. Replace the conductivity of the material domain of the stepped mesh with the effective conductivity, and replace the dielectric constant with the effective dielectric constant.

[0075] After the conductivity and dielectric constant of the material domains of all S404 stepped meshes have been replaced, the first electric field component of all meshes is calculated using the FDTD algorithm.

[0076] In this embodiment, the effective conductivity is determined according to the following formula:

[0077]

[0078] In the formula, N represents the total number of material domain types in the three-dimensional mesh matrix centered on the stepped mesh, n i σ represents the volume percentage of the mesh containing the i-th material domain type. i Let σ represent the conductivity of the i-th material domain type, k represent the preset weight, and σ represent the conductivity of the i-th material domain type. eff The effective conductivity is calculated, where k can be adjusted according to the grid resolution and the radius of curvature of the human body surface at the center grid.

[0079] Furthermore, the effective dielectric constant is determined according to the following formula:

[0080]

[0081] In the formula, ε i ε represents the dielectric constant of the i-th material domain type. eff The effective dielectric constant is calculated.

[0082] Optionally, in step S5 above, the first electric field component at the apex of each material domain of the same organization is linearly interpolated or linearly extrapolated to obtain the second electric field component at the center of the material domain, and the second electric field component is used to calculate the local SAR.

[0083] Figure 4 This is a schematic diagram illustrating the calculation of local SAR provided in an embodiment of the present invention. Figure 4 The calculation process of local SAR is explained using the example shown. Figure 4 All known first electric field components are located at the vertices of various material domains within the same structure.

[0084] First, interpolation is used to calculate the electric field components.

[0085]

[0086] Continue interpolating to calculate the electric field components.

[0087]

[0088] Then, interpolation is used to calculate the material domain. The second electric field component at the center

[0089]

[0090] Similarly, the second electric field component is calculated. and

[0091] Finally, the computational material domain Local SAR at the center:

[0092]

[0093] In the formula: and These represent the electrical conductivity and mass density of the material domain, respectively.

[0094] Materials Domain The second electric field component at the center Extrapolation using electric field components is required:

[0095]

[0096] Continue to obtain the second electric field component Finally, the material domain was calculated. Local SAR at the center.

[0097] Additionally, the range of interpolation and extrapolation can be expanded to avoid using the first electric field component at the vertices of the material domain in a stepped mesh, and instead use only the first electric field component at the vertices of the material domain in a non-stepped mesh for interpolation and extrapolation. For stepped meshes, local SAR can be calculated using the effective conductivity, and the mass density can be modified proportionally based on the ratio of the effective conductivity to the original conductivity.

[0098] Figure 5 This is a schematic diagram of a uniform sphere model of the cornea or skin with a resolution of 0.25 mm, provided in an embodiment of the present invention. Please refer to... Figure 5 This embodiment uses a three-dimensional uniform spherical model of the cornea and skin as an example to further illustrate the FDTD nonlinear conformal filtering method for suppressing the step error of the specific absorption rate.

[0099] Specifically, the operating frequency is 2.45 GHz and the power density is 1.0 mW / cm². 2 The uniform plane wave radiation radius is 8 mm. The sphere is divided into a grid model with a resolution of 0.25 mm. The grid side length is less than 1 / 67 of the wavelength or skin depth. The electric field distribution in each grid is approximately linear. The numerical dispersion in the computational space is less than 0.1°.

[0100] The local SAR at the center of the material domain of the stepped grid on the surface of the sphere model is calculated using the standard FDTD algorithm and the FDTD nonlinear conformal filtering method with the step error of the suppression ratio absorptivity. The accurate theoretical results are calculated using the Mie algorithm. Then, the step error of the local SAR is calculated using the standard FDTD algorithm and the FDTD nonlinear conformal filtering method with the step error of the suppression ratio absorptivity, and the average is taken to obtain the average SAR step error.

[0101] Figure 6 This is a schematic diagram illustrating the variation of the average SAR step error of the corneal and skin sphere models provided in this embodiment of the invention with a preset weight k. Figure 6 As shown, the relationship between the average SAR step error of the FDTD nonlinear conformal filtering method for suppressing the step error of the specific absorption rate and the preset weight k in equations (1) and (2) is further explained.

[0102] k = 1.75 and k = 2.25 correspond to the minimum average SAR step error for the corneal and skin spheroid models, respectively. In the corneal spheroid model, the average SAR step error of the standard FDTD algorithm is 11.2%, while the FDTD nonlinear conformal filtering method, which suppresses the step error of specific absorptivity, reduces the error to 4.49%. In the skin spheroid model, the average SAR step error of the standard FDTD algorithm is 10.8%, while the FDTD nonlinear conformal filtering method, which suppresses the step error of specific absorptivity, reduces the error to 4.82%.

