Functional magnetic resonance based anxiety disorder transcranial electrical stimulation intervention system and method

Abstract

This invention relates to the fields of medical devices and neuromodulation technology, and discloses a transcranial electrical stimulation intervention system and method for anxiety disorders based on functional magnetic resonance imaging (fMRI). The system includes: a brain functional imaging analysis module, used to acquire fMRI data of subjects performing a task involving attentional bias towards threatening stimuli, extracting BOLD time series of the dorsolateral prefrontal cortex and amygdala, calculating the effective connectivity strength value from the dorsolateral prefrontal cortex to the amygdala, the functional connectivity value between the dorsolateral prefrontal cortex and the amygdala, recording the subject's behavioral response time index to threatening stimuli, and generating an individualized stimulation plan based on the above three indicators using a dynamic hierarchical control algorithm; a transcranial direct current stimulation intervention module, which applies electrical stimulation to the target brain region according to the individualized stimulation plan; and a central control and data processing module, connected to the first two modules respectively, for coordinating the workflow. This invention improves the accuracy and individualization of intervention.

Description

Technical Field

[0001] This invention relates to the fields of medical devices and neuromodulation technology, specifically to a transcranial electrical stimulation intervention system and method for anxiety disorders based on functional magnetic resonance imaging. Background Technology

[0002] Anxiety disorders are characterized by a high relapse rate, and one of their core pathological features is an abnormal attentional bias towards threatening stimuli. Transcranial direct current stimulation (tDCS), as a non-invasive neuromodulation technique, has shown potential in treating anxiety disorders by modulating cortical excitability. However, existing tDCS intervention protocols are mostly "open-loop," employing fixed stimulation targets (such as the F3 position in the international 10-20 system) and uniform stimulation parameters (current intensity, number of treatments, etc.). This approach fails to adequately consider the individual differences in neural network structure and function among patients, resulting in inconsistent intervention outcomes and insufficient precision and individualization.

[0003] Functional magnetic resonance imaging (fMRI) can non-invasively reveal patterns of brain functional connectivity. Previous studies have shown that weakened top-down regulation of the amygdala by the dorsolateral prefrontal cortex (dlPFC) is an important neural mechanism in anxiety disorders. Furthermore, studies combining fMRI and tDCS have revealed that anodic stimulation of the dlPFC in individuals with high trait anxiety significantly reduces the amygdala's response to threatening stimuli, confirming the causal direction of prefrontal regulation of amygdala function. Granger causality analysis based on resting-state fMRI further confirms the failure of top-down inhibition of the amygdala by the dlPFC in patients with generalized anxiety disorder. However, most of these studies remain at the level of causal verification of neural mechanisms or fixed-parameter intervention, failing to address the following practical issues: First, existing methods lack multidimensional quantitative assessment tools for individualized neural circuit abnormalities in patients. Most approaches rely solely on single neuroimaging indicators (such as functional connectivity strength or local activation levels) or clinical scale scores, making it difficult to comprehensively characterize the specific damage patterns of "top-down" inhibitory function in the prefrontal-amygdala emotion regulation circuit (e.g., whether effective connectivity is impaired in isolation, or whether both functional and effective connectivity are impaired), resulting in an inability to accurately match intervention strategies.

[0004] Second, existing methods lack standardized mapping rules from individualized neural circuit assessment results to tDCS stimulation parameters. The determination of key parameters such as stimulation targets, current intensity, and number of treatment sessions often relies on the subjective experience of clinicians, resulting in significant differences in protocols between different doctors or institutions, and poor repeatability and consistency.

[0005] Third, existing methods lack a cross-treatment iterative optimization mechanism based on quantitative indicators. After completing a course of treatment, there is a lack of objective and quantitative decision-making basis for whether and how to adjust subsequent treatment plans, which still relies mainly on clinical experience and makes it difficult to achieve closed-loop, dynamic optimization and precise control.

[0006] Therefore, how to quantify individualized neural circuit abnormalities into intervention strategies for transcranial direct current stimulation, and how to make clear and dynamic adjustments based on the recovery of circuit function after intervention, are technical problems that have not yet been effectively solved. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a transcranial direct current stimulation intervention system and method for anxiety disorders based on functional magnetic resonance imaging (fMRI) data. This addresses the technical problems of low individualization of intervention programs, limited dimensions of neural circuit assessment, and lack of objective quantitative basis for cross-treatment program adjustments in existing technologies. This invention integrates effective connectivity analysis, functional connectivity analysis, and behavioral indicators from task-based fMRI to construct deviation scoring rules and dynamic hierarchical control algorithms. This achieves standardized closed-loop control from neural circuit assessment to stimulation parameter generation and cross-treatment program iteration, thereby improving the accuracy and individualization of interventions.

[0008] To achieve the above objectives, the following technical solution is adopted: In a first aspect, embodiments of the present invention provide a transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging, comprising: a brain functional imaging analysis module, a transcranial direct current stimulation intervention module, and a central control and data processing module; wherein: The brain functional imaging analysis module is used to acquire functional magnetic resonance imaging data of subjects when performing a task that requires attention to threatening stimuli, extract the BOLD time series of the dorsolateral prefrontal cortex and amygdala, calculate the effective connectivity strength value from the dorsolateral prefrontal cortex to the amygdala and the functional connectivity value between the dorsolateral prefrontal cortex and the amygdala, record the subjects' behavioral reaction time indices to threatening stimuli, and generate individualized stimulation plans through a dynamic hierarchical control algorithm based on the effective connectivity strength value, the functional connectivity value, and the behavioral reaction time indices. The transcranial direct current stimulation intervention module is a transcranial direct current stimulator, including stimulation electrodes, a control unit and a communication unit, used to apply electrical stimulation to the target brain region according to the individualized stimulation plan; The central control and data processing module is connected to the brain function imaging analysis module and the transcranial direct current stimulation intervention module, respectively, to coordinate the workflow.

