Low-frequency conductivity tensor modeling method, system, medium, and electronic device for brain

Abstract

The present invention provides a method, system, medium, and electronic device for modeling a low-frequency conductivity tensor of the brain. The method includes the following steps: acquiring a phase image of a fast spin echo sequence of the brain; acquiring a high-frequency conductivity image of the brain based on the phase image; acquiring a diffusion-weighted amplitude image of the brain based on a multi-excitation diffusion-weighted sequence; acquiring diffusion microstructure parameters of the brain using a multi-layer model based on the diffusion-weighted amplitude image; generating a low-frequency conductivity tensor image of the brain based on the high-frequency conductivity image and the diffusion microstructure parameters; generating a digital skull model based on a magnetization-prepared fast gradient echo sequence of the brain; and generating a low-frequency conductivity tensor model of the brain based on the low-frequency conductivity tensor image and the digital skull model. The method, system, medium, and electronic device for modeling a low-frequency conductivity tensor of the brain of the present invention provide more accurate target planning for tTIS.

Description

Technical Field

[0001] The present invention relates to the technical field of biomedical engineering, and in particular to a method, system, medium, and electronic equipment for modeling low-frequency conductivity tensor of the brain. Background Art

[0002] Tissue conductivity is a crucial biophysical parameter. High-frequency conductivity plays a crucial role in estimating the specific absorption ratio (SAR). Low-frequency conductivity, on the other hand, plays a crucial role in neuromodulation. For example, an accurate low-frequency conductivity model of the brain can improve source localization and enhance the accuracy of transcranial electrical stimulation field calculations.

[0003] Traditional transcranial electrical stimulation, including transcranial direct current stimulation (tDCS) and transcranial alternating current stimulation (tACS), can typically only stimulate superficial brain regions and lack a focused function. Therefore, the use of a personalized brain conductivity model has little impact on accuracy.

[0004] Transcranial temporal interference stimulation (tTIS) is a novel, noninvasive transcranial neuromodulation method. This method uses multiple pairs of electrodes to inject high-frequency sinusoidal waves with slight frequency differences into the brain, generating low-frequency amplitude-modulated waves in the target brain region. This technique, therefore, offers strong focusing capabilities. However, differences in conductivity models significantly influence target location and electrode placement, making its effectiveness in treating neurological disorders highly dependent on the accuracy of the stimulation target. Localizing the target nuclei and estimating the electric field intensity within the target region requires constructing a brain conductivity model for the individual subject. However, to date, all tTIS methods have used empirically derived brain conductivity values, meaning that all subjects follow the recommended conductivity values ​​from the literature, without considering individual variability in conductivity. Studies have shown that brain conductivity varies with physiological and pathological factors. Therefore, developing an accurate, personalized low-frequency conductivity tensor imaging method to construct a brain conductivity model is crucial for accurate tTIS target location and subsequent neurological disease treatment.

[0005] Conductivity tensor imaging (CTI) is a method that can reconstruct the anisotropic low-frequency conductivity tensor of biological tissues. Its basic principle is as follows: high-frequency current can ignore the permeability of the cell membrane, allowing high-frequency current to penetrate the cell, and therefore the high-frequency current is contributed by both the intracellular and extracellular conductivities. Low-frequency current, on the other hand, is obstructed by the cell membrane and flows primarily in the extracellular space, so the low-frequency conductivity is mainly contributed by the extracellular conductivity. Conductivity tensor imaging is an imaging method that decomposes high-frequency conductivity based on the difference in intracellular and extracellular volume fractions to obtain low-frequency conductivity. In the prior art, magnetic resonance electrical property imaging can measure high-frequency conductivity by inverting high-frequency conductivity based on the B1 field, while diffusion microstructure imaging can measure cell volume fraction, diffusivity, and fiber orientation. Conductivity tensor imaging combines these two methods, decomposing the high-frequency conductivity to obtain a low-frequency conductivity tensor. Studies have shown that volume fraction plays a major role in the decomposition process. Therefore, using diffusion microstructure imaging to extract accurate intracellular and extracellular volume fractions is a difficulty in the CTI field. It also determines the accuracy of the tTIS brain conductivity model, thereby affecting the accuracy of subsequent target planning and ultimately affecting the effectiveness of tTIS treatment.

[0006] In addition to conductivity tensor imaging, magnetic resonance electrical impedance imaging is a method of injecting current into the target body, using magnetic resonance to measure the phase disturbance caused by the injected current, and ultimately reconstructing the low-frequency conductivity tensor. However, this method requires multiple additional external devices, including magnetic resonance-compatible transcranial electrical stimulation devices, and requires the development of special sequences to synchronize the magnetic resonance and electrical stimulation devices. This greatly increases the cost of use and operation time, and the imaging signal-to-noise ratio is low, and image reconstruction is an ill-posed inverse problem, so its clinical practicality is very limited. Compared with magnetic resonance electrical impedance imaging, conductivity tensor imaging does not require any external additional equipment. It only requires a commercial magnetic resonance sequence to collect all the information required for imaging, and its clinical practicality is very high.

