An evaluation system for epilepsy electroacupuncture treatment target based on amygdala neural response prediction

Abstract

The application discloses an evaluation system for an epilepsy electroacupuncture treatment target based on an amygdala nerve response prediction, relates to the technical field of nerve regulation and brain function modeling, and realizes the multi-modal synchronous collection of brain region potentials, blood oxygen and metabolic activities through high-density EEG, electrophysiological sensors, fMRI and PET, and constructs a standardized nerve response dataset and a brain region function mapping matrix. An amygdala prediction response index is calculated and compared with a response threshold, a stimulation response compliance judgment and parameter correction are realized. The nerve response propagation delay, cooperative coupling coefficient and epilepsy wave front tracing time are calculated, a comprehensive target evaluation coefficient is obtained, and effective target points are screened. Based on the treatment scheme implementation of the effective target point set, the nerve feedback data in the continuous treatment cycle is monitored in real time, and a nerve plasticity correction index is calculated, compared with a plasticity stability threshold, an adaptive feedback regulation closed loop is formed, and precise prediction and dynamic optimization control of epilepsy electroacupuncture targets are realized.

Description

Technical Field

[0001] This invention relates to the field of neural modulation and brain function modeling technology, specifically to an epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction. Background Technology

[0002] Epilepsy is a chronic neurological disorder characterized by abnormal electrical discharges in the brain. Its seizures take many forms and involve complex abnormal neural network activity. Currently, commonly used clinical treatments include medication, surgical resection, and neuromodulation. Among these, electroacupuncture, as a non-invasive neuromodulation method, has been widely studied because it can regulate central nervous system activity and improve functional connectivity in brain regions through stimulation of specific acupoints. However, the efficacy of electroacupuncture for epilepsy exhibits significant individual differences, with key issues lying in the insufficient accuracy of identifying effective stimulation targets and predicting neural responses.

[0003] Traditional methods for assessing electroacupuncture targets often rely on experience or single EEG indicators, lacking multimodal dynamic coupling analysis of brain regions related to emotion and seizure regulation, such as the amygdala, thalamus, and cortex, and thus failing to reflect complex neural network interactions. Furthermore, current technologies for detecting electroacupuncture stimulation responses largely remain at the EEG signal level, lacking fusion processing with functional magnetic resonance imaging (fMRI), positron emission tomography (PET), and other brain functional imaging signals. This results in limited target localization accuracy and makes it difficult to achieve a comprehensive assessment of individualized neurological responses in patients.

[0004] During electroacupuncture treatment, brain responses exhibit time-varying and plasticity. Subtle adjustments to stimulation parameters (such as frequency, pulse width, and waveform) can affect neural synchronization patterns and firing activity in key brain regions like the amygdala. Existing electroacupuncture systems generally lack real-time feedback mechanisms and adaptive adjustment strategies, making it impossible to automatically correct stimulation parameters based on changes in neural plasticity during treatment. This can easily lead to unstable therapeutic effects or overstimulation. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction, in order to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction, comprising: The neural signal acquisition module is used to perform multimodal synchronous monitoring and preprocessing of the patient's brain neural activity under electroacupuncture stimulation: it acquires brain region potential changes and frequency band energy through a high-density EEG electrode array, blood oxygenation changes and metabolic activity through fMRI and PET, and local neural potentials and synaptic response timing through a high-resolution sensor array; after preprocessing, it forms a standardized neural response dataset and constructs a brain region functional mapping matrix that reflects the correlation coefficient and connection weight of signals between brain regions; The amygdala neural response prediction module is used to extract multimodal neural feature parameters based on a standardized neural response dataset, including phase lock value (PLV), power spectral density (PSD), anomalous synchronization ratio (ASR), emotion regulation connectivity strength (EAC), and situational perturbation intensity (Cctx). It calculates the amygdala predicted response index (ARI) and compares it with the amygdala response threshold (Ath) to determine whether the amygdala response meets the standard under the current stimulus parameters. If it does, a comprehensive parameter dataset is generated; if it does not, dynamic correction of the stimulus parameters is triggered until the response meets the standard. The neural effect coupling analysis module is used to cluster and spatially reconstruct electrical stimulation response features based on a comprehensive parameter set and a brain region function mapping matrix, generating a candidate set of electroacupuncture targets; for each candidate target, the neural response propagation delay is extracted. amygdala-thalamus-cortex co-coupling coefficient Epilepsy wavefront retrospective time Calculate the comprehensive target evaluation coefficient The target is compared with the target validity threshold Tth. If a target is deemed valid, it is included in the set of valid targets. If a target is deemed invalid, it is removed and the analysis iteration is re-executed. The neural response feedback adjustment module is used to collect neural feedback data in real time during continuous treatment cycles based on an effective target set, and to acquire the rate of change in electrical activity. The rate of change of power of epileptic waves and the rate of change in the strength of the emotion regulation network The neural plasticity correction index (NPI) is calculated and compared with the plasticity stability threshold (Nth) to determine whether neural plasticity is stable under the current treatment state. If it is stable, the current treatment plan is maintained; if it is unstable, feedback correction is triggered to dynamically adjust the stimulation parameters and model weights to form an adaptive treatment closed loop.

[0007] Preferably, the neural signal acquisition module includes an electroencephalogram (EEG) acquisition unit, a brain functional imaging unit, a neural electrical stimulation response detection unit, a data preprocessing unit, and a mapping matrix construction unit; The EEG acquisition unit is used to monitor the electrical activity state of the patient's cerebral cortex in real time; by deploying a high-density EEG electrode array on the scalp surface, it acquires brain region potential change signals and EEG frequency band energy distribution information; The brain functional imaging unit is used to simultaneously detect changes in brain blood oxygenation levels, metabolic activity, and blood flow distribution based on functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) devices, and to acquire fMRI blood oxygenation change signals, PET metabolic activity signals, and blood flow distribution data in the anterior cingulate cortex and amygdala regions. The neural electrical stimulation response detection unit is used to monitor local neural potential changes, synaptic response timing, and conduction path characteristics using a high-resolution neural response sensor array under electroacupuncture stimulation or simulated stimulation conditions. The data preprocessing unit is used to perform time synchronization, noise suppression and spatial registration on multi-source signals, eliminate signal offset caused by device delay and individual differences, and form a standardized neural response dataset. The mapping matrix construction unit is used to perform feature extraction and correlation modeling on EEG frequency band energy distribution information, fMRI blood oxygenation change signals and PET metabolic activity signals after data preprocessing, using cross-correlation analysis and functional connectivity strength estimation algorithms. It analyzes the signal correlation coefficients and functional connectivity weight matrix parameters of each brain region to construct the brain region functional mapping matrix Mbrain.

