A positive problem modeling method and device, electronic equipment and storage medium

Abstract

This invention discloses a method, apparatus, electronic device, and storage medium for modeling forward EEG problems. The method includes: segmenting brain tissue structures based on brain structural images; dividing the head domain into a global mesh based on the segmented brain tissue structures; creating a skin-like electrode template based on the electrode line distribution of skin-like electrodes; locally refining the mesh based on the center position, shape, and electrode line cross-sectional diameter of the skin-like electrodes to obtain a head model; obtaining a skin-like electrode model on the head model with electrode locations based on the skin-like electrode template; obtaining a corresponding source model based on mesh division of the cerebral cortex surface; and constructing a personalized EEG forward EEG model using the finite element method based on the head model, source model, and skin-like electrode model. This invention addresses the problem of complex electric field distribution in the scalp region covered by the electrodes due to the complex fractal serpentine mesh structure of skin-like electrodes, thus improving the modeling accuracy of forward EEG problems.

Description

Technical Field

[0001] This invention relates to the field of physiological signal processing technology, specifically to a method, apparatus, electronic device, and storage medium for modeling positive problems. Background Technology

[0002] Non-invasive EEG is a method of acquiring extracranial EEG signals. Since the source of EEG signals is located inside the skull, scientific research requires inferring the electrical activity of intracranial discharge sources based on the EEG signals measured outside the skull. This process is called EEG Source Imaging (ESI).

[0003] Source imaging involves both forward problem modeling and inverse problem solving. Forward problem modeling involves solving for the numerical solution of scalp potentials given a head model, intracranial current source discharge distribution, and scalp electrode distribution. The inverse problem involves solving for the distribution of intracranial current sources given extracranial measurement signals and the forward problem model. Therefore, a high-precision EEG forward problem model is crucial for ensuring the accuracy of source imaging, and improving the accuracy of forward problem modeling is of great significance.

[0004] Please see Figures 1-3 Skin-like electrodes have seen rapid development in recent years. They eliminate the need for conductive gel and conform to the scalp surface, offering significant advantages over traditional electrodes and leading to their increasing practical applications. However, targeted research on forward problem modeling methods based on skin-like electrodes has been lacking. Traditional forward problem modeling simplifies the electrode model to a point or planar structure. However, because skin-like electrodes possess a fractal serpentine mesh structure, directly applying these methods to them fails to model the complex electric field distribution in the scalp area covered by the electrodes, thus affecting the accuracy of forward problem modeling. The main reasons are: 1) The serpentine structure of skin-like electrodes has a cross-sectional diameter of only tens of micrometers, while current EEG forward problem modeling, balancing computational complexity with the spatial resolution of existing structural brain images, uses a mesh side length of 1 mm, making it impossible for existing forward problem models to accurately describe the shape of skin-like electrodes; 2) The fractal serpentine structure of skin-like electrodes has an irregular shape that cannot be described by simple mathematical formulas. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a method, apparatus, electronic device, and storage medium for positive problem modeling, in order to solve the problem of low accuracy in using traditional methods to model positive problems of skin-like electrodes.

[0006] According to a first aspect, embodiments of the present invention provide a forward problem modeling method, comprising:

[0007] Based on brain structural images, brain tissue structures are segmented to obtain at least the following brain tissue structures: scalp, skull, and brain.

[0008] Based on the brain tissue structures obtained from the segmentation, the head domain is divided into an overall grid to obtain a primary grid.

[0009] Skin-like electrode templates were fabricated based on the electrode line distribution of skin-like electrodes;

[0010] The primary mesh is locally refined based on the center position, shape, and electrode line cross-sectional diameter of the skin-like electrodes to obtain the head model;

[0011] Based on the skin-like electrode template, a skin-like electrode model is obtained on a head model with electrode locations;

[0012] The corresponding source model is obtained based on the grid division of the cerebral cortex surface;

[0013] Based on the head model, the source model, and the skin-like electrode model, a personalized EEG positive problem model is constructed using the finite element method.

[0014] In some optional embodiments, the fabrication of the skin-like electrode template based on the electrode line distribution of the skin-like electrode includes:

[0015] The skin-like electrode mesh unit is obtained by meshing the outer rectangular region of the skin-like electrode.

[0016] The electrode line units in the skin-like electrode grid are set to 1 and the non-electrode line units are set to 0, and the data is stored using a matrix method to obtain the skin-like electrode template.