[0007] Traditional CTI methods suffer from significant errors in measuring cell volume fraction, particularly in gray matter. Specifically, traditional methods employ biophysical models—stick, airship, and sphere models—to extract information about intraneurite, extraneurite, and free water, respectively. Intraneurite includes highly anisotropic dendrites and axons, while extraneurite includes highly isotropic neurons, glial cells, and the extracellular space. Therefore, the extraneurite space still contains a significant amount of intracellular components, and traditional methods cannot accurately distinguish between cellular and extracellular components in the extraneurite space. Therefore, when constructing models, they only approximate the intraneurite volume fraction as the intracellular volume fraction, resulting in significant model deviations in gray matter. For example, traditional methods yield an intraneurite volume fraction of approximately 0.30 in gray matter. This suggests that nearly 70% of the volume of gray matter is extracellular water, which contradicts both common sense and existing physiological and anatomical evidence. Traditional CTI modeling methods further affect the accuracy of the downstream tTIS brain conductivity model, and ultimately affect the accuracy of tTIS neural regulation.

[0008] In 2021, researchers proposed using a more advanced biophysical diffusion model (Apparent Soma and Neurite Imaging, SANDI) to extract intracellular and extracellular volume fractions. However, this reconstruction method places high demands on the measuring instruments and data, requiring a high gradient magnetic field from the magnetic resonance imaging device and very long imaging times, thus limiting its clinical application. Summary of the Invention

[0009] In view of the above problems, the purpose of the present invention is to provide a method, system, medium, and electronic equipment for modeling low-frequency conductivity tensor of the cranial brain, which can accurately construct a personalized low-frequency conductivity tensor model of the cranial brain, thereby providing more accurate target planning for tTIS.

[0010] In a first aspect, the present invention provides a method for modeling a low-frequency conductivity tensor of the brain, the method comprising the following steps: acquiring a phase image of a fast spin echo sequence of the brain; acquiring a high-frequency conductivity image of the brain based on the phase image; acquiring a diffusion-weighted amplitude image of the brain based on a multi-excitation diffusion-weighted sequence; acquiring diffusion microstructure parameters of the brain using a multi-layer model based on the diffusion-weighted amplitude image; generating a low-frequency conductivity tensor image of the brain based on the high-frequency conductivity image and the diffusion microstructure parameters; generating a digital skull model based on a magnetization-prepared fast gradient echo sequence of the brain; and generating a low-frequency conductivity tensor model of the brain based on the low-frequency conductivity tensor image and the digital skull model.

[0011] In an implementation of the first aspect, acquiring a high-frequency conductivity image of the brain based on the phase image includes:

[0012] Based on the phase image, the phase convection-diffusion reaction equation is used. Acquire the high-frequency conductivity image Where ρ represents the resistivity, Indicates the transmit and receive phase, represents the Hamiltonian operator, ω0 represents the precession angular frequency of the magnetic resonance; μ0 represents the magnetic permeability constant in vacuum, and c represents the regularization parameter.

[0013] In an implementation of the first aspect, acquiring the diffusion microstructure parameters of the brain using a multi-layer model based on the diffusion-weighted amplitude image includes the following steps:

[0014] The diffusion weighted amplitude image is fitted using a multi-layer model to obtain the diffusion microstructure parameters, which include intracellular diffusion rate, intracellular volume fraction, extracellular diffusion rate and extracellular diffusion tensor.

[0015] In an implementation of the first aspect, fitting the diffusion weighted amplitude image using a multi-layer model to obtain the diffusion microstructure parameters includes the following steps:

[0016] Generate simulation labels where f s represents the volume fraction of the stick interlayer, D s represents the stick interlayer diffusion rate, and represent the airship parallel and vertical diffusivities, respectively, f ext represents the extracellular volume fraction, D ext represents the extracellular diffusion rate, p l is the l-th order rotation invariant;

[0017] generating a simulation signal based on the simulation tag;

[0018] Using a polynomial regression model to fit the mapping space between the simulation label and the simulation signal;

[0019] The measured diffusion weighted amplitude data is input into the mapping space to obtain the estimated values ​​of the corresponding diffusion microstructure parameters.

[0020] In an implementation of the first aspect, generating a low-frequency conductivity tensor image of the brain based on the high-frequency conductivity image and the diffusion microstructure parameter includes:

[0021] according to Generate low-frequency conductivity tensor image σ L , where σH represents the high-frequency conductivity image, f ext represents the extracellular volume fraction, D int represents the intracellular diffusion rate, D ext represents the extracellular diffusion rate; β represents the ratio of ion concentration inside and outside the cell.