[0008] Preferably, the amygdala neural response prediction module includes a neural response parameter extraction unit, a first calculation unit, and a first analysis unit; The neural response parameter extraction unit is used to perform feature decoding on brain region potential change signals based on a standardized neural response dataset. It establishes a mapping relationship between the electrical stimulation input signal and the amygdala response signal using a trained deep temporal network model, and extracts the phase-locked value (PLV), a characteristic parameter of epileptic abnormal synchronicity. For EEG frequency band energy distribution information, it uses power spectral decomposition and multi-scale wavelet analysis to calculate brain region discharge energy distribution characteristics and obtain the power spectral density parameter (PSD). Based on fMRI blood oxygenation change signals and PET metabolic activity signals, it uses correlation modeling and functional connectivity analysis algorithms to perform temporal correlation analysis on the synchronous response patterns between the amygdala and thalamus, obtaining the abnormal synchronization ratio (ASR). For blood flow distribution data in the anterior cingulate cortex and amygdala region, it uses a functional connectivity strength estimation algorithm to calculate the emotion regulation connectivity strength (EAC). For local neural potential changes, synaptic response timing, and conduction path characteristics, it uses synaptic temporal-dependent plasticity (STDP) modeling and neural pathway perturbation analysis methods to obtain the situational perturbation strength (Cctx).

[0009] Preferably, the first calculation unit is used to construct the amygdala prediction index ARI by using the acquired epilepsy abnormal synchronization characteristic parameters, such as phase lock value (PLV), power spectral density parameter (PSD), abnormal synchronization ratio (ASR), emotion regulation connection strength (EAC), and situational perturbation strength (Cctx), after dimensionless normalization. The first analysis unit is used to obtain a first evaluation result by comparing the amygdala predicted index (ARI) with the amygdala response threshold (Ath) using a preset amygdala response threshold (Ath): When the amygdala predicts the corresponding index ARI ≥ amygdala response threshold Ath, it indicates that the amygdala response meets the target under the current stimulation parameters. A preliminary target assessment model is automatically generated, and a comprehensive parameter set Mtarget including local neural potential characteristics, synaptic response coupling strength and pathway activation weight is recorded. When the amygdala predicts the corresponding index ARI < amygdala response threshold Ath, it indicates that the amygdala response under the current stimulation parameters has not met the standard, triggering the first warning instruction and generating the first strategy: dynamically reducing the step size of the electroacupuncture stimulation frequency, adjusting the pulse width and waveform duty cycle, and combining the optimal parameter range of similar individuals in historical data to generate a set of corrected stimulation parameters, re-collecting data, and recalculating until the amygdala predicts the corresponding index ARI ≥ amygdala response threshold Ath.

[0010] Preferably, the neural effect coupling analysis module includes a target generation unit, a target neurodynamic parameter extraction unit, a second calculation unit, and a second analysis unit; The target generation unit is used to construct an electrical stimulation response distribution model by performing cluster analysis and spatial projection reconstruction on neural response patterns under different stimulation parameters based on the comprehensive parameter set Mtarget combined with the brain region functional mapping matrix Mbrain, using multidimensional feature clustering and spatial correlation modeling techniques. Through the model, the response characteristics of the amygdala and its functional connectivity regions are spatially mapped and path-weighted analyzed to generate an electroacupuncture target candidate set Tcand, and the amygdala functional connectivity region, neural conduction path and response intensity index information corresponding to each target are recorded.

[0011] Preferably, the target neurodynamic parameter extraction unit is used to calculate key neural effect parameters for each target t in the electroacupuncture target candidate set Tcand, based on the temporal electrophysiological characteristics and functional magnetic resonance dynamic signal change information in the standardized neural response dataset. Specifically, this includes: based on brain region potential change signals and EEG frequency band energy distribution information, using cross-correlation time delay analysis and phase synchronization tracking methods, estimating the time delay and aligning the phase of the temporal characteristics of brain region potential change signals, extracting the average signal transmission time difference between the amygdala and the target brain region, and obtaining the neural response propagation delay. Based on local neural potential changes, synaptic response timing, and conduction pathway characteristics, and combined with the brain region functional mapping matrix Mbrain, this study employs dynamic causal modeling (DCM) and partial minimum correlation analysis (PLSC) to jointly model and perform directional coupling analysis on local neural potential changes and the functional connectivity weights shown in the Mbrain mapping matrix. This quantifies the functional connectivity strength and directional coupling characteristics between the amygdala, thalamus, and cortex, obtaining the amygdala-thalamus-cortex co-coupling coefficient. Based on a standardized neural response dataset, a cross-modal time calibration and abnormal potential event detection method was used to synchronously label brain region potential change signals and local neural potential changes in the standardized neural response dataset. The amygdala abnormal discharge detection time (tabn) and epileptic seizure onset time (tonset) were obtained, and the epileptic wavefront retrospective time was further obtained. .

[0012] Preferably, the second computing unit is used to obtain the propagation delay of the neural response. amygdala-thalamus-cortex co-coupling coefficient Epilepsy wavefront retrospective time After dimensionless normalization, the comprehensive target evaluation coefficient is calculated and obtained. ; The second analysis unit is used to pass a preset target effectiveness threshold Tth and integrate the comprehensive target evaluation coefficient. A comparative analysis was performed with the target effectiveness threshold Tth to obtain the second evaluation results, including: When the comprehensive target evaluation coefficient When the value is greater than or equal to the target effectiveness threshold Tth, it indicates that the current target's neural effect is qualified, and the current target is determined to be an effective target and included in the effective target set Teff for continuous monitoring. When the comprehensive target evaluation coefficient When the value is less than the target effectiveness threshold Tth, it indicates that the current target's neural effect is unqualified. The current target is determined to be an invalid target, triggering a second warning instruction and generating a second strategy: remove the current target, update the electroacupuncture target candidate set Tcand, and re-execute the target generation and coupling analysis process until all target evaluations are completed.