[0017] In some optional embodiments, the step of locally refining the primary mesh based on the center position, shape, and electrode line cross-sectional diameter of the skin-like electrodes to obtain the head model includes:

[0018] Based on the skin-like electrode template and the center position of the electrode, the primary grid on the outer surface of the scalp covered by the skin-like electrode is locally refined to obtain a secondary grid after local refinement; wherein, the size of the refined grid is smaller than the original network size, which is adapted to the size of the skin-like electrode mesh structure.

[0019] In some optional embodiments, the skin-like electrode model includes mesh node information and mesh cell information of the locally refined mesh in the head model, as well as a skin-like electrode template loaded with conductivity information.

[0020] In some optional implementations, the step of constructing a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element algorithm includes:

[0021] By applying EEG mathematical equations and boundary conditions to the head model, the source model, and the skin-like electrode model, a personalized EEG positive problem conduction matrix is ​​calculated.

[0022] The mathematical equation for electroencephalography is:

[0023]

[0024] Where σ represents tissue conductivity, u represents electrical potential, and J p The initial current is represented by Ω, the head model is represented by inΩ, the entire domain of the head model is represented by x, and a coordinate point within the entire domain of the head model is represented by x.

[0025] The boundary conditions are as follows:

[0026]

[0027]

[0028]

[0029] in, The region on E represents the boundary area excluding the portion covered by the skin-like electrode, and Z represents the portion of the boundary area covered by the skin-like electrode. l U represents the contact resistance of the skin-like electrode to the scalp. l e represents the potential at the l-th skin-like electrode. l Let L represent the coverage area of ​​the l-th type of skin electrode, L represent the number of electrodes, and n represent the outward normal vector perpendicular to the surface.

[0030] In some optional implementations, the step of constructing a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element algorithm includes:

[0031] The mathematical equations are weakened, and the calculations are simplified by integration by parts:

[0032]

[0033] in, Let U be the applied shape function, and U = [U1 … U L ] represents the potential at L electrodes.

[0034] In some optional implementations, the step of constructing a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element algorithm includes:

[0035] The head domain is discretized using a hexahedron, and the discretization is represented as follows:

[0036]

[0037] Among them, Ω h Representing the discrete domain, u h (x) represents the voltage in the discrete domain, v j Let P represent the voltage at the j-th discrete grid vertex, and let P represent the number of vertices in the discrete grid. The shape function representing the j-th discrete grid vertex;

[0038] in, It satisfies the following properties:

[0039] x = (x, y, z)

[0040]

[0041] in,

[0042] Apply the weakening equation to the discrete domain:

[0043]

[0044] Where, v = (v1 … v P Let X represent the voltage at the vertices of P discrete grid cells, where X = [X1 … X2]. M [] indicates M intracranial discharge sources.

[0045] In some optional implementations, the step of constructing a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element algorithm includes:

[0046] The current flowing out of the electrode covering the scalp region is constrained to be the same as the inflow current. Applying this to the discrete domain, we have:

[0047]

[0048] In some optional embodiments, the method further includes:

[0049] By assembling the equations, we obtain:

[0050]

[0051] By simplifying the matrix using the elimination method, the transmission matrix L can be obtained. CEM for:

[0052] L CEM =R(B T A -1 BC) -1 B T A-1 G

[0053] in, and And it satisfies:

[0054]

[0055]

[0056]

[0057]

[0058] Where, j k is the shape function of the source model.

[0059] According to a second aspect, embodiments of the present invention provide a forward problem modeling apparatus, comprising:

[0060] The segmentation module is used to segment brain tissue structures based on brain structural images, obtaining at least the following brain tissue structures: scalp, skull, and brain.

[0061] The grid division module is used to perform overall grid division of the head domain based on the segmented brain tissue structures to obtain a primary grid.

[0062] A skin-like electrode template fabrication module is used to fabricate skin-like electrode templates based on the electrode line distribution of skin-like electrodes.

[0063] The mesh refinement module is used to locally refine the primary mesh based on the center position, shape, and electrode line cross-sectional diameter of the skin-like electrodes to obtain a head model.

[0064] A registration module is used to obtain a skin-like electrode model on a head model with electrode positions based on the skin-like electrode template.

[0065] The source model acquisition module is used to obtain the corresponding source model based on the mesh division of the cerebral cortex surface;

[0066] The positive problem model construction module is used to construct a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element algorithm.

[0067] According to a third aspect, embodiments of the present invention provide an electronic device, comprising:

[0068] The memory and the processor are interconnected, the memory is used to store a computer program, and when the computer program is executed by the processor, it implements any of the positive problem modeling methods described in the first aspect above.

[0069] According to a fourth aspect, embodiments of the present invention provide a computer-readable storage medium for storing a computer program, which, when executed by a processor, implements any of the forward problem modeling methods described in the first aspect.