[0009] Preferably, the extraction of BOLD time series from the dorsolateral prefrontal cortex and amygdala includes: Based on standardized brain atlases, the dorsolateral prefrontal cortex is defined as Brodman areas 9 and 46, which correspond to the dorsolateral part of the bilateral superior frontal gyrus and the middle frontal gyrus in the AALv3 atlas; the amygdala is defined as the bilateral amygdala in the AALv3 atlas. The average BOLD signal of all voxels within the defined region was extracted and used as the BOLD time series of the dorsolateral prefrontal cortex and amygdala.

[0010] Preferably, the calculation of the effective connectivity strength value from the dorsolateral prefrontal cortex to the amygdala and the functional connectivity value between the dorsolateral prefrontal cortex and the amygdala includes: Granger causal analysis based on a multivariate autoregressive model was used to calculate the path coefficient from the dorsolateral prefrontal cortex to the amygdala, which was used as the effective connectivity strength value. The Pearson correlation coefficient between the BOLD time series of the dorsolateral prefrontal cortex and the amygdala was calculated and then subjected to Fisher's Z-transformation to obtain the functional connectivity values.

[0011] Preferably, when calculating the effective connectivity strength value, the brain functional imaging analysis module is further configured as follows: The stationarity of the extracted BOLD time series was tested. The optimal order of a multivariate autoregressive model is automatically determined using the Bayesian information criterion. The statistical significance of Granger causality was determined by a bootstrap permutation test based on residuals, and the false discovery rate method was used to correct the results of multiple comparisons.

[0012] Preferably, the dynamic hierarchical control algorithm calculates the total score based on the deviation scoring rule, which adopts the Z-score standardized scoring method: calculating the standardized scores of the effective connectivity strength value, functional connectivity value, and behavioral reaction time index relative to the healthy control norm database, and summing the standardized scores of each index that deviate only from the pathological direction to obtain the total score; wherein, the deviation from the pathological direction refers to: the standardized scores of the effective connectivity strength value and functional connectivity value being negative, and the standardized score of the behavioral reaction time index being positive.

[0013] Preferably, the dynamic hierarchical control algorithm divides the subject into multiple control levels based on the total score, including: When the total score is within the first preset score range, it is determined to be the first control level, and the first stimulation intensity and the first number of treatment courses are adopted; When the total score is within the second preset score range, it is determined to be the second control level, and the second stimulation intensity and the second number of treatment courses are adopted; When the total score is within the third preset score range, it is determined to be the third control level, and the third stimulation intensity and the third number of treatment courses are adopted; Among them, the intensity of the third stimulus is greater than or equal to the intensity of the second stimulus, which is greater than the intensity of the first stimulus, and the number of treatments in the third course is greater than the number of treatments in the second course, which is greater than the number of treatments in the first course.

[0014] Preferably, the dynamic hierarchical control algorithm further determines a stimulus target localization strategy based on the deviation score of the effective connectivity strength value and the deviation score of the functional connectivity value, wherein the stimulus target localization strategy includes: If the deviation score of the effective connectivity strength value is not less than 2 points and the deviation score of the functional connectivity value is not greater than 1 point, then the stimulation target point will be adjusted from the standard F3 position to the projection point of the individual dorsolateral prefrontal cortex activation peak coordinates on the scalp surface. If the deviation score of the effective connection strength value is not less than 2 points and the deviation score of the functional connection value is greater than 1 point, then the standard F3 position is maintained. If the deviation score of the effective connection strength value is less than 2 points, the standard F3 position is maintained; Furthermore, when the condition of adjusting the stimulation target point to the projection point of the individual's dorsolateral prefrontal cortex activation peak coordinates on the scalp surface is met, if no dorsolateral prefrontal cortex activation peak above the uncorrected statistical threshold p < 0.001 is detected in the functional magnetic resonance imaging task, the system will revert to the standard F3 position.

[0015] Preferably, the control unit of the transcranial direct current stimulation intervention module includes a current monitoring and impedance adaptive circuit. The current monitoring and impedance adaptive circuit is used to ensure that the deviation between the output current and the set value does not exceed a preset range, and uses a four-wire method to monitor the electrode-skin contact impedance in real time. When the monitored impedance exceeds a preset threshold, the stimulation output is automatically paused and an alarm signal is issued.

[0016] Preferably, the brain functional imaging analysis module is further configured to reacquire and analyze the subject's functional magnetic resonance imaging data after a treatment course, recalculate the multidimensional neural circuit assessment indicators, and generate an individualized stimulation plan for the next treatment course through the dynamic hierarchical regulation algorithm; when generating the next treatment course plan, a hysteresis mechanism is introduced, and when the recalculated total score is in the boundary region between adjacent regulation levels and the change in the total score compared with the previous assessment is less than a preset value, the original regulation level is maintained.

[0017] Secondly, embodiments of the present invention also provide a transcranial electrical stimulation intervention method for anxiety disorders based on functional magnetic resonance imaging, applied to the system described in the first aspect, comprising the following steps: S1: Obtain functional magnetic resonance imaging data of subjects when performing a task that causes attentional bias towards threatening stimuli, extract the BOLD time series of the dorsolateral prefrontal cortex and amygdala, calculate the effective connectivity strength value from the dorsolateral prefrontal cortex to the amygdala, the functional connectivity value between the dorsolateral prefrontal cortex and the amygdala, and record the subjects' behavioral response time in response to threatening stimuli. S2: According to the preset deviation scoring rules, calculate the deviation scores of the effective connection strength value, functional connection value and behavioral reaction time index respectively, and sum them to obtain the total score. Based on the total score, determine the stimulation intensity and number of treatment sessions to generate an individualized stimulation plan. S3: According to the individualized stimulation program, the anode electrode of the transcranial direct current stimulation is positioned in the scalp area corresponding to the left dorsolateral prefrontal cortex, and the cathode electrode is placed in the contralateral supraorbital margin area, and electrical stimulation is applied according to preset parameters. S4: After a course of treatment, repeat steps S1 to S3, recalculate the total score, and adjust the stimulation plan for the next course of treatment based on the changes in the total score.