[0013] Preferably, the neural response feedback adjustment module includes a comprehensive neural feedback acquisition unit, a third calculation unit, and a third analysis unit; The integrated neurofeedback acquisition unit is used for the implementation of treatment plans based on the effective target set Teff, and for real-time monitoring of neurofeedback data during continuous treatment cycles. It also acquires electrical activity signals from the patient's cerebral cortex in real time using an EEG electrode array, and employs time-series analysis and signal transformation techniques to obtain the rate of change in electrical activity in the amygdala region. Epilepsy waveform signals were acquired in real time using an EEG electrode array, and the power change rate of the epilepsy waves was obtained using wavelet transform and power spectrum analysis methods. Combining fMRI and PET scans, we monitored blood flow changes and metabolic activity in the anterior cingulate cortex and amygdala. Using dynamic causal modeling (DCM) and correlation modeling methods, we obtained the rate of change in the strength of the emotion regulation network. .

[0014] Preferably, the third calculation unit is used to obtain the rate of change of electrical activity through data acquisition. The rate of change of power of epileptic waves and the rate of change in the strength of the emotion regulation network After dimensionless normalization, the neural plasticity correction index (NPI) is calculated and obtained.

[0015] Preferably, the third analysis unit is used to preset the plasticity stabilization threshold Nth, and compare the neural plasticity correction index NPI with the plasticity stabilization threshold Nth to obtain the third evaluation result, including: When the neural plasticity correction index NPI is greater than or equal to the plasticity stability threshold Nth, it indicates that neural plasticity is stable under the current treatment state. The current treatment plan should be maintained and continuous monitoring should be carried out. When the neural plasticity correction index NPI is less than the plasticity stability threshold Nth, it indicates that the neural plasticity is unstable under the current treatment state, triggering the third warning instruction and generating the third strategy: automatically executing parameter correction, increasing the intensity of low-frequency stimulation; adjusting the waveform to a biphasic decreasing mode; extending the stimulation period and reducing the overlap rate of adjacent target points; and simultaneously updating the neural response database and the prediction model weight matrix to form an adaptive feedback closed loop.

[0016] This invention provides a target assessment system for epilepsy electroacupuncture treatment based on amygdala neural response prediction. It has the following beneficial effects: (1) This epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction achieves comprehensive monitoring of brain region electrical activity, blood oxygen metabolism and synaptic response through multimodal synchronous acquisition and fusion processing of EEG, fMRI and PET, combined with a high-resolution neural response sensor array; after time synchronization, spatial registration and functional connectivity analysis, a standardized neural response dataset and brain region functional mapping matrix are constructed to provide a unified and accurate quantitative basis for subsequent neural prediction and target assessment.

[0017] (2) This epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction extracts multidimensional features such as phase lock value (PLV), power spectral density (PSD), abnormal synchronization ratio (ASR), emotion regulation connection strength (EAC), and situational perturbation intensity (Cctx) to construct the amygdala predictive response index (ARI), thereby realizing the quantitative prediction of amygdala neural response under electroacupuncture stimulation. The model can dynamically adjust stimulation parameters until the response reaches the target, thereby significantly improving the accuracy and individual matching degree of target selection.

[0018] (3) This epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction establishes a spatial clustering model of electrical stimulation response through the brain region functional mapping matrix Mbrain and the comprehensive parameter set Mtarget, and extracts the neural response propagation delay. amygdala-thalamus-cortex co-coupling coefficient Epilepsy wavefront retrospective time Key neurodynamic parameters, calculating the comprehensive target evaluation coefficient This mechanism enables the quantification and multi-level assessment of target effects, effectively screening out preferred targets with stable functional connectivity and high therapeutic potential.

[0019] (4) This epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction monitors the rate of change in electrical activity in real time. The rate of change of power of epileptic waves and the rate of change in the strength of the emotion regulation network The system calculates the neural plasticity correction index (NPI) and compares it with the plasticity stability threshold (Nth) to achieve dynamic evaluation of the treatment status. When plasticity is unstable, the system automatically performs stimulation parameter correction and model weight update, forming an adaptive closed-loop control based on neural feedback, which effectively improves the safety, stability and individual response adaptation of the electroacupuncture treatment process. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a target assessment system for epilepsy electroacupuncture treatment based on amygdala neural response prediction, as described in this invention. Detailed Implementation

[0021] 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.

[0022] Example 1: Please refer to Figure 1This invention provides a target assessment system for epilepsy electroacupuncture treatment based on amygdala neural response prediction, comprising: The neural signal acquisition module is used to perform multimodal synchronous monitoring and preprocessing of the patient's brain neural activity under electroacupuncture stimulation: it acquires brain region potential changes and frequency band energy through a high-density EEG electrode array, blood oxygenation changes and metabolic activity through fMRI and PET, and local neural potentials and synaptic response timing through a high-resolution sensor array; after preprocessing, it forms a standardized neural response dataset and constructs a brain region functional mapping matrix that reflects the correlation coefficient and connection weight of signals between brain regions; The amygdala neural response prediction module is used to extract multimodal neural feature parameters based on a standardized neural response dataset, including phase lock value (PLV), power spectral density (PSD), anomalous synchronization ratio (ASR), emotion regulation connectivity strength (EAC), and situational perturbation intensity (Cctx). It calculates the amygdala predicted response index (ARI) and compares it with the amygdala response threshold (Ath) to determine whether the amygdala response meets the standard under the current stimulus parameters. If it does, a comprehensive parameter dataset is generated; if it does not, dynamic correction of the stimulus parameters is triggered until the response meets the standard. The neural effect coupling analysis module is used to cluster and spatially reconstruct electrical stimulation response features based on a comprehensive parameter set and a brain region function mapping matrix, generating a candidate set of electroacupuncture targets; for each candidate target, the neural response propagation delay is extracted. amygdala-thalamus-cortex co-coupling coefficient Epilepsy wavefront retrospective time Calculate the comprehensive target evaluation coefficient The target is compared with the target validity threshold Tth. If a target is deemed valid, it is included in the set of valid targets. If a target is deemed invalid, it is removed and the analysis iteration is re-executed. The neural response feedback adjustment module is used to collect neural feedback data in real time during continuous treatment cycles based on an effective target set, and to acquire the rate of change in electrical activity. The rate of change of power of epileptic waves and the rate of change in the strength of the emotion regulation network The neural plasticity correction index (NPI) is calculated and compared with the plasticity stability threshold (Nth) to determine whether neural plasticity is stable under the current treatment state. If it is stable, the current treatment plan is maintained; if it is unstable, feedback correction is triggered to dynamically adjust the stimulation parameters and model weights to form an adaptive treatment closed loop.