[0070] Traditional forward problem modeling simplifies electrode models to point or planar structures. However, because skin-like electrodes possess a fractal serpentine mesh structure, directly applying the above methods to them fails to model the complex electric field distribution in the scalp area covered by the skin-like electrodes. The main reasons are: 1) The diameter of the serpentine electrode lines in skin-like electrodes is only tens of micrometers, while the current mesh resolution for forward problem modeling of EEG is 1 mm, making it impossible to accurately describe their shape; 2) The fractal serpentine structure of skin-like electrodes is irregular and cannot be described by simple mathematical formulas. To address the first problem, describing the serpentine electrode lines of skin-like electrodes requires increasing the resolution of traditional models. However, this increases the model resolution, leading to an excessively large stiffness matrix in the finite element method (FED) calculation, making it difficult or even impossible to solve. Furthermore, for areas where high resolution is not required, increasing the resolution can add excessive detail, affecting computation. Therefore, to avoid excessive computational burden and ensure accuracy, a less refined mesh should be used to describe the serpentine electrode lines. Therefore, this invention proposes a method based on local mesh refinement to address the problem of difficulty in describing serpentine fractal structures caused by the low resolution of traditional models. Secondly, addressing the complexity of serpentine fractal structures, this invention cleverly utilizes a unified skin-like electrode template, based on the fact that electrode fabrication is based on a unified template, to minimize repetitive work, reduce computational burden, and improve computational speed.

[0071] In this embodiment of the invention, the proposed skin-like electrode complete electrode model achieves a fine simulation of the potential distribution of the positive problem through mesh subdivision and conformal mapping of the skin-like electrode, thereby improving the accuracy of the positive problem model. Attached Figure Description

[0072] The features and advantages of the invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way. In the drawings:

[0073] Figure 1 A schematic diagram illustrating conformal contact between a skin-like electrode and a headpiece surface;

[0074] Figure 2 Another schematic diagram illustrating conformal conformal contact between a skin-like electrode and a headpiece surface;

[0075] Figure 3 Diagram showing the connection of skin-like electrodes;

[0076] Figure 4 A flowchart illustrating a forward problem modeling method provided in an embodiment of the present invention;

[0077] Figure 5 A schematic diagram illustrating the process of the positive problem modeling method provided in this embodiment of the invention;

[0078] Figure 6 This is a schematic diagram of the structure of a forward problem modeling device provided in an embodiment of the present invention;

[0079] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0080] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0081] It should be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. Furthermore, the terms "first," "second," etc., 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. In the following descriptions of embodiments, "a plurality of" means two or more, unless otherwise expressly specified.

[0082] Definition of noun:

[0083]

[0084] Please see Figure 4 This invention provides a method for modeling positive problems, comprising the following steps:

[0085] S101: Based on brain structural images, segment brain tissue structures to obtain at least the following brain tissue structures: scalp, skull, and brain;

[0086] Specifically, brain tissue structure segmentation can be divided into: gray matter, white matter, cerebrospinal fluid, scalp, and skull, etc. To improve calculation speed, it can also be simply divided into three layers: scalp, skull, and brain. In addition, the segmented brain tissue structure can also include the eyeball and / or eye socket, etc.

[0087] Brain structural images include images acquired using computed tomography (CT) and / or magnetic resonance imaging (MRI);

[0088] S102: Based on the brain tissue structures obtained from the segmentation, the head domain (defined) is divided into an overall grid to obtain a primary grid; in order to distinguish it from the subsequent local refinement grid, the grid divided in this step is called a primary grid;

[0089] At this point, the mesh generation accuracy can be referenced from the mesh generation accuracy in traditional forward problem modeling methods, considering that the spatial resolution of brain structural images is typically 1 mm. 3 In terms of computational complexity, the grid resolution can be directly set to 1mm. 3 .

[0090] S103: Fabrication of a skin-like electrode template based on the electrode line distribution of a skin-like electrode. Specifically, this involves constructing a skin-like electrode template based on real skin-like electrodes. Please refer to [link / reference]. Figure 5 Step a in the text;

[0091] S104: Based on the center position, shape, and electrode line cross-sectional diameter of the skin-like electrodes, the primary mesh is locally refined to obtain a secondary mesh, i.e., the head model. (See also...) Figure 5 Step b in the process;

[0092] S105: Based on the skin-like electrode template, a skin-like electrode model, i.e., an electrode model, is obtained on the head model with electrode positions; specifically, the skin-like electrode model can be obtained by mapping the fractal serpentine structure of the skin-like electrode line to the corresponding refined region of the head model based on the skin-like electrode template.