[0103] Figure 7 This is a schematic diagram of the SAR stepped error distribution of the corneal and skin spherical model cross-section provided in an embodiment of the present invention. Specifically, Figure 7 (a) shows the results of the corneal sphere model. Figure 7 (b) shows the results of the skin sphere model. Please refer to the results. Figure 7 (a)-7(b) On the cross section of the sphere model that passes through the center of the sphere and is parallel to the incident electric field and the propagation direction, referring to the Mie theory results of the SAR distribution, at the above k values, the step error of the local SAR of the FDTD nonlinear conformal filtering method that suppresses the step error of the specific absorption rate is significantly smaller than the step error of the local SAR of the standard FDTD algorithm.

[0104] Figure 8 This is a schematic diagram of a two-dimensional eyeball anatomical model with resolutions of 0.025mm and 0.25mm provided in an embodiment of the present invention. Further, as... Figure 8 As shown, this embodiment uses a two-dimensional eyeball anatomical model to verify the effectiveness of the above-mentioned FDTD nonlinear conformal filtering method for suppressing the step error of the absorption ratio.

[0105] Specifically, the operating frequency is 2.45 GHz and the power density is 1.0 mW / cm². 2 A two-dimensional anatomical model of the eyeball is constructed using uniform plane wave radiation, with the magnetic field of the uniform plane wave perpendicular to the cross-section of the eyeball. The eyeball is divided into two mesh models with resolutions of 0.025 mm and 0.25 mm, respectively. The 0.25 mm mesh model is generated by merging every 10 × 10 meshes of the 0.025 mm mesh model and is used to calculate the SAR distribution with step error using the standard FDTD algorithm and the FDTD nonlinear conformal filtering method to suppress the step error of the specific absorption rate. The 0.025 mm mesh model is used to calculate the accurate SAR distribution.

[0106] The accurate SAR distribution under the 0.025mm grid model is calculated using the standard FDTD algorithm. In order to eliminate the step error, this embodiment replaces the local SAR of the material domain of the body grid with the local SAR of the material domain of the body grid.

[0107] Figure 9 This is a schematic diagram illustrating the local SAR of replacing the material domain of the body surface mesh with the local SAR of the material domain of the body mesh, as provided in an embodiment of the present invention. Figure 9 As shown, the inventors discovered during their research that only the two layers of mesh on the body surface exhibited significant step errors in local SAR. Therefore, the replacement was performed only on these two layers. A search circle was drawn with the center of the material domain of each body surface mesh as its center. The radius of the search circle was gradually increased. When the material domain type of the body surface mesh was the same as that of the body surface mesh, and the center of the body surface mesh's material domain fell within the search circle, the local SAR of that body surface mesh's material domain would participate in the replacement. The search was performed with a step size equal to the side length of a mesh, increasing the radius until at least one replaceable material domain was found within the search circle. The local SAR of the searched body surface mesh's material domains was averaged and used as the local SAR of the body surface mesh's material domain, thus completing the replacement.

[0108] Next, based on the accurate SAR distribution with a resolution of 0.025 mm, each 10×10 grid is merged, and the local SARs are averaged to obtain an accurate SAR distribution with a resolution of 0.25 mm. Figure 10This is a schematic diagram of the accurate SAR distribution in the eye grid models with resolutions of 0.025mm and 0.25mm provided in the embodiments of the present invention. After the above replacement, merging and averaging, the SAR distribution of the surface area of ​​the eye grid model with resolution of 0.25mm is smooth, the step error is successfully eliminated, and the SAR distribution can be used as an accurate result to evaluate the step error of local SAR in the standard FDTD algorithm and the FDTD nonlinear conformal filtering method for suppressing the step error of the specific absorption rate.

[0109] Next, we will further explain the step error of local SAR using the standard FDTD algorithm and the FDTD nonlinear conformal filtering method for suppressing step error of the ratio of absorption.

[0110] Figure 11a This is a schematic diagram of the SAR distribution with step error in the surface region of a 0.25mm resolution eye grid model provided in the embodiment of the present invention. The SAR distribution of the standard FDTD algorithm has obvious fluctuations between adjacent grids and the step error is significant. The SAR distribution of the FDTD nonlinear conformal filtering method that suppresses the step error of specific absorption rate can smoothly transition between adjacent grids, and the step error is effectively suppressed.

[0111] Figure 11b This is a schematic diagram of the step error of local SAR in the surface region of a 0.25mm resolution eye-shaped mesh model provided in this embodiment of the invention. In each column of the mesh perpendicular to the x-axis in the surface region, the maximum local SAR step error is taken. Figure 11b The distribution of the maximum local SAR step error along the x-axis is shown in the figure. Averaging this distribution, the average maximum SAR step error of the standard FDTD algorithm is 26.1%, while the FDTD nonlinear conformal filtering method for suppressing the step error of specific absorptivity reduces the average maximum SAR step error to 12.6%.

[0112] Referring to Table 1, the FDTD nonlinear conformal filtering method for suppressing the step error of specific absorptivity also helps to improve the accuracy of SAR statistical results in the eye grid model:

[0113] Table 1

[0114]

[0115] It is evident that in cornea and skin, which are significantly affected by step error, the error of tissue-averaged SAR is reduced after using the FDTD nonlinear conformal filtering method for suppressing step error of specific absorptivity provided by this invention. For the maximum value of 10g-SAR of the eyeball, this method can still achieve high accuracy.