[0023] In this embodiment, a multi-module collaborative system consisting of neural signal acquisition, amygdala neural response prediction, neural effect coupling analysis, and neural response feedback adjustment is constructed to achieve closed-loop control of the entire process from multimodal neural signal acquisition to target prediction and evaluation, effect coupling analysis, and adaptive treatment feedback. This system can simultaneously monitor brain neural activity in multiple modalities under electroacupuncture stimulation, extract key neurodynamic parameters, and construct a brain region functional mapping matrix. By calculating the amygdala predicted response index and the comprehensive target evaluation coefficient, it accurately selects the optimal target. Simultaneously, combined with real-time feedback regulation of the neural plasticity correction index, it achieves individualized and adaptive dynamic optimization of stimulation parameters, effectively improving the accuracy of target localization, the stability of neural modulation, and the sustainability of therapeutic effects in epilepsy electroacupuncture treatment.

[0024] Example 2: This example is an explanation of Example 1. Please refer to the example provided. Figure 1 Specifically, the neural signal acquisition module includes an electroencephalogram (EEG) acquisition unit, a brain functional imaging unit, a neural electrical stimulation response detection unit, a data preprocessing unit, and a mapping matrix construction unit. The EEG acquisition unit is used to monitor the electrical activity state of the patient's cerebral cortex in real time; by deploying a high-density EEG electrode array on the scalp surface, it acquires brain region potential change signals and EEG frequency band energy distribution information; The brain functional imaging unit is used to simultaneously detect changes in brain blood oxygenation levels, metabolic activity, and blood flow distribution based on functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) devices, and to acquire fMRI blood oxygenation change signals, PET metabolic activity signals, and blood flow distribution data in the anterior cingulate cortex and amygdala regions. The neural electrical stimulation response detection unit is used to monitor local neural potential changes, synaptic response timing, and conduction path characteristics using a high-resolution neural response sensor array under electroacupuncture stimulation or simulated stimulation conditions. The data preprocessing unit is used to perform time synchronization, noise suppression and spatial registration on multi-source signals, eliminate signal offset caused by device delay and individual differences, and form a standardized neural response dataset. The mapping matrix construction unit is used to perform feature extraction and correlation modeling on EEG frequency band energy distribution information, fMRI blood oxygenation change signals and PET metabolic activity signals after data preprocessing, using cross-correlation analysis and functional connectivity strength estimation algorithms. It analyzes the signal correlation coefficients and functional connectivity weight matrix parameters of each brain region to construct the brain region functional mapping matrix Mbrain.

[0025] In this embodiment, a multimodal neural signal acquisition system was constructed by setting up an EEG acquisition unit, a brain functional imaging unit, a neural electrical stimulation response detection unit, a data preprocessing unit, and a mapping matrix construction unit. This system can achieve spatiotemporal synchronous acquisition and feature fusion of EEG signals, electrophysiological signals, and brain functional imaging signals under electroacupuncture stimulation. Through data preprocessing and functional connectivity modeling, a brain region functional mapping matrix (Mbrain) reflecting the signal correlation and functional coupling characteristics between brain regions is formed. This enables precise characterization and quantitative expression of the neural activity state of brain regions, providing highly consistent and high-resolution data support for subsequent amygdala response prediction and target assessment, effectively improving the accuracy and repeatability of neural modulation in electroacupuncture treatment of epilepsy.

[0026] Example 3: This example is an explanation of Example 1. Please refer to the provided text. Figure 1 Specifically, the amygdala neural response prediction module includes a neural response parameter extraction unit, a first calculation unit, and a first analysis unit; The neural response parameter extraction unit is used to perform feature decoding on brain region potential change signals based on a standardized neural response dataset. It establishes a mapping relationship between the electrical stimulation input signal and the amygdala response signal using a trained deep temporal network model, and extracts the phase-locked value (PLV), a characteristic parameter of epileptic abnormal synchronicity. For EEG frequency band energy distribution information, it uses power spectral decomposition and multi-scale wavelet analysis to calculate brain region discharge energy distribution characteristics and obtain the power spectral density parameter (PSD). Based on fMRI blood oxygenation change signals and PET metabolic activity signals, it uses correlation modeling and functional connectivity analysis algorithms to perform temporal correlation analysis on the synchronous response patterns between the amygdala and thalamus, obtaining the abnormal synchronization ratio (ASR). For blood flow distribution data in the anterior cingulate cortex and amygdala region, it uses a functional connectivity strength estimation algorithm to calculate the emotion regulation connectivity strength (EAC). For local neural potential changes, synaptic response timing, and conduction path characteristics, it uses synaptic temporal-dependent plasticity (STDP) modeling and neural pathway perturbation analysis methods to obtain the situational perturbation strength (Cctx).

[0027] In this embodiment, an amygdala response prediction mechanism based on multimodal neural signals is constructed by setting up a neural response parameter extraction unit, a first calculation unit, and a first analysis unit. This module can integrate multi-source data from EEG, fMRI, and PET to jointly model brain region discharge characteristics, blood oxygenation changes, and metabolic activity, and extract key parameters such as phase-locked value (PLV), power spectral density (PSD), abnormal synchronization ratio (ASR), emotion regulation connectivity strength (EAC), and situational perturbation intensity (Cctx). Through deep temporal networks and functional connectivity analysis algorithms, a nonlinear mapping between electrical stimulation input and amygdala response is achieved, effectively revealing the dynamic characteristics of abnormal synchronous activity in epilepsy. This significantly improves the accuracy and timeliness of amygdala response prediction, providing a reliable quantitative basis for subsequent target assessment and stimulation parameter optimization.