[0018] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention constructs a deviation scoring rule by integrating quantitative assessment results from three dimensions: effective connectivity index (EC), functional connectivity index (FC), and behavioral index (RT). This enables a multi-dimensional quantitative characterization of the abnormal function of the prefrontal-amygdala emotion regulation circuit in patients with anxiety disorders. Compared to existing technologies that rely on single neuroimaging indicators or fixed thresholds, this invention can more comprehensively reflect the specific patterns of neural circuit damage in patients, providing richer decision-making basis for the generation of subsequent intervention plans.

[0019] 2. This invention uses a dynamic hierarchical control algorithm to map deviation scores to specific combinations of stimulation parameters (including stimulation current intensity, number of treatment sessions, and stimulation target localization strategies), establishing a standardized rule path from neural circuit assessment to intervention plan generation. This rule path has clear quantitative boundaries and logical branches, avoiding the consistency and repeatability issues caused by relying on clinicians' subjective experience to adjust parameters in existing technologies.

[0020] 3. This invention introduces a case-based judgment rule into the stimulation target localization strategy: when the effective connectivity index (EC) deviates significantly while the functional connectivity index (FC) remains relatively stable, it is determined that the inhibitory function of the prefrontal cortex on the amygdala is isolatedly damaged, and in this case, the activation peak position of individual functional magnetic resonance imaging is used for precise target localization; when both indices deviate significantly, it is determined to be extensive circuit damage, and in this case, stimulation is performed at the standard anatomical position F3. This case-based strategy differs from the single mode of "fixed target" or "fully individualized localization" in existing technologies, and can select an appropriate target localization method according to the actual damage characteristics of the patient's neural circuit.

[0021] 4. This invention introduces a hysteresis mechanism in the cross-treatment iterative control step. When the total score S is in the boundary region of adjacent levels and the change is small, the original stimulation level remains unchanged, thereby avoiding frequent switching of stimulation levels due to small fluctuations in the score, and improving the stability and clinical operability of the cross-treatment plan.

[0022] 5. In the efficient connection calculation process, this invention improves the statistical rigor and reliability of Granger causality analysis in the application of functional magnetic resonance imaging data by performing augmented Dickey-Fuller stationarity test before modeling, automatically determining the model order using Bayesian information criterion, and using bootstrap permutation test combined with error detection rate correction for significance determination.

[0023] It should be understood that the description in the Summary of the Invention is not intended to limit the key or essential features of the embodiments of the present invention, nor is it intended to restrict the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] The above and other features, advantages, and aspects of the various embodiments of the present invention will become more apparent from the accompanying drawings and the following detailed description. The drawings are provided for a better understanding of the invention and are not intended to limit the invention. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein: Figure 1 This is a schematic diagram of the architecture of the transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of electrode positioning according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the process by which the activation peak forms a projection point on the scalp after registration in an embodiment of the present invention; Figure 4 This is a schematic flowchart of a transcranial electrical stimulation intervention method for anxiety disorders based on functional magnetic resonance imaging, according to an embodiment of the present invention. Detailed Implementation

[0025] 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, and 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.

[0026] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0027] Figure 1 This is a schematic diagram of the architecture of a transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging, provided in an embodiment of the present invention. Figure 1 As shown, a transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging includes: a brain functional imaging analysis module 110, a transcranial direct current stimulation intervention module 120, and a central control and data processing module 130; wherein: The brain functional imaging analysis module 110 is used to acquire functional magnetic resonance imaging (fMRI) data of subjects when performing a task that requires attention to threatening stimuli, extract the BOLD time series of the dorsolateral prefrontal cortex and amygdala, calculate the effective connectivity strength value from the dorsolateral prefrontal cortex to the amygdala and the functional connectivity value between the dorsolateral prefrontal cortex and the amygdala, record the subjects' behavioral reaction time indices to threatening stimuli, and generate individualized stimulation plans through a dynamic hierarchical control algorithm based on the effective connectivity strength value, the functional connectivity value and the behavioral reaction time indices. Brain functional imaging analysis module 110 is configured as follows: (1) Individualized neural network evaluation: Based on standardized brain atlases, BOLD time series of key brain regions—dorsolateral prefrontal cortex and amygdala—were extracted. This included: defining the dorsolateral prefrontal cortex as Broadman areas 9 and 46 based on standardized brain atlases, corresponding to the dorsolateral part of the bilateral superior frontal gyrus (AAL tags 3 and 4) and the middle frontal gyrus (AAL tags 7 and 8) in the AALv3 atlas; defining the amygdala as the bilateral amygdala in the AALv3 atlas (AAL tags 41 and 42); and extracting the average BOLD signal of all voxels within the defined regions as the BOLD time series of the dorsolateral prefrontal cortex and amygdala.

[0028] The effective connectivity strength (EC) between key brain regions and the functional connectivity value (FC) between the dorsolateral prefrontal cortex and the amygdala were calculated, and the behavioral response time (RT) of subjects to threatening stimuli was recorded.

[0029] The effective connectivity strength (EC) from the dorsolateral prefrontal cortex (PBF) to the amygdala and the functional connectivity (FC) between the PBF and the amygdala are calculated. This includes: using effective connectivity analysis to calculate the path coefficient from the PBF to the amygdala as the effective connectivity strength (EC); calculating the Pearson correlation coefficient of the BOLD time series between the PBF and the amygdala, and then performing Fisher's Z-transformation to obtain the functional connectivity (FC). The preferred effective connectivity analysis method is Granger causality analysis based on a multivariate autoregressive model, but it can also be replaced by dynamic causal modeling or psychophysiological interaction analysis methods. Furthermore, the behavioral response time (RT) of the subjects to threatening stimuli is recorded.

[0030] Preferably, when calculating the effective connectivity strength value, the brain functional imaging analysis module 110 performs a stationarity test (such as an augmented Dickey-Fuller test) on the extracted BOLD time series before analysis to ensure that the vector autoregressive model assumptions are met; the optimal order of the multivariate autoregressive model is automatically determined by the Bayesian information criterion, and the order selection range is set from 1 to 15 according to the length of the fMRI time series (e.g., 180 time points); the model parameter estimation adopts the ordinary least squares method; the statistical significance of the Granger causality value is determined by the bootstrap permutation test based on the residuals (e.g., the number of repetitions is not less than 1000), and the multiple comparison correction is performed on the results of the multiple connectivity test using the false discovery rate method.