[0022] In an implementation of the first aspect, generating a digital skull model based on the magnetization-prepared rapid gradient echo sequence of the brain includes the following steps:

[0023] Acquiring a magnetic resonance image of the brain based on a magnetization-prepared rapid gradient echo sequence;

[0024] A digital skull model is generated based on the magnetic resonance image.

[0025] In an implementation of the first aspect, generating a low-frequency conductivity tensor model of the brain based on the low-frequency conductivity tensor image and the digital skull model includes the following steps:

[0026] Segmenting the digital skull model into intracranial and extracranial parts;

[0027] Acquiring an intra-skull conductivity tensor and an extra-skull conductivity tensor based on the low-frequency conductivity tensor image;

[0028] A low-frequency conductivity tensor model of the brain is generated based on the intra-skull conductivity tensor and the extra-skull conductivity tensor.

[0029] In a second aspect, the present invention provides a cranial low-frequency conductivity tensor modeling system, the system comprising an acquisition module, a first acquisition module, a second acquisition module, a third acquisition module, a first generation module, a second generation module, and a third generation module;

[0030] The acquisition module is used to obtain a phase image of a fast spin echo sequence of the brain;

[0031] The first acquisition module is used to acquire a high-frequency conductivity image of the brain based on the phase image;

[0032] The second acquisition module is used to acquire the diffusion-weighted amplitude image of the brain based on a multi-excitation diffusion-weighted sequence;

[0033] The third acquisition module is used to acquire the diffusion microstructure parameters of the brain using a multi-layer model based on the diffusion weighted amplitude image;

[0034] The first generating module is used to generate a low-frequency conductivity tensor image of the brain based on the high-frequency conductivity image and the diffusion microstructure parameter;

[0035] The second generating module is used to generate a digital skull model based on the magnetization prepared rapid gradient echo sequence of the skull;

[0036] The third generating module is used to generate a low-frequency conductivity tensor model of the brain based on the low-frequency conductivity tensor image and the digital skull model.

[0037] In a third aspect, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for modeling the low-frequency conductivity tensor of the brain.

[0038] In a fourth aspect, the present invention provides an electronic device, comprising: a processor and a memory;

[0039] The memory is used to store computer programs;

[0040] The processor is used to execute the computer program stored in the memory, so that the electronic device executes the above-mentioned brain low-frequency conductivity tensor modeling method.

[0041] As described above, the method, system, medium, and electronic device for modeling low-frequency conductivity tensors of the brain of the present invention have the following beneficial effects:

[0042] (1) Ability to accurately construct a personalized low-frequency conductivity tensor model of the brain, thereby providing more accurate target planning for tTIS;

[0043] (2) Accurately estimate the extracellular volume fraction based on a data-driven polynomial regression model, thereby obtaining more accurate low-frequency conductivity tensor values;

[0044] (3) The problem of low modeling accuracy of the traditional CTI method was solved. By constructing a low-frequency conductivity tensor model of the brain, it facilitated the subsequent neural regulation modeling and target planning of tTIS. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Shown is a flow chart of a method for modeling low-frequency conductivity tensor of the brain according to one embodiment of the present invention;

[0046] Figure 2 Shown is a schematic diagram of the structure of a method for modeling low-frequency conductivity tensor of the brain according to one embodiment of the present invention;

[0047] Figure 3 Shown is a schematic diagram comparing the prior art and the brain low-frequency conductivity tensor modeling method of the present invention in one embodiment;

[0048] Figure 4 Shown is a schematic structural diagram of a cranial low-frequency conductivity tensor modeling system according to one embodiment of the present invention;

[0049] Figure 5 FIG. 1 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0051] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0052] The technical solutions in the embodiments of the present invention will be described in detail below with reference to the accompanying drawings in the embodiments of the present invention.

[0053] like Figure 1 and Figure 2 As shown, in one embodiment, the method for modeling low-frequency conductivity tensor of the brain of the present invention includes steps S1 to S7.

[0054] Step S1: Acquire a phase image of a fast spin echo sequence of the brain.

[0055] Specifically, a fast spin echo sequence begins with a 90° excitation pulse, followed by multiple 180° pulses, generating multiple echo signals that are placed in the same k-space to generate an image. For images with the same spatial resolution, its imaging speed is significantly faster than that of a spin echo, and the image weight is determined by the echo time (TE) value of the central region of k-space. In the present invention, phase images of the brain are acquired using a fast spin echo sequence.

[0056] Step S2: Acquire a high-frequency conductivity image of the brain based on the phase image.