[0093] S106: Obtain the corresponding source model based on the mesh division of the cerebral cortex surface; additionally, for the cortical surface division based on the template, the source model needs to be registered to the head model, please refer to [link to relevant documentation]. Figure 5 Step d;

[0094] S107: Based on the head model, the source model, and the skin-like electrode model, a personalized EEG forward problem model is constructed using the finite element method (see [link to model]). Figure 5Step e). Specifically, based on the locally refined head model (i.e., the head model), the source model, and the skin-like electrode model, the finite element method is used to calculate the conduction matrix describing the relationship between EEG measurement voltage and intracranial sources, i.e., the EEG positive problem model. Methods for handling source singularities include the Saint-Venant method, partial integration, subtraction methods, and the Whitney method, among others.

[0095] Traditional forward problem modeling simplifies electrode models to point or planar structures. However, because skin-like electrodes possess a fractal serpentine mesh structure, directly applying the above methods to them fails to model the complex electric field distribution in the scalp area covered by the skin-like electrodes. The main reasons are: 1) The diameter of the serpentine electrode lines in skin-like electrodes is only tens of micrometers, while the current mesh resolution for forward problem modeling of EEG is 1 mm, making it impossible to accurately describe their shape; 2) The fractal serpentine structure of skin-like electrodes is irregular and cannot be described by simple mathematical formulas. To address the first problem, describing the serpentine electrode lines of skin-like electrodes requires increasing the resolution of traditional models. However, this increases the model resolution, leading to an excessively large stiffness matrix in the finite element method (FED) calculation, making it difficult or even impossible to solve. Furthermore, for areas where high resolution is not required, increasing the resolution can add excessive detail, affecting computation. Therefore, to avoid excessive computational burden and ensure accuracy, a less refined mesh should be used to describe the serpentine electrode lines. Therefore, this invention proposes a method based on local mesh refinement to address the problem of difficulty in describing serpentine fractal structures caused by the low resolution of traditional models. Secondly, addressing the complexity of serpentine fractal structures, this invention cleverly utilizes a unified skin-like electrode template, based on the fact that electrode fabrication is based on a unified template, to minimize repetitive work, reduce computational burden, and improve computational speed.

[0096] In this embodiment of the invention, the proposed skin-like electrode complete electrode model achieves a fine simulation of the potential distribution of the positive problem through mesh subdivision and conformal mapping of the skin-like electrode, thereby improving the accuracy of the positive problem model.

[0097] In some specific embodiments, the fabrication of the skin-like electrode template based on the electrode line distribution of the skin-like electrode includes:

[0098] a) Based on the outer rectangular region of the skin-like electrode, mesh division is performed to obtain the skin-like electrode mesh unit;

[0099] b) Set the electrode line units in the skin-like electrode grid to 1 and the non-electrode line (blank) units to 0, and store them using a matrix method to obtain the skin-like electrode template. (See [link to relevant documentation]). Figure 5 Step a.

[0100] Based on the above steps, the resolution of the skin-like electrode mesh is the minimum value (um or nm level) that can identify the cross-sectional radius of the electrode line. The obtained skin-like electrode mesh has a total of M1*M2 units, where the values ​​of M1 and M2 are: the length of the circumscribed rectangle / mesh resolution and the width of the circumscribed rectangle / mesh resolution.

[0101] The embodiments of the present invention fabricate skin-like electrode templates based on the electrode line distribution of skin-like electrodes, which can achieve rapid mapping of skin-like electrodes.

[0102] In some optional implementations, the step of locally refining the primary mesh based on the center position, shape, and electrode line cross-sectional diameter of the skin-like electrodes to obtain the head model includes:

[0103] Based on the skin-like electrode template and the electrode center position, the primary mesh on the outer surface of the scalp covered by the skin-like electrode is locally refined to obtain a secondary mesh. The size of the refined mesh is smaller than the original mesh size, thus matching the size of the skin-like electrode mesh structure.

[0104] Specifically, firstly, local interpolation is performed on each primary mesh in the electrode coverage area to obtain a local super-resolution mesh. Secondly, after obtaining the locally refined secondary mesh, the newly generated mesh nodes and mesh connectivity relationships are added to the primary mesh and the skin-like electrode model (see...). Figure 5 Step b).