[0031] In one specific embodiment, the brain functional imaging analysis module 110 is a high-performance computing workstation deployed with a customized analysis pipeline based on Python and MATLAB. The brain functional imaging analysis module 110 first preprocesses the raw DICOM data using SPM12 software, including: head motion correction (excluding time points with FD > 0.5 mm), spatial normalization to MNI space, 6 mm smoothing, and de-linearization. Subsequently, the region of interest time series is extracted based on the AALv3 atlas: the dorsolateral prefrontal cortex is defined as the union of labels 3, 4, 7, and 8; the amygdala is defined as the union of labels 41 and 42. After signal extraction, physiological noise from heartbeats and respiration is removed using regression methods (physiological signals are recorded synchronously during scanning), or noise reduction is performed using the CompCor method based on white matter and cerebrospinal fluid signals. Then, blind deconvolution is performed using a task-state fMRI deconvolution method based on Wiener filtering (double gamma function standard HRF, NSR = 0.1). The stationarity of the deconvolutioned time series is verified using the augmented Dickey-Fuller test. Then, the multivariate Granger causality analysis toolbox (such as MVGC toolbox v1.0) was used, setting the model order to a range of 1 to 15. The optimal order was automatically selected using the Bayesian information criterion, and the regression mode was set to OLS. The average path coefficient from the bilateral dorsolateral prefrontal cortex to the bilateral amygdala was calculated as the EC value. The statistical significance of the Granger causality value was determined by the bootstrap permutation test (1000 times), and the results were corrected for false discovery rate (q < 0.05). Functional connectivity (FC) was obtained by calculating the Pearson correlation coefficient between the deconvolutioned dorsolateral prefrontal cortex and amygdala time series and then performing Fisher's Z-transform. Behavioral reaction times were automatically recorded and the mean was output by the E-Prime task software (attentional bias difference can also be calculated).

[0032] (2) Dynamic hierarchical control algorithm generates individualized schemes: The above indicators are input into the dynamic hierarchical regulation algorithm unit. The dynamic hierarchical regulation algorithm unit uses the dynamic hierarchical regulation algorithm to classify the subject's neuromodulation needs based on three indicators: the effective connectivity strength value from the dorsolateral prefrontal cortex to the amygdala, the functional connectivity value between the dorsolateral prefrontal cortex and the amygdala, and the subject's behavioral response time to threatening stimuli. The unit then generates corresponding stimulus plans.

[0033] The dynamic hierarchical regulation algorithm calculates the total score S based on the three indicators calculated in step (1), classifies the subject's neuromodulation needs, and generates corresponding stimulation plans. The algorithm calculates the total score S based on the deviation scoring rule.

[0034] In one embodiment, the deviation scoring rule adopts a fixed-unit scoring method as shown in Table 1 to calculate the total score S and determine the regulation level and target fine-tuning suggestions, ultimately generating an individualized stimulation plan. This algorithm, based on deviation scoring rules rather than simple threshold judgment, integrates quantitative indicators from three dimensions: effective connectivity, functional connectivity, and behavioral aspects. It determines the stimulation intensity, target coordinates, and treatment plan according to the deviation pattern.

[0035] Table 1. Deviation Scoring Rules

[0036] The pathological significance of deviations in each indicator is explained as follows: Patients with anxiety disorders exhibit insufficient prefrontal cortex-amygdala inhibition; therefore, lower effective connectivity values ​​indicate more severe damage. Weakened functional connectivity reflects decreased neural circuit synchronicity. Prolonged reaction time to threatening stimuli reflects difficulty in disengaging attentional bias. Scoring was performed using rounding down (i.e., the difference was divided by the unit score, and the decimal part was discarded) to ensure consistency. The scoring units (0.01, 0.05, 20 ms) were determined through statistical distribution analysis of healthy controls and patient samples, ensuring that the contribution of each indicator to the total score was on the same order of magnitude.

[0037] The aforementioned reference ranges were independently determined by each implementing institution based on its local fMRI database of healthy controls. The reference values ​​were obtained statistically from data of no fewer than 30 healthy control subjects, and the normal range was determined using the mean ± 1.96 times the standard deviation. For example, the reference values ​​could be set as follows: EC normal range 0.08 to 0.15, FC normal range 0.35 to 0.65, and RT normal range 450 to 550 ms.

[0038] In another preferred embodiment, the deviation scoring rule adopts the Z-score standardized scoring method: the standardized scores of the effective connectivity strength value (EC), functional connectivity value (FC), and behavioral reaction time (RT) relative to the healthy control norm database are calculated respectively, and the standardized scores of each index that deviate only from the pathological direction are summed. The sum of the three deviation scores yields the total score S; where deviation from the pathological direction means that the standardized scores of the effective connectivity strength value (EC) and functional connectivity value (FC) are negative, and the standardized score of the behavioral reaction time (RT) is positive. Specifically: First, calculate the Z-score of each indicator relative to a norm database (no fewer than 30 healthy controls):

[0039] in, : The Z-score standardized value of the effective connectivity metric EC; : Original value of effective connectivity strength from the dorsolateral prefrontal cortex to the amygdala; : The mean value of the effective connectivity index EC in healthy controls in the norm database; : The standard deviation of the effective connectivity index EC in healthy controls in the norm database; Z-score normalized value of functional connectivity index FC; FC: raw value of functional connectivity strength between dorsolateral prefrontal cortex and amygdala; : The mean of the functional connectivity index (FC) of healthy controls in the norm database; : The standard deviation of the functional connectivity index (FC) in healthy controls in the norm database; Z-score standardized value of reaction time (RT), a behavioral indicator; RT: raw value of the average reaction time of the subject to a threatening stimulus; : The mean value of the behavioral reaction time index RT of healthy controls in the norm database; : The standard deviation of RT, a behavioral reaction time index, in healthy controls in the norm database.

[0040] Total deviation score This means that only standardized scores deviating from the pathological direction are included in the summation. This continuous scoring method eliminates the problem of inconsistent scoring scales and provides a more granular grading capability. In addition, during cross-treatment iterations, the weight of each indicator in the total score can be dynamically adjusted according to its contribution to the improvement of efficacy (for example, if multiple iterations show that the improvement of effective connectivity is highly correlated with clinical efficacy, the weight coefficient of EC can be appropriately increased), forming a semi-adaptive optimization mechanism.