[0028] Example 4: This example is an explanation of Example 3. Please refer to the example provided. Figure 1 Specifically, the first calculation unit is used to construct the amygdala prediction index ARI by using the acquired epilepsy abnormal synchronicity characteristic parameters, including phase lock value (PLV), power spectral density parameter (PSD), abnormal synchronization ratio (ASR), emotion regulation connectivity strength (EAC), and situational perturbation strength (Cctx), after dimensionless normalization. The formula is as follows: ; In the formula, a1, a2, a3, a4 and a5 represent weighting coefficients; The phase-locked value (PLV) characterizes the impact of the amygdala predictive response index. It has a high weight and is a key indicator that directly reflects the contribution of abnormal discharge synchronization to the amygdala's excitation state. The power spectral density (PSD) characterizes the effect of the amygdala predicted response index, accounting for a moderate weight, and reflects the modulation effect of the discharge energy distribution on the intensity of the amygdala response. The anomalous synchronization ratio (ASR) represents the influence of the amygdala predictive response index, and has a moderate weight, reflecting the degree of disturbance of the stability of the amygdala response caused by thalamic-cortical synchronization abnormalities. The influence of Emotion Regulation Connection Strength (EAC) on the amygdala predictive response index is represented by a moderate weight, reflecting the buffering effect of the emotional network's regulatory capacity on amygdala abnormal activation. The influence of situational perturbation intensity Cctx on the amygdala predictive response index is characterized by a minor weight, reflecting the inhibitory effect of external environmental stimuli or cognitive load on the amygdala response threshold.

[0029] By constructing an amygdala predictive response index that is a weighted fusion of phase synchronization, power spectral energy, abnormal synchronization ratio, emotional regulation intensity, and situational perturbation intensity, the abnormal response trend of individual neural activity under different stimulus states can be quantitatively characterized, providing a scientific basis for epileptic seizure prediction and individualized electrical stimulation parameter setting.

[0030] The first analysis unit is used to obtain a first evaluation result by comparing the amygdala predicted index (ARI) with the amygdala response threshold (Ath) using a preset amygdala response threshold (Ath): When the amygdala predicts the corresponding index ARI ≥ amygdala response threshold Ath, it indicates that the amygdala response meets the target under the current stimulation parameters. A preliminary target assessment model is automatically generated, and a comprehensive parameter set Mtarget including local neural potential characteristics, synaptic response coupling strength and pathway activation weight is recorded. When the amygdala predicts the corresponding index ARI < amygdala response threshold Ath, it indicates that the amygdala response under the current stimulation parameters has not met the standard, triggering the first warning instruction and generating the first strategy: dynamically reducing the step size of the electroacupuncture stimulation frequency, adjusting the pulse width and waveform duty cycle, and combining the optimal parameter range of similar individuals in historical data to generate a set of corrected stimulation parameters, re-collecting data, and recalculating until the amygdala predicts the corresponding index ARI ≥ amygdala response threshold Ath.

[0031] The amygdala response threshold (Ath) was obtained by statistically analyzing EEG and fMRI data from a large number of epilepsy patients and healthy controls. The amplitude changes in electrical activity and the distribution range of blood flow response amplitude in the amygdala region under electrical stimulation or emotionally induced tasks were extracted. Combining the characteristics of neural signal propagation with the physiological safety limits of electroacupuncture stimulation, the stability and reversibility of the amygdala's response under different stimulation intensities were comprehensively analyzed. Referring to neuromodulation safety guidelines, quantitative analysis standards for medical imaging, and the experience of clinical experts, a reasonable amygdala response threshold (Ath) was finally determined to accurately reflect the sensitivity of the amygdala to electroacupuncture stimulation, enabling early identification of abnormal excitation responses and ensuring the safety of neuromodulation.

[0032] In this embodiment, a dynamic calculation and adaptive judgment mechanism for the amygdala predictive response index (ARI) is realized by setting up a first calculation unit and a first analysis unit. This module normalizes and fuses multimodal neural feature parameters to construct a quantitative response index, and compares it with a preset threshold Ath in real time to accurately determine the response state of the amygdala under the current electroacupuncture stimulation parameters. When the response is insufficient, the system can automatically generate a parameter correction strategy, dynamically adjust the stimulation frequency, pulse width, and waveform duty cycle, forming a closed-loop optimization process. This can effectively improve the individualized adaptability of electroacupuncture stimulation and the accuracy of target response matching, and significantly improve the stability and efficacy consistency of amygdala regulation during epilepsy treatment.

[0033] Example 5: This example is an explanation of Example 1. Please refer to the example provided. Figure 1 Specifically, the neural effect coupling analysis module includes a target generation unit, a target neural dynamic parameter extraction unit, a second calculation unit, and a second analysis unit. The target generation unit is used to construct an electrical stimulation response distribution model by performing cluster analysis and spatial projection reconstruction on neural response patterns under different stimulation parameters based on the comprehensive parameter set Mtarget combined with the brain region functional mapping matrix Mbrain, using multidimensional feature clustering and spatial correlation modeling techniques. Through the model, the response characteristics of the amygdala and its functional connectivity regions are spatially mapped and path-weighted analyzed to generate an electroacupuncture target candidate set Tcand, and the amygdala functional connectivity region, neural conduction path and response intensity index information corresponding to each target are recorded.

[0034] In this embodiment, by setting up a target generation unit and utilizing joint modeling of the comprehensive parameter set Mtarget and the brain region functional mapping matrix Mbrain, clustering analysis and spatial reconstruction of multidimensional neural features are achieved. This design can accurately map and perform path-weighted analysis on the response patterns of the amygdala and its functional connectivity regions under different stimulation parameter conditions, thereby generating an electroacupuncture target candidate set Tcand. This process can significantly improve the spatial resolution and functional specificity of target screening, realizing the mapping association from macroscopic brain region function to microscopic stimulation response, and providing a high-precision spatial basis for subsequent target evaluation and optimization.

