Anatomical Connection Prediction Method Based on Functional Connectivity and Kinetic Model

Abstract

The present invention discloses an anatomical connection prediction method based on functional connectivity and kinetic models. First, a neural kinetic model is constructed, with a random matrix as the initial input. Combining functional connectivity data and the neural kinetic model, a coupling matrix that better conforms to neural kinetic characteristics is generated through iterative optimization of the random matrix, and this coupling matrix is used to predict anatomical connections, replacing anatomical connection data to construct a high-precision brain model. The method of the present invention realizes non-invasive anatomical connection prediction by iteratively optimizing the initial random matrix, enabling the generated coupling matrix to gradually approach the true anatomical connection, significantly reducing the dependence on invasive structural connection data. Compared with existing neural kinetic modeling methods that rely on dMRI data, the DMF model used has a higher degree of fit in terms of structure and function, stronger flexibility and adaptability, helps to reveal the deep connection between functional connectivity and anatomical connection, and provides more efficient technical support for brain network research.

Description

Technical Field

[0001] The present invention belongs to the cross - technical field of neuroscience and computer science, and particularly relates to a method for predicting anatomical connections based on functional connectivity and kinetic models. Background Art

[0002] In neuroscience research, understanding the structure and function of the brain network has always been a core challenge, and anatomical connectivity and functional connectivity are two key concepts. Anatomical connectivity represents the physical connections between neurons in the brain and forms the basis of the brain network. In the prior art, anatomical connections are obtained through invasive means, such as using tracer injection techniques in non - human primates (such as macaques). Such techniques can accurately track the connections between brain regions and provide ground truth data on brain connections. However, due to the highly invasive nature of these methods, they cannot be widely used in human studies. Therefore, how to effectively obtain anatomical connection data has always been a difficult problem in research.

[0003] To solve this problem, researchers have tried to indirectly infer anatomical connections through non - invasive means such as diffusion tensor imaging (dMRI) and functional magnetic resonance imaging (fMRI). Diffusion MRI can infer the orientation of nerve fibers by tracking the diffusion path of water molecules in brain tissue, thereby constructing a structural connectivity map of the brain. However, the resolution and accuracy of this method are limited by technology, especially in resolving complex crossing fibers. In addition, diffusion MRI requires a large amount of data and computational resources, making the research process cumbersome and time - consuming. In contrast, functional connectivity measures the interaction relationship between different brain regions by analyzing their co - activity. Resting - state functional magnetic resonance imaging (rs - fMRI) is a commonly used method that can capture the spontaneous functional activities of the brain in the resting state, thereby inferring the functional connectivity between brain regions. This method has the advantages of non - invasiveness, easy acquisition, and high efficiency, and has been widely used in the research of brain networks. Research shows that resting - state functional connectivity can effectively identify direct anatomical connections in the brain and reflect the characteristics of anatomical connections. However, since functional connectivity not only reflects direct anatomical connections but may also be affected by polysynaptic or other complex network factors, inferring anatomical connections based on functional connectivity becomes more complex and challenging.

[0004] The neural dynamics model aims to combine anatomy and dynamics to explain the formation of resting-state networks. Theoretically, anatomical connections can be inferred through functional connectivity and computational models. However, existing neural dynamics models often construct a structural connectivity matrix based on dMRI and then simulate the dynamic behavior of the brain network through these structural connections. Due to the low imaging accuracy of dMRI and the limited structural connection information obtained, the model often fails to achieve the expected accuracy when analyzing complex networks and simulating real brain functional networks. To solve this problem, researchers have begun to explore methods to optimize anatomical connections using functional connectivity data. Recent research has proposed a method for iteratively optimizing structural connections based on functional connectivity. This method uses functional connectivity data to iteratively optimize the initial dMRI structural connectivity matrix to make it more consistent with the characteristics of functional connectivity, thereby improving the accuracy and predictive ability of the neural dynamics model. Although this method has improved the accuracy of the model to a certain extent, it still relies on the initial dMRI connection data, limiting its application in cases where there is no anatomical or scarce anatomical data. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a method for predicting anatomical connections based on functional connectivity and a dynamics model, which solves the problem of lack of high-precision anatomical connection data. By constructing a connection matrix that conforms to dynamic coupling, it effectively predicts anatomical connections, thereby realizing a high-precision brain model.