[0041] Wherein, S: the total deviation score that integrates multi-dimensional deviation information; Only the contribution value of effective connection deviation is considered for deviations in the pathological direction (effective connections are lower than normal). Only the contribution value of functional connectivity deviation is counted for deviations in the pathological direction (functional connectivity is lower than normal); Only the behavioral deviation contribution value of deviation in the pathological direction (reaction time higher than normal) is counted.

[0042] Furthermore, the dynamic graded control algorithm divides the subject into multiple control levels based on the total score S, including: when the total score is within a first preset score range, it is determined to be the first control level, using the first stimulation intensity and the first number of treatment courses; when the total score is within a second preset score range, it is determined to be the second control level, using the second stimulation intensity and the second number of treatment courses; when the total score is within a third preset score range, it is determined to be the third control level, using the third stimulation intensity and the third number of treatment courses; wherein, the third stimulation intensity ≥ the second stimulation intensity > the first stimulation intensity, and the third number of treatment courses > the second number of treatment courses > the first number of treatment courses.

[0043] In a preferred embodiment, subjects are divided into three regulatory levels based on a total score S: Level 1 (mild, S score 1 to 3): Stimulation intensity 1.5mA, once daily for a total of 10 times; Level 2 (moderate, S score 4 to 7): Stimulation intensity 2.0 mA, once daily for a total of 15 times; Level 3 (severe, S score not less than 8): Stimulation intensity 2.0mA, once daily for a total of 20 times (extension of treatment course).

[0044] The stimulation parameters and treatment course settings are based on existing routine clinical protocols for transcranial direct current stimulation.

[0045] Furthermore, the dynamic hierarchical regulation algorithm also determines the stimulus target localization strategy based on the deviation scores of effective connectivity strength values ​​and functional connectivity values. The stimulus target localization strategy includes: If the deviation score of the effective connectivity strength value (EC) is not less than 2 points and the deviation score of the functional connectivity value (FC) is not greater than 1 point, it indicates that the inhibitory function of the dorsolateral prefrontal cortex on the amygdala is relatively isolated and impaired. The stimulation target point is adjusted from the standard F3 position to the projection point of the individual's dorsolateral prefrontal cortex activation peak coordinates onto the scalp surface, i.e., target fine-tuning. The specific operation is as follows: First, obtain the functional magnetic resonance imaging activation map of the subject when performing a threatening stimulus attention bias task. Extract the maximum activation peak coordinates (MNI coordinates) within the dorsolateral prefrontal cortex region under a statistical threshold p < 0.001 (uncorrected). Then, register these MNI coordinates to the individual's scalp surface using a nonlinear transformation algorithm in a neural navigation system (such as Brainsight or Localite), and combine this with a scalp surface reconstruction model to generate the corresponding scalp projection point. Finally, place the anode electrode (e.g., 5 × 5 cm) onto the scalp surface. 2 The center of the square sponge electrode is moved from the standard F3 position to the scalp projection point position, thereby achieving fine-tuning of the individual activation peak coordinates;

[0046] If the deviation score of the effective connection strength value EC is not less than 2 points and the deviation score of the functional connection value FC is greater than 1 point, it indicates that both the functional connection and the effective connection are severely damaged, and the standard F3 position should be maintained. If the deviation score of the effective connection strength value EC is less than 2 points, the standard F3 position is maintained regardless of the FC score. Furthermore, when the condition of adjusting the stimulation target to the projection point of the individual's dorsolateral prefrontal cortex activation peak coordinates on the scalp surface is met, if no dorsolateral prefrontal cortex activation peak above p<0.001 (uncorrected) is detected in the functional magnetic resonance imaging (fMRI) task, the system reverts to the standard F3 position.

[0047] The transcranial direct current stimulation intervention module 120 is a transcranial direct current stimulator, including stimulation electrodes 121, a control unit 122 and a communication unit 123, used to apply electrical stimulation to the target brain region according to the individualized stimulation plan; Specifically: The transcranial direct current stimulation intervention module 120 is a transcranial direct current stimulator, including: Stimulation electrode 121: Used to apply electrical stimulation with specific parameters to a target brain region according to the individualized stimulation protocol. The stimulation is performed in a non-magnetic resonance environment.

[0048] Control unit 122: Receives the stimulation protocol and precisely controls the intensity, frequency, and duration of the stimulation current. Control unit 122 includes current monitoring and impedance adaptation circuitry to ensure that the output current deviates from the set value within a preset range (e.g., ±0.1 mA, achieved through a high-precision operational amplifier and a 16-bit DAC). Control unit 122 uses a four-wire method to monitor the electrode-skin contact impedance in real time, with the excitation signal being 220Hz, 100Hz. The alternating current. When the impedance exceeds a preset threshold, for example, 5... If a faulty connection is detected, the stimulation output is automatically paused and an alarm signal is issued. The control unit 122 further adjusts the stimulation parameters automatically based on the output results of the dynamic hierarchical control algorithm, and combines impedance monitoring during the stimulation process to achieve closed-loop safety control.

[0049] Communication unit 123: Used for data interaction with the host computer, receiving schemes and uploading records.