[0116] As can be seen from the above embodiments, the beneficial effects of the present invention are as follows:

[0117] This invention provides an FDTD nonlinear conformal filtering method for suppressing step error in specific absorptivity. First, a three-dimensional mesh matrix is ​​used to scan the human tissue / air interface grid-by-grid to determine the step mesh. Then, the effective conductivity and effective dielectric constant of the material domains surrounding the step mesh are optimized to obtain the effective conductivity and effective dielectric constant of the material domains within the step mesh. Next, after obtaining the first electric field component using the FDTD algorithm, the second electric field component at the center of each material domain within the same tissue is calculated through linear interpolation or linear extrapolation using the first electric field components at the vertices of each material domain. Local SAR and tissue-averaged SAR for that tissue are then calculated. Finally, after calculating the local SAR at the centers of all material domains for different tissues, 10-g SAR is calculated. This invention effectively suppresses the step error in local SAR on the human body surface and is suitable for calculating the SAR distribution and SAR statistical results of complex human body surface structures.

[0118] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0119] In the description of this specification, the references to terms such as "embodiment," "example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. In addition, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0120] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce a good effect.

[0121] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A FDTD nonlinear conformal filtering method to suppress the step error of specific absorption rate, characterized in that, The method comprises the following steps: acquiring a grid model uniformly divided in advance at a human tissue / air interface; determining, according to the positions of the conductivity and permittivity of the material in the FDTD algorithm on the grid, that one-eighth of each grid is a material domain; the material domain type includes a tissue domain and an air domain; sequentially taking each grid as a center grid and scanning by using a preset 3*3*3 three-dimensional grid matrix; judging whether the center grid is a step grid when the center grid contains a tissue domain and at least one air domain in the remaining 26 adjacent grids in the three-dimensional grid matrix, or when the center grid contains an air domain and at least one tissue domain in the remaining 26 adjacent grids in the three-dimensional grid matrix; after scanning the grid model of the human tissue / air interface, obtaining step grids on both sides of the human tissue / air interface; optimizing the conductivity and permittivity of the material domain of the step grid according to the volume ratio of the grid containing different material domain types in the three-dimensional grid matrix with the step grid as the center grid, and after the optimization of all step grids is completed, calculating the first electric field component of all grids by using the FDTD algorithm; linearly interpolating or linearly extrapolating the first electric field component of the vertex of each material domain of the same tissue to obtain the second electric field component of the center of the material domain, and calculating the local SAR by using the second electric field component; after obtaining the local SAR of all material domains of the same tissue, statistically calculating the tissue average SAR of the tissue, and after obtaining the local SAR of all material domains of different tissues, statistically calculating the 10-g SAR containing different tissues.

2. The FDTD nonlinear conformal filtering method of suppressing the step error of specific absorption rate according to claim 1, wherein, The preset 3*3*3 three-dimensional grid matrix comprises a first vertex; The step of judging whether the center grid is a step grid comprises: establishing an xyz coordinate system with the first vertex of the three-dimensional grid matrix as the origin, and the three-dimensional grid matrix is uniformly divided into 9 grid groups, each grid group comprising 3 grids, and the 3 grids in each grid group are coincident in x direction, y direction or z direction projection; when the sorting of the material domain types of the 3 grids in at least two grid groups in the 9 grid groups is not completely the same in x direction, y direction or z direction, the center grid is a step grid.

3. The FDTD nonlinear conformal filtering method of suppressing the step error of specific absorption rate according to claim 1, wherein, The step of optimizing the conductivity and permittivity of the material domain of the step grid according to the volume ratio of the grid containing different material domain types in the three-dimensional grid matrix with the step grid as the center grid, and after the optimization of all step grids is completed, calculating the first electric field component of all grids by using the FDTD algorithm, comprises: determining the volume ratio of the grid containing different material domain types in the three-dimensional grid matrix with the step grid as the center grid; performing nonlinear weighted average on the conductivity and permittivity of different material domain types by using the volume ratio to obtain effective conductivity and effective permittivity; replacing the conductivity of the material domain of the step grid with the effective conductivity and replacing the permittivity with the effective permittivity; after the replacement of the conductivity and permittivity of the material domain of all step grids is completed, calculating the first electric field component of all grids by using the FDTD algorithm.

4. The FDTD nonlinear conformal filtering method of suppressing the step error of specific absorption rate according to claim 3, wherein, The effective conductivity is determined according to the following formula: In the formula, N represents the total number of material domain types in a three-dimensional grid matrix with a stepped grid as a central grid, n i represents the volume fraction of the grid containing the i-th material domain type, σ i represents the electrical conductivity of the i-th material domain type, k represents a preset weight, and σ eff is the calculated effective electrical conductivity.

5. The FDTD nonlinear conformal filtering method of suppressing the step error of specific absorption rate according to claim 3, wherein, The effective permittivity is determined according to the following formula: where ε i represents the dielectric constant of the i-th material domain type, ε eff is the calculated effective dielectric constant.

Images