[0105] Furthermore, it's important to note that when the mesh resolution of the head model is an integer multiple of the diameter of the non-electrode line cross-section, simple interpolation cannot achieve refinement. To reduce computational complexity, a simple diameter dilation interpolation method is used. That is, the local refinement mesh resolution of the secondary mesh is the closest value to the electrode line cross-section diameter that allows for local interpolation (e.g., when the mesh resolution is 1 mm and the electrode diameter is 0.18 mm, a 0.2 mm local refinement mesh will be used). This method sacrifices computational accuracy to some extent, but it reduces computational complexity and significantly decreases the number of meshes, thus improving computational speed.

[0106] In this embodiment of the invention, on the one hand, to achieve the mapping of the distribution of skin-like electrode lines, it is necessary to improve the resolution of the head grid; on the other hand, to avoid the data curse caused by an overly dense grid, the number of locally refined grid nodes should be minimized. Therefore, the resolution of the head grid covered by the skin-like electrodes is at least the same as or slightly greater than the resolution of the electrode grid. At this point, the resolution of the head grid covered by the skin-like electrodes is improved from the mm level to the level of the minimum cross-sectional area of ​​the skin-like electrode lines, i.e., the μm level. This embodiment of the invention achieves local refinement of the head region covered by the skin-like electrodes based on a local interpolation method, and builds an EEG positive problem model based on the skin-like electrode boundaries based on this refinement model.

[0107] The skin-like electrode model includes mesh node information and mesh cell information of the locally refined mesh in the head model, as well as a skin-like electrode template loaded with conductivity information.

[0108] In some optional embodiments, obtaining a skin-like electrode model on a head model with electrode locations based on the skin-like electrode template includes:

[0109] a) Add the mesh node information and mesh cell information of the locally refined mesh in the head model to the skin-like electrode model. Specifically, match the mesh node information and cell information with the electrode labels;

[0110] b) Load the conductivity information onto the skin-like electrode template and add it (i.e., the skin-like electrode template loaded with conductivity information) to the skin-like electrode model. Specifically, multiply the conductivity and the electrode template matrix.

[0111] The final skin-like electrode model includes at least the following information: grid node coordinates, grid cell matrix, grid cell labels, and the skin-like electrode template loaded with conductivity information.

[0112] In some specific implementations, the construction of a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element algorithm includes:

[0113] By applying EEG mathematical equations and boundary conditions to the head model, the source model, and the skin-like electrode model, a personalized EEG positive problem conduction matrix is ​​calculated.

[0114] The mathematical equation for electroencephalography is:

[0115]

[0116] Where σ represents tissue conductivity, u represents electrical potential, and J pThe initial current is represented by Ω, the head model is represented by inΩ, the entire domain of the head model is represented by x, and a coordinate point within the entire domain of the head model is represented by x.

[0117] The boundary conditions are as follows:

[0118]

[0119]

[0120]

[0121] in, The region on E represents the boundary area excluding the portion covered by the skin-like electrode, and Z represents the portion of the boundary area covered by the skin-like electrode. l U represents the contact resistance of the skin-like electrode to the scalp. l e represents the potential at the l-th skin-like electrode. l Let L represent the coverage area of ​​the l-th type of skin electrode, L represent the number of electrodes, and n represent the outward normal vector perpendicular to the surface.

[0122] The process of determining the mathematical equations for electroencephalography (EEG) is as follows:

[0123] Because the signal frequency in the brain is relatively low, the quasi-static Maxwell equations are commonly used to describe the relationship between neural discharge sources and brain electrical signals. The brain is described as a conductor model with different conductivities, typically divided into five layers (scalp, skull, cerebrospinal fluid, gray matter, and white matter). The quasi-static Maxwell equations are as follows:

[0124]

[0125]

[0126]

[0127]

[0128] Where E represents the electric field strength, B represents the magnetic field strength, ρ represents the charge density, ε represents the dielectric constant, μ represents the permeability, and J represents the current density.

[0129] Furthermore:

[0130]

[0131]

[0132] And because Right now Therefore, the mathematical equation for electroencephalography (EEG) is:

[0133]

[0134] In some specific implementations, the construction of a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element algorithm includes:

[0135] The mathematical equations are weakened, and the calculations are simplified by integration by parts:

[0136]

[0137] in, Let U be the applied shape function, and U = [U1 … U L ] represents the potential at L electrodes.

[0138] In some specific implementations, the construction of a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element algorithm includes:

[0139] The head domain is discretized using a hexahedron, and the discretization is represented as follows:

[0140]

[0141] Among them, Ω h Representing the discrete domain, u h (x) represents the voltage in the discrete domain, v j Let P represent the voltage at the j-th discrete grid vertex, and let P represent the number of vertices in the discrete grid. The shape function representing the j-th discrete grid vertex;

[0142] in, It satisfies the following properties:

[0143] x = (x, y, z)

[0144]

[0145] in,

[0146] a j b j c j d j e j f j g j h j The coefficients of the shape function;

[0147] Apply the weakening equation to the discrete domain:

[0148]

[0149] Where, v = (v1 … v P Let X represent the voltage at the vertices of P discrete grid cells, where X = [X1 … X2]. M ] represents M intracranial discharge sources, as shown above: U = [U1 … U L ] represents the potential at L electrodes.