[0006] The technical solution adopted by the present invention is as follows: A method for predicting anatomical connections based on functional connectivity and a dynamics model, the specific steps are as follows:

[0007] S1. Construct a neural dynamics model, including: an input module, a DMF module, and an output module;

[0008] The neural dynamics model adopts a dynamic mean field (DMF) model, and the functional connectivity matrix obtained by simulation is used as the output.

[0009] Among them, the input module receives structural connection data as the initial input of the model, representing the connection strength between different brain regions; the DMF module adopts the DMF model, describes the electrical activities of different brain regions through neural dynamics equations, and converts electrical signals into blood oxygenation level-dependent (BOLD) signals through a transduction function; the output module generates a simulated functional connectivity matrix, uses the brain region activity signals output by the DMF module, calculates the functional connection strength between each brain region, and finally generates a simulated functional connectivity matrix.

[0010] S2. The input module initializes a random matrix as the initial coupling matrix of the DMF model;

[0011] Initialize a random matrix as the initial coupling matrix C of the model(0) , i.e., the initial input matrix of the DMF model, which will serve as the starting point for optimization.

[0012] Among them, each element in the input matrix represents the connection relationship between specific brain regions. represents the initial structural connection between brain regions n and p.

[0013] S3. Feed the initial input matrix obtained in step S2 into the DMF module, and convert the electrical signal output by the model into a blood oxygenation level-dependent (BOLD) signal.

[0014] The DMF module uses a neurodynamics equation to describe the electrical activity of different brain regions. Then, the DMF model at the whole-brain level is represented by a coupled differential equation system, and the specific expression is as follows:

[0015]

[0016]

[0017] Among them, represents the input current of the excitatory (E) or inhibitory (I) neuron population in brain region n. represents the firing rate of the excitatory (E) or inhibitory (I) neuron population in brain region n. represents the average excitatory (E) or inhibitory (I) synaptic gating variable of brain region n, and t represents time. And the total effective external input I0 = 0.382 nA, which is scaled by the effective coupling constants W E = 1 and W I = 0.7 respectively in the excitatory pool and the inhibitory pool. The local excitatory recurrence ω + = 1.4, G represents the global coupling parameter, C np represents the structural connection matrix between brain regions n and p, and all excitatory synaptic couplings J NMDA = 0.15 nA. represents the current transduction function, which converts the incoming total input current into a firing rate. The gains of the transduction function are g E = 310 nC -1 , g I = 615 nC -1 , and the other parameters are the threshold current The noise factor d E = 0.16, d I = 0.087. And in the transduction function, it is necessary to ensure that the denominator is not zero, that is The kinetic parameter γ = 0.64 1 / 1000, and the factor 1000 represents all contents in milliseconds. The effective time constants are τ E = τNMDA = 100 ms and τI = τGABA = 10 ms. v n represents Gaussian noise with an amplitude of σ = 0.01 nA. J n represents the local feedback inhibition weight of the brain region, and the J of each brain region is adjusted using the Feedback Inhibition Control (FIC) algorithm n to keep the firing rate of the excitatory pool at a low firing rate, i.e.,

[0018] Then, using the generalized hemodynamic model, the mean-field activity simulated in the DMF model is converted into the BOLD signal. Regulated by the self-regulatory feedback mechanism, when the firing rate of neurons increases, it triggers the vasodilation signal x n to increase, and the blood flow f proportional to the vasodilation signal n causes changes in the deoxyhemoglobin content q n and blood volume y n The specific expressions are as follows:

[0019]

[0020] where k represents the signal attenuation rate, e represents the elimination rate dependent on blood flow, ρ represents the resting oxygen extraction fraction, α represents the venous resistance, and τ represents the time constant.