[0035] Example 6: This example is an explanation of Example 5. Please refer to the example provided. Figure 1 Specifically, the target neurodynamic parameter extraction unit is used to calculate key neural effect parameters for each target t in the electroacupuncture target candidate set Tcand, based on the temporal electrophysiological characteristics and functional magnetic resonance dynamic signal change information in the standardized neural response dataset. Specifically, this includes: based on brain region potential change signals and EEG frequency band energy distribution information, using cross-correlation time delay analysis and phase synchronization tracking methods to estimate the time delay and phase align the temporal characteristics of brain region potential change signals, extracting the average signal transmission time difference between the amygdala and the target brain region, and obtaining the neural response propagation delay. Based on local neural potential changes, synaptic response timing, and conduction pathway characteristics, and combined with the brain region functional mapping matrix Mbrain, this study employs dynamic causal modeling (DCM) and partial minimum correlation analysis (PLSC) to jointly model and perform directional coupling analysis on local neural potential changes and the functional connectivity weights shown in the Mbrain mapping matrix. This quantifies the functional connectivity strength and directional coupling characteristics between the amygdala, thalamus, and cortex, obtaining the amygdala-thalamus-cortex co-coupling coefficient. Based on a standardized neural response dataset, a cross-modal time calibration and abnormal potential event detection method was used to synchronously label brain region potential change signals and local neural potential changes in the standardized neural response dataset. The amygdala abnormal discharge detection time (tabn) and epileptic seizure onset time (tonset) were obtained, and the epileptic wavefront retrospective time was further obtained. .

[0036] In this embodiment, by setting up a target neurodynamic parameter extraction unit, and utilizing the joint analysis of temporal electrophysiological characteristics and functional magnetic resonance imaging (fMRI) dynamic signals, key dynamic parameters such as neural response propagation delay, amygdala-thalamus-cortex synergistic coupling coefficient, and epileptic wavefront latency can be simultaneously quantified. This design achieves a multidimensional characterization of neural activity from the temporal, spatial, and directional connectivity levels, accurately revealing the neural conduction pathways and functional coupling patterns between target points. This provides a scientific basis for the quantitative assessment of electroacupuncture targets and the optimization of individualized treatment plans, significantly improving the accuracy and reliability of target assessment.

[0037] Example 7: This example is an explanation of Example 5. Please refer to the example provided. Figure 1 Specifically, the second computing unit is used to obtain the propagation delay of the neural response. amygdala-thalamus-cortex co-coupling coefficient Epilepsy wavefront retrospective time After dimensionless normalization, the comprehensive target evaluation coefficient is calculated and obtained. The formula is as follows: ; In the formula, w1 and w2 represent weight coefficients, and z1 and z2 represent weight coefficients; Characterizing the wavefront retrospective time in epilepsy The impact on the target assessment results carries a high weight and is a key indicator that directly reflects the early response capability of the target area in the early stage of the disease, which is of great significance for early intervention. Characterizing the delay in neural response propagation and the amygdala-thalamus-cortex co-coupling coefficient The impact on target assessment results has the second highest weight, reflecting the synergy of signal propagation between neural circuits and the effectiveness of target regulation; : Characterizes the influence of the local magnification correction coefficient on the cooperative coupling strength, and has a high weight, reflecting the weight correction ratio of the coupling characteristics of the target region under different neural pathways; This characterizes the impact of neural response propagation delay on target assessment results, and has a minor weight, reflecting the constraint effect of target response delay on neural signal transmission efficiency and control timing. By constructing a comprehensive target evaluation coefficient that is weighted by wavefront advance, synergistic coupling strength, and propagation delay, the regulatory potential of the target region in multi-pathway signal propagation can be quantified, providing a decision-making basis for the dynamic screening and optimization of effective therapeutic targets.

[0038] The second analysis unit is used to pass a preset target effectiveness threshold Tth and integrate the comprehensive target evaluation coefficient. A comparative analysis was performed with the target effectiveness threshold Tth to obtain the second evaluation results, including: When the comprehensive target evaluation coefficient When the value is greater than or equal to the target effectiveness threshold Tth, it indicates that the current target's neural effect is qualified, and the current target is determined to be an effective target and included in the effective target set Teff for continuous monitoring. When the comprehensive target evaluation coefficient When the value is less than the target effectiveness threshold Tth, it indicates that the current target's neural effect is unqualified. The current target is determined to be an invalid target, triggering a second warning instruction and generating a second strategy: remove the current target, update the electroacupuncture target candidate set Tcand, and re-execute the target generation and coupling analysis process until all target evaluations are completed.

[0039] The target effectiveness threshold Tth was obtained by systematically statistically analyzing and modeling the distribution of neural effect parameters from multiple patients under electroacupuncture stimulation. Typical distribution intervals of the comprehensive target evaluation coefficient were extracted by analyzing the functional improvement rate, changes in epileptic seizure frequency, and the degree of recovery of the emotion regulation network after stimulation of different targets. Combining the stability constraints of the neural function remodeling model and the clinical reproducibility of target responses, and referencing the effect evaluation standards in the field of brain function regulation and the target identification experience of clinicians, a reasonable target effectiveness threshold was determined to accurately distinguish between qualified and unqualified targets, thereby improving the scientific rigor and reliability of target selection.

[0040] In this embodiment, a comprehensive target evaluation coefficient is introduced. The computational mechanism enables the quantitative identification and dynamic optimization of electroacupuncture treatment targets for epilepsy. This method comprehensively considers key parameters such as neural response propagation delay, synergistic coupling strength, and epileptic wavefront retrospective time. It utilizes dimensionless normalization and weighted calculation to achieve multi-index fusion evaluation, which not only improves the objectivity and accuracy of target evaluation, but also automatically identifies and eliminates invalid targets by comparing them with the target effectiveness threshold Tth, dynamically updates the candidate set, and forms a closed-loop optimization process, thereby significantly improving the certainty of electroacupuncture treatment targets and the precision of neural modulation.

[0041] Example 8: This example is an explanation of Example 1. Please refer to the example provided. Figure 1 Specifically, the neural response feedback adjustment module includes a comprehensive neural feedback acquisition unit, a third calculation unit, and a third analysis unit; The integrated neurofeedback acquisition unit is used for the implementation of treatment plans based on the effective target set Teff, and for real-time monitoring of neurofeedback data during continuous treatment cycles. It also acquires electrical activity signals from the patient's cerebral cortex in real time using an EEG electrode array, and employs time-series analysis and signal transformation techniques to obtain the rate of change in electrical activity in the amygdala region. Epilepsy waveform signals were acquired in real time using an EEG electrode array, and the power change rate of the epilepsy waves was obtained using wavelet transform and power spectrum analysis methods. Combining fMRI and PET scans, we monitored blood flow changes and metabolic activity in the anterior cingulate cortex and amygdala. Using dynamic causal modeling (DCM) and correlation modeling methods, we obtained the rate of change in the strength of the emotion regulation network. .