[0057] Specifically, according to Maxwell's equations, we can get:

[0058]

[0059] Where E(r) is the electric field, J(r) is the current density, H(r) is the magnetic field intensity, i is the imaginary unit, μ0 is the magnetic permeability of free space, κ is the node characteristic, ω is the magnetic resonance angular frequency, is the Hamilton operator. Current density J = κE = (σ + iω∈)E is determined by the conduction current J c =σE (generated by the movement of mobile carriers) and displacement current J d :=iω∈E (produced by the polarization of immobile charges). Taking the curl operator we get:

[0060]

[0061] because We can get:

[0062]

[0063] According to the magnetic induction intensity B = μH, we can get the equation expressed by B:

[0064]

[0065] Since the k-space data measured by magnetic resonance is not B=(B x , B y , B z ), but B1 field, so B needs to be converted into B1 field:

[0066]

[0067] Without tedious derivation, we can use the formula The phase-based convection-diffusion MREPT equation is obtained:

[0068]

[0069] Where ρ = 1 / σ H is the resistivity (the inverse of conductivity), is the phase of transmission and reception, that is, the phase of the magnetic resonance image. With amplitude and phase plural With amplitude and phase in Is the transmit and receive phase. The coefficients of the convection-diffusion equation are given by the phase Obtain the equation The solution is the high-frequency conductivity σ H The reciprocal of . is the Hamiltonian operator, which represents the gradient of space; ω0 represents the precession angular frequency of magnetic resonance; μ0 represents the magnetic permeability constant in vacuum. The equation in the diagonal position contains many zero elements, which leads to oscillation of the solution. Therefore, diffusion terms are usually added to form the phase convection-diffusion reaction equation

[0070] Solving the phase convection-diffusion reaction equation, the solution obtained is the high-frequency conductivity σ H The reciprocal of . That is, the high-frequency conductivity image Where c represents the regularization parameter.

[0071] According to the divergence theorem equation Can be converted to The finite difference equation for this equation is as follows:

[0072]

[0073] Finally, it can be organized into the form of Aρ=b, where A is a sparse matrix of N*M*L×N*M*L, and b is a vector of 2ω0μ0.

[0074] Step S3: Acquire a diffusion-weighted amplitude image of the brain based on a multi-excitation diffusion-weighted sequence.

[0075] Specifically, the Readout Segmentation of Long Variable Echo-trains (RESOLVE) sequence is a multi-shot diffusion-weighted sequence that generates diffusion-weighted amplitude images by applying diffusion-weighted weighting to water molecules within the brain. b, the diffusion weighting factor, represents the degree of diffusion weighting. A larger b value indicates a greater water molecule diffusion weighting and a smaller acquired signal.

[0076] Step S4: Based on the diffusion-weighted amplitude image, a multi-layer model is used to obtain the diffusion microstructure parameters of the brain.

[0077] Specifically, the multi-layer model is a set of mathematical formulas, which are given based on the characteristics of diffusion. Among them, diffusion can adopt a stick model, an airship model, and a sphere model. The stick model indicates that water molecules can only diffuse along the direction of the stick, showing high anisotropy. The airship model indicates that they can diffuse in all directions, showing low anisotropy. The sphere model indicates that they can diffuse in all directions, showing isotropy. The volume fractions of the stick model, airship model, and sphere model are weighted to form the kernel function of the multi-layer model. When the multi-layer model is used for fitting, this set of mathematical formulas is used to infinitely approximate the data. When the minimum error is reached, the parameters of the mathematical formula, that is, the volume fraction and diffusion rate, can be obtained.

[0078] In the present invention, a multi-layer model is used to fit the diffusion weighted amplitude image to obtain the diffusion microstructure parameters. The diffusion microstructure parameters include the intracellular diffusion rate D int , extracellular volume fraction f ext , extracellular diffusion rate D ext and the extracellular diffusion tensor D ext .

[0079] The nerve fiber bundle is equivalent to a kernel function, and then the diffusion signal of the voxel is equal to the convolution of the kernel function and the fiber direction distribution function (fODF). Specifically, in the field of diffusion magnetic resonance, when the diffusion time is long enough, the diffusion signal can be written as the convolution of the kernel function and the fiber direction distribution function:

[0080]

[0081] in, is the fiber direction distribution function, is the kernel function, is the diffusion-weighted amplitude signal, S0 is the amplitude signal without diffusion weighting, is the direction of the fiber, is the direction of the magnetic resonance pulse diffusion gradient, represents a two-dimensional sphere, and b is the diffusion weighting factor. The kernel function formula is:

[0082]

[0083] in, yes exist The projection of the direction, f s is the volume fraction of the stick interlayer, D s is the stick interlayer diffusion rate, and f is the airship parallel and vertical diffusivity, ext is the extracellular volume fraction, D ext is the extracellular diffusion rate.

[0084] The direction of the fibers is determined by The fiber direction distribution function is described in the form of decomposition on the spherical harmonic (SH) basis:

[0085]

[0086] Among them, L max is the order of spherical harmonic decomposition, when l max = 2, its unknown parameters are the same as the 3×3 symmetric tensor. m represents the degree of spherical harmonics, p lm is the spherical harmonic coefficient, Y lm is the spherical harmonic basis function.