[0021] The BOLD signal is a static non-linear function of the deoxyhemoglobin content q n and blood volume y n and includes the volume-weighted sum of extravascular and intravascular signals. For each region n, the BOLD signal is denoted by B n and the expression is as follows:

[0022]

[0023] where the resting blood volume fraction V0 = 0.02, the dimensionless parameters k1 = 7ρ, k2 = 2, k3 = 2ρ - 0.2.

[0024] S4. Based on the BOLD signal obtained in step S3, the initial simulated functional connectivity matrix is output through the output module;

[0025] The initial simulated functional connectivity matrix FC is generated by calculating the Pearson correlation coefficient between the simulated BOLD signals of different brain regions sim For the BOLD signal time series of brain regions i and j, the calculation expression of the Pearson correlation coefficient r ij is as follows:

[0026]

[0027] Among them, T represents the total time length, and B i，t represents the BOLD signal value of brain region i at time t, and represents the average value of the BOLD signal of brain region i.

[0028] Then, the initial simulated functional connectivity matrix is compared with the real functional connectivity data FC emp to calculate the Pearson correlation coefficient and the mean square error MSE between the two. The calculation expression of MSE is as follows:

[0029]

[0030] Among them, N represents the total number of connections in the matrix.

[0031] Finally, the differences between the simulated functional connectivity and the real functional connectivity data are evaluated using the two metrics of Pearson correlation coefficient and mean square error as the optimization objective function.

[0032] S5. Based on step S4, the coupling matrix is updated using the gradient descent method;

[0033] Using the gradient descent method, the coupling matrix is adjusted using the difference between the simulated and real functional connectivity matrices, so that the simulated functional connectivity gradually approaches the real functional connectivity. The update expression of the coupling matrix is as follows:

[0034]

[0035] Among them, C (m) represents the coupling matrix input in the m-th iteration, ∈ represents the learning rate, which controls the update amplitude of each step, and m represents the m-th iteration.

[0036] Then, the updated C (m+1) is used to re-run the DMF model to generate new BOLD signals, and then a new simulated functional connectivity matrix is obtained by calculating the Pearson correlation coefficient of the new BOLD signals and the difference between the newly simulated functional connectivity and the real functional connectivity data in the current iteration is calculated. This difference is compared with the difference obtained in the previous iteration. If the difference no longer decreases and is close to the preset fitting accuracy, it is considered that the optimization has converged; otherwise, continue the iteration until the preset fitting accuracy is reached.

[0037] S6. Based on step S5, after the iterative optimization is completed, an optimized coupling matrix, that is, the optimal coupling matrix, is generated for anatomical connection prediction;

[0038] After the iterative optimization is completed, a coupling matrix C optimized that conforms to the neurodynamics characteristics is generated as the final input of the DMF model for predicting anatomical connections and replacing the anatomical connection data to construct a brain model with higher accuracy.

[0039] Advantages of the present invention: The method of the present invention first constructs a neural dynamics model. With a random matrix as the initial input, combined with functional connectivity data and the neural dynamics model, a coupling matrix that better conforms to neural dynamics characteristics is generated by iteratively optimizing the random matrix, and this coupling matrix is used to predict anatomical connectivity, replacing anatomical connectivity data to construct a high-precision brain model. The method of the present invention iteratively optimizes the initial random matrix, enabling the generated coupling matrix to gradually approach the true anatomical connectivity, realizing non-invasive prediction of anatomical connectivity, significantly reducing the dependence on invasive structural connectivity data. Compared with existing neural dynamics modeling methods that rely on dMRI data, the DMF model used has a higher degree of fit in terms of structure and function, provides a new direction for the construction of high-precision digital twin brain models, has stronger flexibility and adaptability, helps to reveal the deep connection between functional connectivity and anatomical connectivity, and provides more efficient technical support for brain network research. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flowchart of a method for predicting anatomical connectivity based on functional connectivity and a dynamics model of the present invention.

[0041] Figure 2 It is a structural diagram of the neural dynamics model in an embodiment of the present invention.

[0042] Figure 3 It is a schematic diagram of the result of predicting anatomical connectivity with the optimal coupling matrix in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The method of the present invention will be further described below in conjunction with the drawings and embodiments.