[0050] In one specific embodiment, the transcranial direct current stimulation intervention module 120 performs a precise neuromodulation process as follows: Stimulating electrode 121 is placed in a non-magnetic resonance environment. The anode electrode is positioned in the scalp region corresponding to the left dorsolateral prefrontal cortex. When the generated dynamic hierarchical modulation algorithm determines the target point that needs fine-tuning by generating an individualized plan, the operator imports the patient's individual high-resolution T1 structural image and the fMRI statistical activation map obtained in the individualized neural network evaluation step (1) into a neuronavigation system (such as Brainsight or Localite). Taking the maximum activation point in the dorsolateral prefrontal cortex region under the statistical threshold p<0.001 (uncorrected) as the target point, the navigation system automatically registers the MNI coordinates to the individual scalp surface and combines them with the scalp surface reconstruction model to complete the spatial mapping from the brain region to the scalp, generating the corresponding scalp projection point. Under the guidance of the navigation probe, the operator places a 5×5cm 2 The center of the square sponge electrode (soaked in 0.9% saline) is aligned with the projection point. The cathode electrode is placed in the contralateral supraorbital region. The transcranial direct current stimulator is activated, and the control unit 122 outputs direct current stimulation according to preset parameters (30 seconds each for the ramp ascent and descent). Figure 2 As shown, the anode electrode is located in the scalp region corresponding to the left dlPFC (F3 position), and the cathode electrode is placed in the contralateral supraorbital region; when fine-tuning is required, the electrode center is aligned with the projection point of the individual's fMRI activation peak on the scalp. Figure 3 As shown, after registration, the activation peak forms a projection point on the scalp, which has a spatial offset relative to the standard F3 position. The operator uses this projection point to accurately place the electrode.

[0051] Preferably, the transcranial direct current stimulation intervention module 120 is a dual-channel transcranial direct current stimulator, whose control unit is based on an STM32 microcontroller. The current output uses a high-precision constant current source circuit (precision operational amplifier with a 16-bit DAC) to ensure that the output current deviates from the set value by no more than ±0.1mA. A four-wire impedance measurement circuit (excitation signal 220Hz, 100μA) monitors the electrode-skin contact impedance in real time. If the detected impedance exceeds 5kΩ, the microcontroller determines it as poor contact, immediately reduces the output current to 0 and triggers a buzzer alarm, while simultaneously reporting the alarm event to the central control module via the ESP8266 Wi-Fi module. The control unit 122 automatically configures the stimulator's operating state according to the control level and stimulation parameters output by the dynamic graded control algorithm, achieving closed-loop linkage between the algorithm and hardware. Stimulation parameters are received via JSON format commands, such as {"mode":"tDCS","current":1.5,"ramp_up":30,"duration":1200,"ramp_down":30}, where the time parameter unit is seconds. All operation events are recorded in structured log format on the MicroSD card. The log format is [YYYY-MM-DD HH:MM:SS]EVENT_CODE PARAM. The event code table is as follows: 001 Stimulation start, 002 Stimulation end normally, 003 Impedance over-limit pause, 004 Current abnormality pause, 005 Device self-test passed, 006 Communication connection successful.

[0052] The central control and data processing module 130 is connected to the brain function imaging analysis module 110 and the transcranial direct current stimulation intervention module 120, respectively, and is used to coordinate the workflow.

[0053] Furthermore, the central control and data processing module 130 uses AES-128 encryption to protect the security of patient data transmission, and TCP / IP communication uses a dynamic port allocation strategy to avoid port conflicts.

[0054] In one specific embodiment, the central control and data processing module is a Windows 10 PC with control software installed. The software interface includes patient management, protocol import, stimulation monitoring, and log viewing functions. The PC and the stimulator communicate via TCP / IP protocol, employing a dynamic port allocation strategy. The initial port is 8888, and if occupied, it automatically increments until an available port is found. All transmitted data is encrypted using AES-128 to ensure compliance with medical data security requirements. During stimulation, the software queries the stimulator's status (current, impedance, remaining time) every 5 seconds and plots real-time monitoring curves.

[0055] Preferably, the brain functional imaging analysis module 110 is further configured to reacquire and analyze the subject's functional magnetic resonance imaging data after a treatment course, recalculate the multidimensional neural circuit assessment indicators, and generate an individualized stimulation plan for the next treatment course through the dynamic hierarchical regulation algorithm; when generating the next treatment course plan, a hysteresis mechanism is introduced, and when the recalculated total score is in the boundary region of adjacent regulation levels and the change in the total score compared with the previous assessment is less than a preset value, the original regulation level is maintained.

[0056] In one specific embodiment, the system performs the following effect evaluation and cross-treatment plan iteration: After one treatment course, the aforementioned steps of individualized neural network evaluation and dynamic grading adjustment algorithm to generate individualized plans are repeated, and the total score S is recalculated. To reduce the unstable grading caused by boundary fluctuations in the total score S, a hysteresis mechanism is introduced: when the total score S is at the boundary between two levels (e.g., between 3 and 4 points), if the change in the total score S compared to the previous evaluation is less than 1 point, the original level is maintained. The next treatment plan is adjusted based on the score change. If the total score S drops to 0: treatment ends and the patient enters the follow-up observation period.

[0057] If the total score S decreases by no less than 3 points but remains greater than 0 points: maintain the current treatment plan and continue treatment.

[0058] If the total score S drops by less than 3 points and the current level is below level three: the current level will be raised by one level for enhanced intervention.

[0059] If the total score S decreases by less than 3 points and the current level is already level three: maintain the level three treatment plan and recommend that clinicians assess whether to combine other treatment methods.

[0060] like Figure 4 As shown, this embodiment of the invention also provides a transcranial electrical stimulation intervention method for anxiety disorders based on functional magnetic resonance imaging, applied to the aforementioned system, including the following steps: S1: Obtain functional magnetic resonance imaging data of subjects when performing a task that causes attentional bias towards threatening stimuli, extract the BOLD time series of the dorsolateral prefrontal cortex and amygdala, calculate the effective connectivity strength value from the dorsolateral prefrontal cortex to the amygdala, the functional connectivity value between the dorsolateral prefrontal cortex and the amygdala, and record the subjects' behavioral response time in response to threatening stimuli. S2: According to the preset deviation scoring rules, calculate the deviation scores of the effective connection strength value, functional connection value and behavioral reaction time index respectively, and sum them to obtain the total score. Based on the total score, determine the stimulation intensity and number of treatment sessions to generate an individualized stimulation plan. S3: According to the individualized stimulation program, the anode electrode of the transcranial direct current stimulation is positioned in the scalp area corresponding to the left dorsolateral prefrontal cortex, and the cathode electrode is placed in the contralateral supraorbital margin area, and electrical stimulation is applied according to preset parameters. S4: After a course of treatment, repeat steps S1 to S3, recalculate the total score, and adjust the stimulation plan for the next course of treatment based on the changes in the total score.