[0087] Spherical harmonics are the expression of Fourier basis in spherical coordinates. The convolution in the spatial domain can be written in the form of a product under the spherical harmonics:

[0088]

[0089] Among them, S lm 、p lm and are the coefficients of the signal, fiber direction distribution function and kernel function under the spherical harmonic basis. In order to reduce the fitting parameters, the rotation invariant is introduced:

[0090]

[0091] equation can be converted to: l=0, 2, ....

[0092] In one embodiment, fitting the diffusion weighted amplitude image using a multi-layer model to obtain the diffusion microstructure parameters includes the following steps:

[0093] (a) Generate simulation labels where f s represents the volume fraction of the stick interlayer, D s represents the stick interlayer diffusion rate, and represent the airship parallel and vertical diffusivities, respectively, f ext represents the extracellular volume fraction, D ext represents the extracellular diffusion rate, p l is the l-th order rotation invariant.

[0094] Here, the scalar parameter is first estimated and rotation-independent basis p l The present invention uses a data-driven machine learning method to estimate the scalar parameter x and the rotation-independent basis p respectively. l .

[0095] (b) generating a simulated signal based on the simulated tag.

[0096] The simulation label is introduced into the convolution formula of the fiber direction distribution function and the kernel function to generate a simulation signal, thereby obtaining a data set of the simulation label and the simulation signal.

[0097] (c) Using a polynomial regression model to fit the mapping space between the simulated label and the simulated signal.

[0098] (d) Inputting the measured diffusion weighted amplitude data into the mapping space to obtain an estimated value of the corresponding diffusion microstructure parameter as the diffusion microstructure parameter.

[0099] The measured diffusion weighted amplitude data is input into the mapping space to obtain the estimated value of the diffusion microstructure parameter corresponding to the measured diffusion weighted amplitude data.

[0100] In one embodiment, the formula of the polynomial regression model is:

[0101]

[0102] in, is the estimated diffusion microstructure parameter, where is the diffusion-weighted amplitude data with noise, W is the order of the polynomial, are the polynomial coefficients of the training. The training labels are generated by simulation, and their range is f S =[0.02,0.98],D s =[1,3]μm 2 / ms, f ext =[0,1],D ext =[0.4,3]μm 2 / ms, p2 = [0, 0.99], p4 = [0, 0.99], the total number of samples is 1e5, and the generated data label x is brought into l=0,2,...,we can obtain the simulation signal set of all label parameters x{S l (b)}, there is a mapping space between the simulation label and the simulation signal, and then the polynomial regression model is used to fit the mapping space to obtain the polynomial regression coefficient representing the mapping space For the estimation of diffusion magnetic resonance parameters, a third-order polynomial regression is sufficient, and higher-order polynomials will result in overfitting. and the polynomial regression model coefficients By multiplying, the diffusion microstructure parameters can be estimated This includes the accurate extracellular volume fraction f ext and the extracellular diffusion rate D ext .

[0103] like Figure 3 As shown, compared with the linear square method in the prior art to obtain the diffusion microstructure parameters, the core of this method is to ensure that the extracellular diffusion rate D of the spherical layer must be calculated during the training phase. ext Add to the simulation tag instead of D ext The value is fixed at 3μm 2 / ms. It should be noted that in the prior art, the stick layer is the space inside the neurite, the airship layer and the ball layer are the space outside the neurite, and the diffusion rate of the ball layer is fixed to free diffusion, so only the information of cerebrospinal fluid can be extracted. The present invention creatively proposes to add the diffusion rate of the ball layer to the simulation label, and the range of values ​​is between non-diffusion and free diffusion, so that the information of isotropic diffusion rate in any space can be adaptively extracted, and the corresponding volume fraction can also be extracted, and the volume fraction of the ball layer is closer to the real extracellular volume fraction. Therefore, it is necessary to redivide the layers, and the stick layer and the airship layer are divided into the intracellular space, and the ball layer is the extracellular space.

[0104] Step S5: generating a low-frequency conductivity tensor image of the brain based on the high-frequency conductivity image and the diffusion microstructure parameters.

[0105] Specifically, according to Generate low-frequency conductivity tensor image σ L , where σ H represents the high-frequency conductivity image, f ext represents the extracellular volume fraction, D int represents the intracellular diffusion rate, D ext represents the extracellular diffusion rate; β represents the ratio of ion concentration inside and outside the cell. For the human brain, the empirical value is 0.41.

[0106] The low-frequency conductivity tensor decomposition formula is as follows:

[0107]

[0108] in, is the average extracellular diffusion rate, D ext is the extracellular diffusion tensor; is the apparent extracellular ion concentration; is the low-frequency conductivity tensor within the skull.