[0044] As Figure 1 shown, a flowchart of a method for predicting anatomical connectivity based on functional connectivity and a dynamics model of the present invention is as follows:

[0045] S1. Construct a neural dynamics model, including: an input module, a DMF module, and an output module;

[0046] As Figure 2 shown, the neural dynamics model adopts a dynamic mean field (DMF) model, and the functional connectivity matrix obtained by simulation is used as the output.

[0047] Among them, the input module receives structural connection data as the initial input of the model, representing the connection strength between different brain regions; the DMF module uses the DMF model to describe the electrical activities of different brain regions through neural dynamics equations, and converts the electrical signals into blood oxygenation level dependent (BOLD) signals through a transduction function. By setting appropriate differential equations, the changes in the membrane potential of each brain region are simulated, thereby reflecting the activation state of neurons. The behavior of each brain region is affected by other brain regions, where the connection strength (or coupling) is a key parameter in the model, determining the degree of influence of one brain region on another. Each brain region is described by a node equation, and the interaction between different brain regions is characterized by the input coupling matrix C; the output module generates a simulated functional connection matrix, uses the brain region activity signals output by the DMF module to calculate the functional connection strength between each brain region, and finally generates a simulated functional connection matrix.

[0048] S2. The input module initializes a random matrix as the initial coupling matrix of the DMF model;

[0049] Initialize a random matrix as the initial coupling matrix C of the model (0) , that is, the initial input matrix of the DMF model, and this matrix will be used as the optimization starting point.

[0050] Among them, each element in the input matrix represents the connection relationship between specific brain regions, representing the initial structural connection between brain regions n and p.

[0051] S3. Send the initial input matrix obtained in step S2 into the DMF module, and convert the electrical signals output by the model into blood oxygenation level dependent BOLD signals;

[0052] In the DMF model, large-scale interconnected excitatory and inhibitory neuron populations are simplified into two mutually coupled aggregates (i.e., pools), and their activities are described by a set of simplified dynamic equations. At the biological level, the inhibitory current (I I ) is mediated by GABA receptors, while the excitatory synaptic current (I E ) is mediated by NMDA receptors. Each brain region can be regarded as a coupled system composed of excitatory and inhibitory neuron populations, and the coupling between nodes n and p in different brain regions only occurs between their excitatory populations. The long-range synaptic connections are mediated by AMPA receptors. In the DMF model, these coupling connections are scaled by the structural connections.

[0053] The DMF module uses neural dynamics equations to describe the electrical activities of different brain regions. Then, the DMF model at the whole-brain level is represented by a coupled differential equation set, and the specific expression is as follows:

[0054]

[0055]

[0056] wherein, represents the input current of the excitatory E or inhibitory I neuron population in brain region n, represents the firing rate of the excitatory E or inhibitory I neuron population in brain region n. represents the average excitatory E or inhibitory I synaptic gating variable of brain region n, and t represents time. And the overall effective external input I0 = 0.382 nA, which is scaled by the effective coupling constants W E = 1 and W I = 0.7 in the excitatory pool and inhibitory pool respectively. The local excitatory recurrence ω + = 1.4, G represents the global coupling parameter, C np represents the structural connection matrix between brain regions n and p, and all excitatory synaptic couplings J NMDA = 0.15 nA. represents the current transduction function, which converts the incoming total input current into a firing rate. The gains of the transduction function are g E = 310 nC -1 , g I = 615 nC -1 , and the other parameters are the threshold current noise factor d E = 0.16, d I = 0.087. And in the transduction function, it is necessary to ensure that the denominator is not zero, that is The kinetic parameter γ = 0.64 1 / 1000 (the factor 1000 represents all contents in milliseconds), and the effective time constants are τ E = τNMDA = 100 ms and τ I = τGABA = 10 ms. v n represents Gaussian noise, and its amplitude is σ = 0.01 nA. In order to better predict the resting-state functional connectivity, each independent node shows the typical noise spontaneous activity observed in electrophysiological experiments and has a low firing rate The feedback inhibition control FIC algorithm is adopted to adjust the local feedback inhibition weight J of each brain region n to keep the firing rate of the excitatory pool around 3 Hz.