[0150] In some optional specific implementations, the step of constructing a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element algorithm includes: constraining the current in the scalp area covered by the outflowing electrode to be consistent with the inflowing current, and applying this constraint (i.e., the constraint condition: the current in the scalp area covered by the outflowing electrode is consistent with the inflowing current) to the discrete domain as follows:

[0151]

[0152] In some specific implementations, the method further includes:

[0153] By assembling the equations, we obtain:

[0154]

[0155] By simplifying the matrix using the elimination method, the transmission matrix L can be obtained. CEM for:

[0156] L CEM =R(B T A -1 BC) -1 B T A -1 G

[0157] in, and And it satisfies:

[0158]

[0159]

[0160]

[0161]

[0162] Where, j k is the shape function of the source model.

[0163] Accordingly, please refer to Figure 6 This invention provides a positive problem modeling apparatus, which includes:

[0164] The segmentation module 601 is used to segment brain tissue structures based on brain structural images, and to obtain at least the following brain tissue structures: scalp, skull and brain.

[0165] The grid division module 602 is used to perform overall grid division of the head domain based on the segmented brain tissue structures to obtain a primary grid.

[0166] Skin-like electrode template fabrication module 603 is used to fabricate a skin-like electrode template based on the electrode line distribution of skin-like electrodes;

[0167] Mesh refinement module 604 is used to locally refine the primary mesh based on the center position, shape and electrode line cross-sectional diameter of the skin-like electrode to obtain a head model;

[0168] The registration module 605 is used to obtain a skin-like electrode model on a head model with electrode positions based on the skin-like electrode template.

[0169] The source model acquisition module 606 is used to obtain the corresponding source model based on the grid division of the surface of the cerebral cortex;

[0170] The positive problem model construction module 607 is used to construct a personalized EEG positive problem model based on the head model, the source model and the skin-like electrode model using the finite element algorithm.

[0171] In some specific embodiments, the skin-like electrode template fabrication module 603 includes:

[0172] The partitioning unit is used to perform meshing based on the outer rectangular region of the skin-like electrode to obtain the skin-like electrode mesh unit;

[0173] The template generation unit is used to set the electrode line units in the skin-like electrode grid unit to 1 and the non-electrode line units to 0, and store them using a matrix method to obtain the skin-like electrode template.

[0174] In some specific implementations, the mesh refinement module 604 is specifically used for:

[0175] Based on the skin-like electrode template and the center position of the electrode, the primary grid on the outer surface of the scalp covered by the skin-like electrode is locally refined to obtain a secondary grid after local refinement; wherein, the size of the refined grid is smaller than the original network size, which is adapted to the size of the skin-like electrode mesh structure.

[0176] In some specific implementations, the skin-like electrode model includes mesh node information and mesh cell information of the locally refined mesh in the head model, as well as a skin-like electrode template loaded with conductivity information.

[0177] In some specific implementations, the positive problem model construction module 607 is specifically used to apply EEG mathematical equations and boundary conditions to the head model, the source model, and the skin-like electrode model to calculate a personalized EEG positive problem conduction matrix;

[0178] The mathematical equation for electroencephalography is:

[0179]

[0180] Where σ represents tissue conductivity, u represents electrical potential, and J p The initial current is represented by Ω, the head model is represented by inΩ, the entire domain of the head model is represented by x, and a coordinate point within the entire domain of the head model is represented by x.

[0181] The boundary conditions are as follows:

[0182]

[0183]

[0184]

[0185] in, The region on E represents the boundary area excluding the portion covered by the skin-like electrode, and Z represents the portion of the boundary area covered by the skin-like electrode. l U represents the contact resistance of the skin-like electrode to the scalp. l e represents the potential at the l-th skin-like electrode. l Let L represent the coverage area of ​​the l-th type of skin electrode, L represent the number of electrodes, and n represent the outward normal vector perpendicular to the surface.

[0186] In some specific implementations, the positive problem model construction module 607 is further specifically used for:

[0187] The mathematical equations are weakened, and the calculations are simplified by integration by parts:

[0188]

[0189] in, Let U be the applied shape function, and U = [U1 … U L ] represents the potential at L electrodes.