[0042] In this embodiment, a neural response feedback adjustment module was implemented to achieve dynamic monitoring and adaptive regulation of the patient's brain neural activity during electroacupuncture treatment. This module integrates multimodal data from EEG, fMRI, and PET, extracting key indicators such as the rate of change in amygdala electrical activity, the rate of change in epileptic wave power, and the rate of change in the strength of the emotion regulation network, constructing a high-dimensional neural feedback dataset to provide continuous quantitative evidence for the therapeutic effect of the target. Through real-time feedback and modeling analysis, the neuromodulation effect can be dynamically evaluated during treatment, and stimulation parameters can be adjusted in a timely manner to achieve individualized and precise electroacupuncture treatment strategies, thereby significantly improving the stability of therapeutic effects and the efficiency of neurological function recovery.

[0043] Example 9: This example is an explanation of Example 8. Please refer to the example provided. Figure 1 Specifically, the third calculation unit is used to acquire the rate of change of electrical activity. The rate of change of power of epileptic waves and the rate of change in the strength of the emotion regulation network After dimensionless normalization, the neural plasticity correction index (NPI) is calculated using the following formula: ;

[0044] In the formula, s1, s2 and s3 represent weighting coefficients.

[0045] The amygdala electrical activity rate is a key indicator that characterizes the impact of changes in amygdala electrical activity on the neural plasticity correction index. It has a high weight and directly reflects the contribution of changes in neural discharge intensity to the trend of plasticity remodeling. The epileptic wave power change rate characterizes the effect of the neural plasticity correction index, accounting for a medium weight, and reflects the regulatory effect of abnormal wave energy decay or enhancement on neural stability. This index represents the impact of the rate of change in the strength of the emotion regulation network on the neural plasticity correction index, and has the second highest weight, reflecting the supporting role of changes in emotion network activity in the recovery of neural plasticity. By constructing a neural plasticity correction index that is a weighted fusion of the rate of change of electrical activity, the rate of change of power spectrum, and the rate of change of emotional network strength, we can comprehensively assess the neurological functional adaptability and stability of patients during continuous treatment cycles, providing a basis for optimizing individualized stimulation strategies and controlling the closed-loop feedback of therapeutic effects.

[0046] In this embodiment, by setting up a third calculation unit and introducing a computational mechanism for the Neuroplasticity Modified Index (NPI), a quantitative assessment of the degree of functional recovery in brain regions after electroacupuncture stimulation is achieved. This index comprehensively considers three types of dynamic parameters: enhanced amygdala electrical activity, suppression of epileptic wave power, and increased strength of the emotion regulation network. After dimensionless normalization and weighted fusion processing, it accurately reflects the trend of plasticity changes in the nervous system during continuous treatment cycles. This allows for precise tracking of the neural function reconstruction process, providing a scientific basis for optimizing electroacupuncture parameters and adjusting individualized treatment courses, thereby improving the targeting and duration of neuromodulation.

[0047] Example 10: This example is an explanation of Example 8. Please refer to the example provided. Figure 1 Specifically, the third analysis unit is used to preset the plasticity stabilization threshold Nth, and compare the neural plasticity correction index NPI with the plasticity stabilization threshold Nth to obtain the third evaluation result, including: When the neural plasticity correction index NPI is greater than or equal to the plasticity stability threshold Nth, it indicates that neural plasticity is stable under the current treatment state. The current treatment plan should be maintained and continuous monitoring should be carried out. When the neural plasticity correction index NPI is less than the plasticity stability threshold Nth, it indicates that the neural plasticity is unstable under the current treatment state, triggering the third warning instruction and generating the third strategy: automatically executing parameter correction, increasing the intensity of low-frequency stimulation; adjusting the waveform to a biphasic decreasing mode; extending the stimulation period and reducing the overlap rate of adjacent target points; and simultaneously updating the neural response database and the prediction model weight matrix to form an adaptive feedback closed loop.

[0048] The method for obtaining the plasticity stability threshold Nth: Based on longitudinal analysis of neurofeedback data within consecutive treatment cycles, the changing trend and steady-state distribution range of the neuroplasticity correction index (NPI) are extracted. By statistically analyzing the long-term fluctuation characteristics of the rate of change in emotion regulation network strength, the rate of change in epileptic wave power, and the rate of change in amygdala electrical activity in different patients, their correlation with treatment efficacy, seizure control rate, and neural network reconstruction stability is analyzed. Referring to clinical evaluation criteria for neuroplasticity regulation, steady-state response thresholds for neurofeedback therapy, and the practical experience of neurorehabilitation experts, this threshold is formulated to accurately reflect the stability of the neuroplasticity state, achieving adaptive parameter correction during treatment and maintaining long-term efficacy.

[0049] In this embodiment, by setting a third analysis unit and introducing a plasticity stability threshold Nth, dynamic discrimination and adaptive control of the neural plasticity state during treatment are achieved. When the neural plasticity correction index NPI is lower than the threshold, the system can automatically trigger a parameter correction strategy to intelligently adjust the stimulation intensity, waveform pattern, and stimulation cycle, and simultaneously update the neural response database and model weight matrix, forming a closed-loop optimization process. This design significantly improves the adaptability and safety of electroacupuncture therapy, enabling the system to maintain the optimal stimulation state in real time during fluctuations in neural plasticity, promoting stable remodeling of neural function and long-term efficacy maintenance.

[0050] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value, it is acceptable.

[0051] The above formulas are all derived from software simulation using a large amount of data and are selected to be close to the actual values. The coefficients in the formulas are set by those skilled in the art according to the actual situation. The above description is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any equivalent substitutions or changes made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the protection scope of the present invention.