[0057] Then, the generalized hemodynamic model is used to convert the mean-field activity simulated in the DMF model into the BOLD signal. Regulated by the self-regulatory feedback mechanism, when the firing rate of neurons rises, a vasodilation signal x is triggeredn Increase the blood flow rate f proportional to the vasodilation signal n Cause a change in the deoxyhemoglobin content q n And the blood volume y n The changes are as follows:

[0058]

[0059] Where k represents the signal attenuation rate, e represents the elimination rate dependent on blood flow, ρ represents the resting oxygen extraction fraction, α represents the venous resistance, and τ represents the time constant.

[0060] The BOLD signal is a static non - linear function of the deoxyhemoglobin content q n And the blood volume y n Including the volume - weighted sum of extra - vascular and intra - vascular signals. For each region n, the BOLD signal is denoted by B n And is expressed as follows:

[0061]

[0062] Where the resting blood volume fraction V0 = 0.02, the dimensionless parameters k1 = 7ρ, k2 = 2, k3 = 2ρ - 0.2.

[0063] S4. Based on the BOLD signal obtained in step S3, output the initial simulated functional connectivity matrix through the output module;

[0064] Generate the initial simulated functional connectivity matrix FC by calculating the Pearson correlation coefficient between the simulated BOLD signals of different brain regions sim For the BOLD signal time series of brain regions i and j, the calculation expression of the Pearson correlation coefficient r ij Is as follows:

[0065]

[0066] Where T represents the total time length, B i,t Represents the BOLD signal value of brain region i at time t, Represents the average value of the BOLD signal of brain region i.

[0067] Then compare the initial simulated functional connectivity matrix with the real functional connectivity data FC emp Calculate the Pearson correlation coefficient and the mean square error (MSE) between the two. The calculation expression of MSE is as follows:

[0068]

[0069] Where N represents the total number of connections in the matrix.

[0070] Finally, the differences between the simulated functional connectivity and the real functional connectivity data are evaluated using two metrics, the Pearson correlation coefficient and the mean squared error, as the optimization objective function.

[0071] S5. Based on step S4, the coupling matrix is updated using the gradient descent method;

[0072] Using the gradient descent method, the coupling matrix is adjusted using the difference between the simulated and real functional connectivity matrices, so that the simulated functional connectivity gradually approaches the real functional connectivity. The update expression of the coupling matrix is as follows:

[0073]

[0074] where C (m) represents the coupling matrix input at the m-th iteration, ∈ represents the learning rate, which controls the update amplitude of each step, and m represents the m-th iteration.

[0075] Then, the updated C (m+1) is used to re-run the DMF model to generate new BOLD signals, and then a new simulated functional connectivity matrix is obtained by calculating the Pearson correlation coefficient of the new BOLD signals And calculate the difference between the newly simulated functional connectivity and the real functional connectivity data in the current iteration, and compare this difference with the difference obtained in the previous iteration. In this embodiment, if the difference no longer decreases (the fluctuation is less than 0.1) and is close to the preset fitting accuracy (the Pearson correlation coefficient is greater than 0.9 and the mean squared error is less than 0.1), it is considered that the optimization has converged; otherwise, continue the iteration until the preset fitting accuracy is reached.

[0076] S6. Based on step S5, after the iterative optimization is completed, an optimized coupling matrix, that is, the optimal coupling matrix, is generated for anatomical connection prediction;

[0077] After the iterative optimization is completed, a coupling matrix C optimized that conforms to the neurodynamics characteristics is generated as the final input of the DMF model for predicting anatomical connections and replacing the anatomical connection data to construct a brain model with higher accuracy.

[0078] In this embodiment, step S7 is also included to evaluate the prediction accuracy, which is specifically as follows:

[0079] Fitting accuracy evaluation: Calculate the difference between the simulated functional connectivity matrix (FC sim ) and the real functional connectivity matrix (FC emp ) output by the optimized final model to evaluate the fitting accuracy, that is, use the mean squared error (MSE) and the Pearson correlation coefficient as the quantification metrics.