[0190] In some specific implementations, the positive problem model construction module 607 is further specifically used for:

[0191] The head domain is discretized using a hexahedron, and the discretization is represented as follows:

[0192]

[0193] Among them, Ω h Representing the discrete domain, u h (x) represents the voltage in the discrete domain, v j Let P represent the voltage at the j-th discrete grid vertex, and let P represent the number of vertices in the discrete grid. The shape function representing the j-th discrete grid vertex;

[0194] in, It satisfies the following properties:

[0195] x = (x, y, z)

[0196]

[0197] in,

[0198] Apply the weakening equation to the discrete domain:

[0199]

[0200] Where, v = (v1 … v P Let X represent the voltage at the vertices of P discrete grid cells, where X = [X1 … X2]. M [] indicates M intracranial discharge sources.

[0201] In some specific implementations, the positive problem model construction module 607 is further specifically used for:

[0202] The current flowing out of the electrode covering the scalp region is constrained to be the same as the inflow current. Applying this to the discrete domain, we have:

[0203]

[0204] In some specific implementations, the positive problem model construction module 607 is further specifically used for:

[0205] By assembling the equations, we obtain:

[0206]

[0207] By simplifying the matrix using the elimination method, the transmission matrix L can be obtained. CEM for:

[0208] L CEM =R(B T A -1 BC) -1 B T A -1 G

[0209] in, and And it satisfies:

[0210]

[0211]

[0212]

[0213] Where, j k is the shape function of the source model.

[0214] The embodiments of the present invention are device embodiments based on the same inventive concept as the method embodiments described above. Therefore, for specific technical details and corresponding technical effects, please refer to the method embodiments described above, and they will not be repeated here.

[0215] This invention also provides an electronic device, such as... Figure 7 As shown, the electronic device may include a processor 71 and a memory 72, wherein the processor 71 and the memory 72 can communicate with each other via a bus or other means. Figure 7 Taking the example of a connection between China and Israel via a bus.

[0216] Processor 71 can be a central processing unit (CPU). Processor 71 can 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 of the above types of chips.

[0217] Memory 72, 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 positive problem modeling method in this embodiment of the invention (e.g., Figure 6 The module shown includes segmentation module 601, mesh generation module 602, skin-like electrode template fabrication module 603, mesh refinement module 604, registration module 605, source model acquisition module 606, and forward problem model construction module 607. The processor 71 executes various processor functions and data processing by running non-transitory software programs, instructions, and modules stored in the memory 72, thereby implementing the forward problem modeling method in the above method embodiments.

[0218] The memory 72 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 the processor 71, etc. Furthermore, the memory 72 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, the memory 72 may optionally include memory remotely located relative to the processor 71, and these remote memories may be connected to the processor 71 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0219] The one or more modules are stored in the memory 72, and when executed by the processor 71, the positive problem modeling method in the above method embodiment is executed.

[0220] The specific details of the aforementioned electronic device can be understood by referring to the relevant descriptions and effects in the method embodiments, and will not be repeated here.

[0221] Accordingly, this embodiment of the invention also provides a computer-readable storage medium for storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described positive problem modeling method embodiment and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0222] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0223] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0224] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A forward problem modeling method, characterized in that, include: Based on brain structural images, brain tissue structures are segmented to obtain at least the following brain tissue structures: scalp, skull, and brain. Based on the brain tissue structures obtained from the segmentation, the head domain is divided into an overall grid to obtain a primary grid. Skin-like electrode templates were fabricated based on the electrode line distribution of skin-like electrodes; The primary mesh is locally refined based on the center position, shape, and electrode line cross-sectional diameter of the skin-like electrodes to obtain the head model; Based on the skin-like electrode template, a skin-like electrode model is obtained on a head model with electrode locations; The corresponding source model is obtained based on the grid division of the cerebral cortex surface; Based on the head model, the source model, and the skin-like electrode model, a personalized EEG positive problem model is constructed using the finite element algorithm. The step of constructing a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element method includes: By applying EEG mathematical equations and boundary conditions to the head model, the source model, and the skin-like electrode model, a personalized EEG positive problem conduction matrix is ​​calculated. The mathematical equation for electroencephalography is: in, Indicates the tissue conductivity. Represents electric potential, Indicates the initial current. Ω This refers to the head model. inΩ This refers to the entire domain of the head model. x This represents a coordinate point within the entire head model area; The boundary conditions are as follows: in, This refers to the boundary region excluding the portion of the boundary region covered by the skin-like electrode. This indicates the partial boundary area covered by the skin-like electrode. Indicates the contact resistance of the skin-like electrode to the scalp. Indicates the first l The potential at the aforementioned skin electrode Indicates the first The coverage area of ​​each type of skin electrode, where L represents the number of electrodes. This represents the outward normal vector perpendicular to the surface.