[0061] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the described method can be referred to the corresponding process in the foregoing system embodiments, and will not be repeated here.

[0062] Furthermore, the following example, using a 30-year-old female subject with generalized anxiety disorder, illustrates the specific implementation process of the method of this invention: S210: Individualized Assessment Subjects completed a point detection task within a 3.0T MRI scanner. An EPI sequence was used, with TR / TE = 2000 / 30ms, 3mm isotropic voxels, 36 slices in total, and a scan duration of 360 seconds, yielding 176 valid time points (the first 4 virtual scans were automatically discarded).

[0063] S220: Region of Interest Extraction and Index Calculation Data were preprocessed using SPM12. Bolding signals from the dorsolateral prefrontal cortex and amygdala were extracted based on the AALv3 map. Regression methods were used to remove synchronously recorded physiological noise related to heartbeats and respiration. After blind deconvolution using Wiener filtering (double gamma function HRF, NSR = 0.1), stationarity was confirmed by the ADF test, and MVGC analysis was performed (model order determined to be 5 using the Bayesian information criterion). The calculated EC = 0.06 (p < 0.05 after error detection rate correction). Functional connectivity FC = 0.28. Behavioral response time (RT) = 610 ms.

[0064] S230: Application of Dynamic Hierarchical Control Algorithm Calculate the deviation score according to the rules in Table 1: (1) EC: 0.08 below the reference lower limit, difference 0.02, 0.02 / 0.01 = 2 points (rounded down).

[0065] (2) FC: 0.35 below the reference lower limit, the difference is 0.07, 0.07 / 0.05 = 1.4, round down to 1 point.

[0066] (3) RT: 550ms higher than the upper limit of the reference, the difference is 60ms, 60 / 20 = 3 minutes.

[0067] The total score S = 2 + 1 + 3 = 6 points, which is judged as secondary regulation. Further analysis of target location conditions: EC deviation score = 2 points (not less than 2 points), FC deviation score = 1 point (not greater than 1 point), which meets the target fine-tuning conditions.

[0068] S240: Generate an intervention plan The protocol was as follows: anodic tDCS stimulation of the left dorsolateral prefrontal cortex at an intensity of 2.0 mA, once daily for 20 minutes each time (including a 30-second rise / fall), for a total of 15 sessions. Target localization was based on individual fMRI activation peaks.

[0069] S250: Implementing Interventions High-resolution T1 structural images and fMRI task activation maps of the subjects were loaded into the Brainsight neuronavigation system. The maximum activation point in the dorsolateral prefrontal cortex region was selected at a statistical threshold p < 0.001 (uncorrected), with MNI coordinates of (-38, 32, 28). The navigation system mapped these coordinates to the individual's scalp surface through nonlinear transformation and combined with a scalp surface reconstruction model. The operator placed a 5×5cm... 2 Align the center of the anode electrode with the scalp projection point indicated by the navigation probe, and place the cathode at the right supraorbital margin. After confirming the impedance is 3.8 kΩ, initiate 2.0 mA stimulation.

[0070] S260: Cross-treatment efficacy evaluation and iteration After 15 stimulations, the subject underwent another fMRI scan, repeating steps S210 to S230. EC increased to 0.12, FC increased to 0.48, and RT decreased to 530 ms. The deviation score was recalculated. (1) EC: Within the reference range, score 0 points.

[0071] (2) FC: Within the reference range, score 0 points.

[0072] (3) RT: Not exceeding the upper limit, score 0 points.

[0073] The total score S = 0. According to the algorithm rules, treatment ends and follow-up begins.

[0074] The specific embodiments described above are merely illustrative examples of the present invention, used to demonstrate the complete workflow from individualized neural network assessment, dynamic hierarchical modulation algorithm generation of stimulation programs, precise neural modulation execution, to cross-treatment effect evaluation and iteration. Those skilled in the art should understand that the transcranial electrical stimulation intervention system and method for anxiety disorders based on functional magnetic resonance imaging proposed in this invention, by integrating multi-dimensional neural circuit assessment indicators with a clearly defined dynamic hierarchical modulation algorithm, achieves individualized, closed-loop, and iterative transcranial electrical stimulation intervention for patients with anxiety disorders, significantly improving the accuracy and repeatability of treatment. In practical applications, the specific implementation methods of each module (such as the selection of the Granger causal analysis toolbox, the brand of the navigation system, the model of the stimulator's control chip, etc.) can be equivalently replaced according to existing technologies, all of which fall within the protection scope of this invention.

[0075] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0076] It should also be noted that, in the embodiments of this application, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, 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. Without further limitations, 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.

[0077] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined in the embodiments of this application may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown in this application, but is to be accorded the widest scope consistent with the principles and novel features disclosed in the embodiments of this application.

Claims

1. A transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging, characterized in that, include: The module comprises a brain functional imaging analysis module, a transcranial direct current stimulation intervention module, and a central control and data processing module; among which: The brain functional imaging analysis module is used to acquire functional magnetic resonance imaging data of subjects when performing a task that requires attention to threatening stimuli, extract the BOLD time series of the dorsolateral prefrontal cortex and amygdala, calculate the effective connectivity strength value from the dorsolateral prefrontal cortex to the amygdala and the functional connectivity value between the dorsolateral prefrontal cortex and the amygdala, record the subjects' behavioral reaction time indices to threatening stimuli, and generate individualized stimulation plans through a dynamic hierarchical control algorithm based on the effective connectivity strength value, the functional connectivity value, and the behavioral reaction time indices. The transcranial direct current stimulation intervention module is a transcranial direct current stimulator, including stimulation electrodes, a control unit and a communication unit, used to apply electrical stimulation to the target brain region according to the individualized stimulation plan; The central control and data processing module is connected to the brain function imaging analysis module and the transcranial direct current stimulation intervention module, respectively, to coordinate the workflow.

2. The transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging according to claim 1, characterized in that, The extracted BOLD time series from the dorsolateral prefrontal cortex and amygdala include: Based on standardized brain atlases, the dorsolateral prefrontal cortex is defined as Brodman areas 9 and 46, which correspond to the dorsolateral part of the bilateral superior frontal gyrus and the middle frontal gyrus in the AALv3 atlas; the amygdala is defined as the bilateral amygdala in the AALv3 atlas. The average BOLD signal of all voxels within the defined region was extracted and used as the BOLD time series of the dorsolateral prefrontal cortex and amygdala.