[0080] Comparison with the anatomical connectivity matrix: The optimized coupling matrix is compared with the real anatomical connectivity matrix to evaluate its predictive ability for anatomical structures.

[0081] Correlation analysis: Calculate the Pearson correlation coefficient between C optimized and the real anatomical connectivity (AC) to determine to what extent the optimized coupling matrix retains the anatomical connectivity characteristics.

[0082] Binary classification metrics: Use metrics such as the area under the receiver operating characteristic curve (ROC) (AUC), specificity, and sensitivity to evaluate the performance of the optimized coupling matrix in predicting anatomical connectivity. Figure 3 For the result display of predicting anatomical connectivity based on the optimal coupling matrix, where Figure 3 (a) is the receiver operating characteristic curve (ROC), Figure 3 (b) are the specificity and sensitivity curves.

[0083] In summary, the method of the present invention aims to solve the problem of the lack of high-precision anatomical connectivity data. By constructing a connectivity matrix that conforms to dynamic coupling, it can effectively predict anatomical connectivity, thereby realizing a high-precision brain model. The neural dynamics model adopts the dynamic mean field DMF model, and the functional connectivity matrix obtained by simulation is used as the output. The optimization process of the model is based on the difference between the simulated functional connectivity and the real functional connectivity, and continuously adjusts the input random matrix until the output functional connectivity matrix reaches the best fit with the real functional connectivity matrix. The finally generated coupling matrix can be used to more accurately predict anatomical connectivity. Compared with the existing neural dynamics modeling methods that rely on dMRI data, the method of the present invention only requires functional connectivity and is simple to model, enabling the model to achieve a higher degree of fit in terms of structure and function, providing a new direction for the construction of a high-precision digital twin brain model. The method of the present invention has stronger flexibility and adaptability, helps to reveal the deep connection between functional connectivity and anatomical connectivity, and provides more efficient technical support for brain network research.

[0084] Those of ordinary skill in the art will realize that the embodiments described herein are for helping the reader understand the principles of the present invention and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. For those skilled in the art, various changes and modifications can be made to the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the scope of the claims of the present invention.