[0109] The water diffusion tensor can be written as the following positive definite matrix:

[0110]

[0111] Among them, S D is the eigenvalue decomposition matrix, is the transpose of the eigenvalue decomposition matrix. The signal intensity of diffusion magnetic resonance can be written as The water diffusion tensor D can be obtained by fitting the diffusion tensor DTI. Where b is the diffusion weighting factor, is the direction of the magnetic resonance pulse diffusion gradient, ρ0 is the signal amplitude without diffusion weighting, ρ Dis the diffusion-weighted signal amplitude. Assuming that the extracellular diffusion tensor and the water diffusion tensor have the same eigenvalue, then D = (1-f w )D int +f w D ext Assuming the same microstructure environment inside and outside the cell, the extracellular diffusion tensor is in, for peace Normalized tensors with the same orientation.

[0112] Step S6: Generate a digital skull model based on the magnetization-prepared rapid gradient echo sequence of the brain.

[0113] Specifically, a magnetic resonance image of the brain is first acquired using a magnetization-prepared rapid gradient echo sequence; then, a digital skull model is generated based on the magnetic resonance image. Generating the digital skull model typically requires segmenting the acquired magnetic resonance image of the brain. Various algorithms are available for segmentation, such as the unified segmentation algorithm used in the SPM12 software, which segments the brain into multiple tissues. A surface is then generated based on the segmented image, and finally, a digital skull model is generated based on the surface.

[0114] Step S7: generating a low-frequency conductivity tensor model of the brain based on the low-frequency conductivity tensor image and the digital skull model.

[0115] Specifically, the digital skull model is segmented into the skull and the skull using traditional segmentation technology. The conductivity tensor inside the skull has been solved above. and the extracranial conductivity tensor Based on the conductivity tensor in the skull and the extracranial electrical conductivity tensor Finally, the brain conductivity tensor C including the inside and outside of the brain can be obtained.

[0116] In addition, when two or more pairs of electrodes are placed on the scalp surface and current is injected into the subject using the tTIS device, according to electrostatic theory, it can be obtained:

[0117]

[0118] Among them, Ω represents the entire brain area, represents the boundary of the brain, ε represents the set of all injection electrodes, g(r) is the injection current, and u(r) is the electric potential.

[0119] It should be noted that SANDI can also estimate the accurate extracellular volume fraction, but the method has high requirements for magnetic resonance equipment and the imaging time is long. To use the SANDI model for imaging, b = 0 / 500 / 1000 / 2000 / 3000 / 4000 / 6000 is required, and the total number of directions is about 200. The method proposed in the present invention only requires b = 0 / 1000 / 2000, and the total number of directions is 34, which is nearly 6 times less data, so the imaging time can be reduced by 6 times. Currently, magnetic resonance electrical impedance imaging can also reconstruct low-frequency conductivity tensors, but this method requires multiple additional external devices, including magnetic resonance-compatible transcranial electrical stimulation devices, and requires the development of special sequences to synchronize magnetic resonance and electrical stimulation devices, which greatly increases the cost and operation time. In addition, the imaging signal-to-noise ratio is low, and image reconstruction is an ill-posed inverse problem, so its clinical practicality is very limited.

[0120] The protection scope of the method for modeling the low-frequency conductivity tensor of the brain described in the embodiment of the present invention is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing, or replacing steps in the existing technology based on the principles of the present invention are included in the protection scope of the present invention.

[0121] An embodiment of the present invention also provides a cranial and cerebral low-frequency conductivity tensor modeling system, which can implement the cranial and cerebral low-frequency conductivity tensor modeling method described in the present invention. However, the implementation device of the cranial and cerebral low-frequency conductivity tensor modeling system described in the present invention includes but is not limited to the structure of the cranial and cerebral low-frequency conductivity tensor modeling system listed in this embodiment. All structural deformations and replacements of the existing technology made according to the principles of the present invention are included in the protection scope of the present invention.

[0122] like Figure 4 As shown, in one embodiment, the low-frequency conductivity tensor modeling system of the brain proposed by the present invention includes an acquisition module 41, a first acquisition module 42, a second acquisition module 43, a third acquisition module 44, a first generation module 45, a second generation module 46 and a third generation module 47.

[0123] The acquisition module 41 is used to acquire a phase image of a fast spin echo sequence of the brain.

[0124] The first acquisition module 42 is connected to the acquisition module 41 and is configured to acquire a high-frequency conductivity image of the brain based on the phase image.

[0125] The second acquisition module 43 is connected to the acquisition module 41 and is configured to acquire a diffusion-weighted amplitude image of the brain based on a multi-excitation diffusion-weighted sequence.

[0126] The third acquisition module 44 is connected to the second acquisition module 43 and is configured to acquire the diffusion microstructure parameters of the brain based on the diffusion weighted amplitude image using a multi-layer model.