3. The transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging according to claim 2, characterized in that, The calculation of the effective connectivity strength from the dorsolateral prefrontal cortex to the amygdala and the functional connectivity between the dorsolateral prefrontal cortex and the amygdala includes: Granger causal analysis based on a multivariate autoregressive model was used to calculate the path coefficient from the dorsolateral prefrontal cortex to the amygdala, which was used as the effective connectivity strength value. The Pearson correlation coefficients of the BOLD time series of the dorsolateral prefrontal cortex and amygdala were calculated and then subjected to Fisher's Z transformation to obtain the functional connectivity values.

4. The transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging according to claim 3, characterized in that, The brain functional imaging analysis module, before employing Granger causal analysis based on a multivariate autoregressive model, is configured to calculate the effective connectivity strength value as follows: The stationarity of the extracted BOLD time series was tested. The optimal order of a multivariate autoregressive model is automatically determined using the Bayesian information criterion. The statistical significance of Granger causality was determined by a bootstrap permutation test based on residuals, and the false discovery rate method was used to correct the results of multiple comparisons.

5. The transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging according to claim 1, characterized in that, The dynamic hierarchical control algorithm calculates the total score based on the deviation scoring rule, which adopts the Z-score standardized scoring method: the standardized scores of the effective connectivity strength value, functional connectivity value, and behavioral reaction time index relative to the healthy control norm database are calculated respectively, and the standardized scores of each index that deviate only from the pathological direction are summed to obtain the total score; wherein, the deviation from the pathological direction refers to: the standardized scores of the effective connectivity strength value and functional connectivity value being negative, and the standardized score of the behavioral reaction time index being positive.

6. The transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging according to claim 5, characterized in that, The dynamic hierarchical control algorithm divides subjects into multiple control levels based on the total score, including: When the total score is within the first preset score range, it is determined to be the first control level, and the first stimulation intensity and the first number of treatment courses are adopted; When the total score is within the second preset score range, it is determined to be the second control level, and the second stimulation intensity and the second number of treatment courses are adopted; When the total score is within the third preset score range, it is determined to be the third control level, and the third stimulation intensity and the third number of treatment courses are adopted; Among them, the intensity of the third stimulus is greater than or equal to the intensity of the second stimulus, which is greater than the intensity of the first stimulus, and the number of treatments in the third course is greater than the number of treatments in the second course, which is greater than the number of treatments in the first course.

7. The transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging according to claim 5, characterized in that, The dynamic hierarchical control algorithm further determines a stimulus target localization strategy based on the deviation scores of the effective connectivity strength value and the functional connectivity value, the stimulus target localization strategy including: If the deviation score of the effective connectivity strength value is not less than 2 points and the deviation score of the functional connectivity value is not greater than 1 point, then the stimulation target point will be adjusted from the standard F3 position to the projection point of the individual dorsolateral prefrontal cortex activation peak coordinates on the scalp surface. If the deviation score of the effective connection strength value is not less than 2 points and the deviation score of the functional connection value is greater than 1 point, then the standard F3 position is maintained. If the deviation score of the effective connection strength value is less than 2 points, the standard F3 position is maintained; Furthermore, when the condition of adjusting the stimulation target to the projection point of the individual's dorsolateral prefrontal cortex activation peak coordinates on the scalp surface is met, if no dorsolateral prefrontal cortex activation peak above the uncorrected statistical threshold p < 0.001 is detected in the functional magnetic resonance imaging task, the system will revert to the standard F3 position.

8. The transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging according to claim 1, characterized in that, The control unit of the transcranial direct current stimulation intervention module includes a current monitoring and impedance adaptive circuit. The current monitoring and impedance adaptive circuit is used to ensure that the deviation between the output current and the set value does not exceed the preset range, and uses a four-wire method to monitor the electrode-skin contact impedance in real time. When the monitored impedance exceeds the preset threshold, the stimulation output is automatically paused and an alarm signal is issued.

9. The transcranial electrical stimulation intervention system for anxiety disorders based on functional magnetic resonance imaging according to claim 1, characterized in that, The brain functional imaging analysis module is also configured to reacquire and analyze the subject's functional magnetic resonance imaging data after a treatment course, recalculate the multidimensional neural circuit assessment indicators, and generate an individualized stimulation plan for the next treatment course through the dynamic hierarchical regulation algorithm. When generating the next treatment plan, a hysteresis mechanism is introduced. When the recalculated total score is in the boundary region between adjacent regulation levels and the change in the total score compared with the previous assessment is less than a preset value, the original regulation level is maintained.

10. A transcranial electrical stimulation intervention method for anxiety disorders based on functional magnetic resonance imaging, applied to the system according to any one of claims 1 to 9, characterized in that, Includes the following steps: S1: Obtain functional magnetic resonance imaging data of subjects when performing a task that causes attentional bias towards threatening stimuli, extract the BOLD time series of the dorsolateral prefrontal cortex and amygdala, calculate the effective connectivity strength value from the dorsolateral prefrontal cortex to the amygdala, the functional connectivity value between the dorsolateral prefrontal cortex and the amygdala, and record the subjects' behavioral response time in response to threatening stimuli. S2: According to the preset deviation scoring rules, calculate the deviation scores of the effective connection strength value, functional connection value and behavioral reaction time index respectively, and sum them to obtain the total score. Based on the total score, determine the stimulation intensity and number of treatment sessions to generate an individualized stimulation plan. S3: According to the individualized stimulation program, the anode electrode of the transcranial direct current stimulation is positioned in the scalp area corresponding to the left dorsolateral prefrontal cortex, and the cathode electrode is placed in the contralateral supraorbital margin area, and electrical stimulation is applied according to preset parameters. S4: After a course of treatment, repeat steps S1 to S3, recalculate the total score, and adjust the stimulation plan for the next course of treatment based on the changes in the total score.

Images