Claims

1. A method for predicting anatomical connectivity based on functional connectivity and dynamics model, the specific steps are as follows: S1. Construct a neural dynamics model, including: Input module, DMF module, output module; The neural dynamics model adopts a dynamic mean field DMF model, and the functional connection matrix obtained by simulation is used as output; The input module receives structural connection data as the initial input of the model, which represents the connection strength between different brain regions. The DMF module adopts the DMF model to describe the electrical activities of different brain regions through neurodynamic equations, and converts the electrical signals into blood oxygen level-dependent BOLD signals through transduction functions. The output module generates a simulated functional connection matrix, and uses the brain region activity signals output by the DMF module to calculate the functional connection strength between each brain region, and finally generates a simulated functional connection matrix. S2, the input module initializes a random matrix as the initial coupling matrix of the DMF model; Initialize a random matrix as the initial coupling matrix C of the model (0) , which is the initial input matrix of the DMF model, and this matrix will be used as the starting point for optimization; Each element in the input matrix represents the connection relationship between specific brain regions. represents the initial structural connection between brain regions n and p; S3, sending the initial input matrix obtained in step S2 to the DMF module, and converting the electrical signal output by the model into a blood oxygen level dependent BOLD signal; The DMF module uses neurodynamic equations to describe the electrical activities of different brain regions. The DMF model at the whole brain level is represented by a set of coupled differential equations. The specific expression is as follows: in, represents the input current of an excitatory E or inhibitory I neuron population in brain region n, represents the firing rate of the excitatory E or inhibitory I neuron population in brain area n; represents the average excitatory E or inhibitory I synaptic gating variable in brain region n, t represents time; and the overall effective external input I0 = 0.382 nA, which is represented by the effective coupling constant W in the excitatory pool and the inhibitory pool, respectively. E =1 and W I = 0.7 scaling; local excitability recursion ω + =1.4, G represents the global coupling parameter, C np represents the structural connectivity matrix between brain regions n and p, with all excitatory synaptic coupling J NMDA =0.15nA; represents the current transduction function, which converts the incoming total input current into the firing rate; the gains of the transduction function are g E =310nC -1 , g I =615nC -1 , and the other parameters are threshold current Noise factor d E =0.16, d I =0.087; and in the transduction function, it is necessary to ensure that the denominator is not zero, that is, The kinetic parameter γ = 0.641 / 1000, the factor 1000 means that everything is in milliseconds, and the effective time constants are τ E =τNMDA=100ms and τ I =τGABA=10ms; v n represents Gaussian noise, whose amplitude is σ=0.01nA; J n represents the local feedback inhibition weight of the brain region, and the feedback inhibition control FIC algorithm is used to adjust the J of each brain region n , so that the firing rate of the excitation pool is kept at a low firing rate, that is The generalized hemodynamic model is then used to convert the simulated mean field activity in the DMF model into a BOLD signal; regulated by a self-regulatory feedback mechanism, when the firing rate of the neuron When it rises, it triggers vasodilation signal x n Increase, blood flow proportional to the vasodilation signal f n Caused by deoxyhemoglobin content q n and blood volume n The specific expression is as follows: Where k represents the signal attenuation rate, e represents the elimination rate dependent on blood flow, ρ represents the resting oxygen uptake fraction, α represents the venous resistance, and τ represents the time constant; BOLD signal is the content of deoxyhemoglobin q n and blood volume n A static nonlinear function of the volume weighted sum of the extravascular and intravascular signals; for each region n, the BOLD signal is expressed as B n The expression is as follows: Among them, resting blood volume fraction V0 = 0.02, dimensionless parameters k1 = 7ρ, k2 = 2, k3 = 2ρ-0.2; S4, based on the BOLD signal obtained in step S3, outputting the initial simulated functional connectivity matrix through the output module; The initial simulated functional connectivity matrix FC was generated by calculating the Pearson correlation coefficient between the simulated BOLD signals of different brain regions. sim , for the BOLD signal time series of brain regions i and j, the Pearson correlation coefficient r ij The calculation expression is as follows: Among them, T represents the overall time length, B i,t represents the BOLD signal value of brain area i at time t, represents the average value of BOLD signal in brain area i; The initial simulated functional connectivity matrix is ​​then compared with the real functional connectivity data FC emp For comparison, the Pearson correlation coefficient and mean square error MSE are calculated. The MSE calculation expression is as follows: Where N represents the total number of connections in the matrix; Finally, the Pearson correlation coefficient and mean square error were used to evaluate the difference between the simulated functional connectivity and the real functional connectivity data as the objective function for optimization; S5, based on step S4, using the gradient descent method to update the coupling matrix; The gradient descent method is used to adjust the coupling matrix using the difference between the simulated and true functional connection matrices, so that the simulated functional connection gradually approaches the true functional connection; the update expression of the coupling matrix is ​​as follows: Among them, C (m) represents the coupling matrix of the input of the mth iteration, ∈ represents the learning rate, which controls the update amplitude of each step, and m represents the mth iteration; Then use the updated C (m+1) Re-run the DMF model to generate a new BOLD signal, and then calculate the Pearson correlation coefficient of the new BOLD signal to obtain a new simulated functional connectivity matrix The difference between the new simulated functional connectivity and the real functional connectivity data in the current iteration is calculated, and this difference is compared with the difference obtained in the previous iteration. If the difference no longer decreases and is close to the preset fitting accuracy, the optimization is considered to have converged; otherwise, the iteration continues until the preset fitting accuracy is reached. S6, based on step S5, after the iterative optimization is completed, an optimized coupling matrix, i.e., an optimal coupling matrix, is generated to perform anatomical connection prediction; After the iterative optimization is completed, the coupling matrix C that conforms to the neural dynamics characteristics is generated optimized , as the final input of the DMF model, is used to predict anatomical connections and replace anatomical connection data to build a higher-precision brain model.

Images