[0127] The first generating module 45 is connected to the first acquiring module 42 and the third acquiring module 44 , and is configured to generate a low-frequency conductivity tensor image of the brain based on the high-frequency conductivity image and the diffusion microstructure parameters.

[0128] The second generating module 46 is configured to generate a digital skull model based on the magnetization prepared rapid gradient echo sequence of the skull.

[0129] The third generating module 47 is connected to the first generating module 45 and the second generating module 46 and is used to generate a low-frequency conductivity tensor model of the brain based on the low-frequency conductivity tensor image and the digital skull model.

[0130] Among them, the structures and principles of the acquisition module 41, the first acquisition module 42, the second acquisition module 43, the third acquisition module 44, the first generation module 45, the second generation module 46 and the third generation module 47 correspond one to one with the above-mentioned low-frequency conductivity tensor modeling method of the brain, so they will not be repeated here.

[0131] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices or methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of modules / units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules or units, which can be electrical, mechanical or other forms.

[0132] Modules / units described as separate components may or may not be physically separate, and components displayed as modules / units may or may not be physical modules, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules / units may be selected based on actual needs to achieve the objectives of the embodiments of the present invention. For example, the functional modules / units in various embodiments of the present invention may be integrated into a single processing module, each module / unit may exist physically separately, or two or more modules / units may be integrated into a single module / unit.

[0133] Those skilled in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the composition and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0134] The embodiment of the present invention also provides a computer-readable storage medium. Those skilled in the art will understand that all or part of the steps in the method for modeling the low-frequency conductivity tensor of the cranial brain of the above embodiment can be completed by instructing a processor through a program, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state drive, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0135] An embodiment of the present invention further provides an electronic device comprising a processor and a memory.

[0136] The memory is used to store computer programs.

[0137] The memory includes various media that can store program codes, such as ROM, RAM, magnetic disk, USB flash drive, memory card or optical disk.

[0138] The processor is connected to the memory and is used to execute the computer program stored in the memory, so that the electronic device executes the above-mentioned cranial low-frequency conductivity tensor modeling method.

[0139] Preferably, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0140] like Figure 5 As shown, the electronic device of the present invention is in the form of a general-purpose computing device. Components of the electronic device may include, but are not limited to: one or more processors or processing units 51, a memory 52, and a bus 53 connecting different system components (including the memory 52 and the processing unit 51).

[0141] Bus 53 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. Examples of these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0142] Electronic devices typically include a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device, including volatile and non-volatile media, removable and non-removable media.

[0143] The memory 52 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 521 and / or cache memory 522. The electronic device may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 523 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 Not shown, often called a "hard drive"). Although Figure 5Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 53 via one or more data medium interfaces. Memory 52 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of various embodiments of the present invention.

[0144] A program / utility 524 having a set (at least one) of program modules 5241 may be stored, for example, in memory 52. ​​Such program modules 5241 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may include an implementation of a network environment. Program modules 5241 generally implement the functions and / or methods of the embodiments described herein.

[0145] The electronic device may also communicate with one or more external devices (e.g., keyboards, pointing devices, displays, etc.), one or more devices that enable a user to interact with the electronic device, and / or any device that enables the electronic device to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via input / output (I / O) interface 54. Furthermore, the electronic device may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 55. Figure 5 As shown, the network adapter 55 communicates with other modules of the electronic device via the bus 53. It should be understood that, although not shown in the figures, other hardware and / or software modules may be used in conjunction with the electronic device, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0146] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical principles disclosed herein are intended to be covered by the claims of the present invention.

Claims

1. A method for modeling low-frequency conductivity tensors of the brain, characterized by: The method comprises the following steps: Acquire phase images of the brain using a fast spin echo sequence; acquiring a high-frequency conductivity image of the brain based on the phase image; Acquiring a diffusion-weighted amplitude image of the brain based on a multi-excitation diffusion-weighted sequence; Based on the diffusion-weighted amplitude image, a multi-layer model is used to obtain diffusion microstructure parameters of the brain; generating a low-frequency conductivity tensor image of the brain based on the high-frequency conductivity image and the diffusion microstructure parameter; generating a digital skull model based on a magnetization-prepared rapid gradient echo sequence of the brain; generating a cranial low-frequency conductivity tensor model based on the low-frequency conductivity tensor image and the digital skull model; Acquiring the diffusion microstructure parameters of the brain using a multi-layer model based on the diffusion-weighted amplitude image includes the following steps: Using a multi-compartment model, the diffusion-weighted amplitude image is fitted to obtain the diffusion microstructure parameters, wherein the diffusion microstructure parameters include intracellular diffusion rate, intracellular volume fraction, extracellular diffusion rate, and extracellular diffusion tensor; Using a multi-layer model to fit the diffusion weighted amplitude image to obtain the diffusion microstructure parameters includes the following steps: Generate Simulation Label where f s represents the volume fraction of the stick interlayer, D s represents the stick interlayer diffusion rate, and are the airship parallel and vertical diffusivities, respectively, and f ext represents the extracellular volume fraction, D ext represents the extracellular diffusion rate, p l is the l-th order rotation invariant; generating a simulation signal based on the simulation tag; Using a polynomial regression model to fit the mapping space between the simulation label and the simulation signal; The measured diffusion weighted amplitude data is input into the mapping space to obtain the estimated values ​​of the corresponding diffusion microstructure parameters.