2. The method according to claim 1, characterized in that, The fabrication of the skin-like electrode template based on the electrode line distribution of the skin-like electrode includes: The skin-like electrode mesh unit is obtained by meshing the outer rectangular region of the skin-like electrode. The electrode line units in the skin-like electrode grid are set to 1 and the non-electrode line units are set to 0, and the data is stored using a matrix method to obtain the skin-like electrode template.

3. The method according to claim 1, characterized in that, The process of locally refining the primary mesh based on the center position, shape, and electrode line cross-sectional diameter of the skin-like electrodes to obtain the head model includes: Based on the skin-like electrode template and the center position of the electrode, the primary grid on the outer surface of the scalp covered by the skin-like electrode is locally refined to obtain a secondary grid after local refinement; wherein, the size of the refined grid is smaller than the original network size, which is adapted to the size of the skin-like electrode mesh structure.

4. The method according to claim 1, characterized in that, The skin-like electrode model includes mesh node information and mesh cell information of the locally refined mesh in the head model, as well as a skin-like electrode template loaded with conductivity information.

5. The method according to claim 1, characterized in that, The method for constructing a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element method includes: The mathematical equations are weakened, and the calculations are simplified by integration by parts: in, For the applied shape function, This represents the potential at the L electrodes.

6. The method according to claim 5, characterized in that, The method for constructing a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element method includes: The head domain is discretized using a hexahedron, and the discretization is represented as follows: in, Represents the discrete domain. Represents voltage in the discrete domain. Indicates the first The voltage at each vertex of a discrete grid, where P represents the number of vertices in the discrete grid. Indicates the first Shape functions of discrete mesh vertices; in, It satisfies the following properties: Apply the weakening equation to the discrete domain: in, This represents the voltage at the vertices of P discrete grid cells. This represents M intracranial discharge sources.

7. The method according to claim 6, characterized in that, The method for constructing a personalized EEG positive problem model based on the head model, the source model, and the skin-like electrode model using the finite element method includes: The current flowing out of the electrode covering the scalp region is constrained to be the same as the inflow current. Applying this to the discrete domain, we have: 。 8. The method according to claim 7, characterized in that, Also includes: By assembling the equations, we obtain: The transformation matrix can be obtained by simplifying the matrix using the elimination method. for: in, ,and , i =1,2… L , j =1,2… L ; And satisfy: in, is the shape function of the source model.

9. A forward problem modeling apparatus, characterized in that, include: The segmentation module is used to segment brain tissue structures based on brain structural images, obtaining at least the following brain tissue structures: scalp, skull, and brain. The grid division module is used to perform overall grid division of the head domain based on the segmented brain tissue structures to obtain a primary grid. A skin-like electrode template fabrication module is used to fabricate skin-like electrode templates based on the electrode line distribution of skin-like electrodes. The mesh refinement module is used to locally refine the primary mesh based on the center position, shape, and electrode line cross-sectional diameter of the skin-like electrodes to obtain a head model. A registration module is used to obtain a skin-like electrode model on a head model with electrode positions based on the skin-like electrode template. The source model acquisition module is used to obtain the corresponding source model based on the mesh division of the cerebral cortex surface; The positive problem model construction module is used to construct a personalized EEG positive problem model based on the head model, the source model and the skin-like electrode model using the finite element algorithm. Specifically, the positive problem model construction module is used to: apply EEG mathematical equations and boundary conditions to the head model, the source model, and the skin-like electrode model to calculate a personalized EEG positive problem conduction matrix; The mathematical equation for electroencephalography is: in, Indicates the tissue conductivity. Represents electric potential, Indicates the initial current. Ω This refers to the head model. inΩ This refers to the entire domain of the head model. x This represents a coordinate point within the entire head model area; The boundary conditions are as follows: in, This refers to the boundary region excluding the portion of the boundary region covered by the skin-like electrode. This indicates the partial boundary area covered by the skin-like electrode. Indicates the contact resistance of the skin-like electrode to the scalp. Indicates the first l The potential at the aforementioned skin electrode Indicates the first The coverage area of ​​each type of skin electrode, where L represents the number of electrodes. This represents the outward normal vector perpendicular to the surface.

10. An electronic device, characterized in that, include: A memory and a processor are interconnected, the memory being used to store a computer program, which, when executed by the processor, implements the forward problem modeling method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, which, when executed by a processor, implements the forward problem modeling method according to any one of claims 1 to 8.

Images