2. The method for modeling low-frequency conductivity tensor of the brain according to claim 1, characterized in that: Acquiring a high-frequency conductivity image of the brain based on the phase image includes: Based on the phase image, the phase convection-diffusion reaction equation is used. Acquire the high-frequency conductivity image Where ρ represents the resistivity, Indicates the transmit and receive phase, represents the Hamiltonian operator, ω0 represents the precession angular frequency of the magnetic resonance; μ0 represents the magnetic permeability constant in vacuum; and c represents the regularization parameter.

3. The method for modeling low-frequency conductivity tensor of the cranial brain according to claim 1, characterized in that: Generating the low-frequency conductivity tensor image of the brain based on the high-frequency conductivity image and the diffusion microstructure parameter includes: according to Generate low-frequency conductivity tensor image σ L , where σ H represents the high-frequency conductivity image, f ext represents the extracellular volume fraction, D int represents the intracellular diffusion rate, D ext represents the extracellular diffusion rate; β represents the ratio of ion concentration inside and outside the cell.

4. The method for modeling low-frequency conductivity tensor of the brain according to claim 1, characterized in that: Generating a digital skull model based on the magnetization-prepared rapid gradient echo sequence of the brain comprises the following steps: Acquiring a magnetic resonance image of the brain based on a magnetization-prepared rapid gradient echo sequence; A digital skull model is generated based on the magnetic resonance image.

5. The method for modeling low-frequency conductivity tensor of the brain according to claim 1, characterized in that: Generating a low-frequency conductivity tensor model of the brain based on the low-frequency conductivity tensor image and the digital skull model comprises the following steps: Segmenting the digital skull model into intracranial and extracranial parts; Acquiring an intra-skull conductivity tensor and an extra-skull conductivity tensor based on the low-frequency conductivity tensor image; A low-frequency conductivity tensor model of the brain is generated based on the intra-skull conductivity tensor and the extra-skull conductivity tensor.

6. A cranial low-frequency conductivity tensor modeling system, characterized by: The system includes an acquisition module, a first acquisition module, a second acquisition module, a third acquisition module, a first generation module, a second generation module and a third generation module; The acquisition module is used to obtain a phase image of a fast spin echo sequence of the brain; The first acquisition module is used to acquire a high-frequency conductivity image of the brain based on the phase image; The second acquisition module is used to acquire the diffusion-weighted amplitude image of the brain based on a multi-excitation diffusion-weighted sequence; The third acquisition module is used to acquire the diffusion microstructure parameters of the brain using a multi-layer model based on the diffusion weighted amplitude image; The first generating module is used to generate a low-frequency conductivity tensor image of the brain based on the high-frequency conductivity image and the diffusion microstructure parameter; The second generating module is used to generate a digital skull model based on the magnetization prepared rapid gradient echo sequence of the skull; The third generating module is used to generate a low-frequency conductivity tensor model of the brain based on the low-frequency conductivity tensor image and the digital skull model; Acquiring the diffusion microstructure parameters of the brain using a multi-layer model based on the diffusion-weighted amplitude image includes the following steps: Using a multi-compartment model, the diffusion-weighted amplitude image is fitted to obtain the diffusion microstructure parameters, wherein the diffusion microstructure parameters include intracellular diffusion rate, intracellular volume fraction, extracellular diffusion rate, and extracellular diffusion tensor; Using a multi-layer model to fit the diffusion weighted amplitude image to obtain the diffusion microstructure parameters includes the following steps: Generate Simulation Label where f s represents the volume fraction of the stick interlayer, D s represents the stick interlayer diffusion rate, and are the airship parallel and vertical diffusivities, respectively, and f ext represents the extracellular volume fraction, D ext represents the extracellular diffusion rate, p l is the l-th order rotation invariant; generating a simulation signal based on the simulation tag; Using a polynomial regression model to fit the mapping space between the simulation label and the simulation signal; The measured diffusion weighted amplitude data is input into the mapping space to obtain the estimated values ​​of the corresponding diffusion microstructure parameters.

7. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for modeling the low-frequency conductivity tensor of the cranial brain according to any one of claims 1 to 5 is implemented.

8. An electronic device, characterized in that: include: processor and memory; The memory is used to store computer programs; The processor is configured to execute the computer program stored in the memory, so that the electronic device executes the cranial low-frequency conductivity tensor modeling method according to any one of claims 1 to 5.

Images