Claims

1. An evaluation system for electroacupuncture treatment target of epilepsy based on amygdala neural response prediction, characterized in that, include: The neural signal acquisition module is used to perform multimodal synchronous monitoring and preprocessing of the patient's brain neural activity under electroacupuncture stimulation: it acquires brain region potential changes and frequency band energy through a high-density EEG electrode array, acquires blood oxygen changes and metabolic activity through fMRI and PET, and acquires local neural potentials and synaptic response timing through a high-resolution sensor array. After preprocessing, a standardized neural response dataset is formed, and a brain region functional mapping matrix reflecting the correlation coefficient and connection weight of signals between brain regions is constructed. The amygdala neural response prediction module is used to extract multimodal neural feature parameters based on a standardized neural response dataset, including phase lock value (PLV), power spectral density (PSD), anomalous synchronization ratio (ASR), emotion regulation connectivity strength (EAC), and situational perturbation intensity (Cctx). It calculates the amygdala predicted response index (ARI) and compares it with the amygdala response threshold (Ath) to determine whether the amygdala response meets the standard under the current stimulus parameters. If it does, a comprehensive parameter dataset is generated; if it does not, dynamic correction of the stimulus parameters is triggered until the response meets the standard. a neural effect coupling analysis module for clustering and spatially reconstructing the electrical stimulation response features based on the comprehensive parameter set and the brain region function mapping matrix to generate a set of candidate target points for electrical acupuncture; for each candidate target point, extracting the neural response propagation delay , the amygdala-thalamus-cortex synergistic coupling coefficient and the epilepsy wave front tracing time , calculating a comprehensive target point evaluation coefficient , comparing the comprehensive target point evaluation coefficient with a target point effectiveness threshold Tth, determining the target point as effective and including the target point in the effective target point set, or determining the target point as ineffective and excluding the target point and re-executing the analysis iteration; a neural response feedback adjustment module, configured to collect neural feedback data in real time based on the effective target point set in a continuous treatment cycle, and calculate a change rate of electrical activity , a power change rate of epileptic waves , and a change rate of the strength of the emotional regulation network , calculate a neural plasticity correction index NPI, and compare the NPI with a plasticity stability threshold Nth to determine whether the neural plasticity is stable under the current treatment state. If the neural plasticity is stable, the current treatment plan is maintained. If the neural plasticity is unstable, a feedback correction is triggered to dynamically adjust the stimulation parameters and the model weights, thereby forming an adaptive treatment closed loop.

2. The system for evaluating the target point of electroacupuncture treatment for epilepsy based on the prediction of amygdala neural response according to claim 1, wherein The neural signal acquisition module includes an electroencephalogram (EEG) acquisition unit, a brain functional imaging unit, a neural electrical stimulation response detection unit, a data preprocessing unit, and a mapping matrix construction unit. The EEG acquisition unit is used to monitor the electrical activity state of the patient's cerebral cortex in real time; By deploying a high-density EEG electrode array on the scalp surface, brain region potential change signals and brainwave frequency band energy distribution information are collected. The brain functional imaging unit is used to simultaneously detect changes in brain blood oxygenation levels, metabolic activity, and blood flow distribution based on functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) devices, and to acquire fMRI blood oxygenation change signals, PET metabolic activity signals, and blood flow distribution data in the anterior cingulate cortex and amygdala regions. The neural electrical stimulation response detection unit is used to monitor local neural potential changes, synaptic response timing, and conduction path characteristics using a high-resolution neural response sensor array under electroacupuncture stimulation or simulated stimulation conditions. The data preprocessing unit is used to perform time synchronization, noise suppression and spatial registration on multi-source signals, eliminate signal offset caused by device delay and individual differences, and form a standardized neural response dataset. The mapping matrix construction unit is used to perform feature extraction and correlation modeling on EEG frequency band energy distribution information, fMRI blood oxygenation change signals and PET metabolic activity signals after data preprocessing, using cross-correlation analysis and functional connectivity strength estimation algorithms. It analyzes the signal correlation coefficients and functional connectivity weight matrix parameters of each brain region to construct the brain region functional mapping matrix Mbrain.

3. The epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction according to claim 1, characterized in that, The amygdala neural response prediction module includes a neural response parameter extraction unit, a first calculation unit, and a first analysis unit; The neural response parameter extraction unit is used to perform feature decoding on brain region potential change signals based on a standardized neural response dataset. It establishes a mapping relationship between the electrical stimulation input signal and the amygdala response signal using a trained deep temporal network model, and extracts the phase-locked value (PLV), a characteristic parameter of epileptic abnormal synchronicity. For EEG frequency band energy distribution information, it uses power spectral decomposition and multi-scale wavelet analysis to calculate brain region discharge energy distribution characteristics and obtain the power spectral density parameter (PSD). Based on fMRI blood oxygenation change signals and PET metabolic activity signals, it uses correlation modeling and functional connectivity analysis algorithms to perform temporal correlation analysis on the synchronous response patterns between the amygdala and thalamus, obtaining the abnormal synchronization ratio (ASR). For blood flow distribution data in the anterior cingulate cortex and amygdala region, it uses a functional connectivity strength estimation algorithm to calculate the emotion regulation connectivity strength (EAC). For local neural potential changes, synaptic response timing, and conduction path characteristics, it uses synaptic temporal-dependent plasticity (STDP) modeling and neural pathway perturbation analysis methods to obtain the situational perturbation strength (Cctx).

4. The epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction according to claim 3, characterized in that, The first calculation unit is used to construct the amygdala prediction index ARI by using the acquired epilepsy abnormal synchronization characteristic parameters, such as phase lock value PLV, power spectral density parameter PSD, abnormal synchronization ratio ASR, emotion regulation connection strength EAC, and situational perturbation strength Cctx, after dimensionless normalization. The first analysis unit is used to obtain a first evaluation result by comparing the amygdala predicted index (ARI) with the amygdala response threshold (Ath) using a preset amygdala response threshold (Ath): When the amygdala predicts the corresponding index ARI ≥ amygdala response threshold Ath, it indicates that the amygdala response meets the target under the current stimulation parameters. A preliminary target assessment model is automatically generated, and a comprehensive parameter set Mtarget including local neural potential characteristics, synaptic response coupling strength and pathway activation weight is recorded. When the amygdala predicts the corresponding index ARI < amygdala response threshold Ath, it indicates that the amygdala response under the current stimulation parameters has not met the standard, triggering the first warning instruction and generating the first strategy: dynamically reducing the step size of the electroacupuncture stimulation frequency, adjusting the pulse width and waveform duty cycle, and combining the optimal parameter range of similar individuals in historical data to generate a set of corrected stimulation parameters, re-collecting data, and recalculating until the amygdala predicts the corresponding index ARI ≥ amygdala response threshold Ath.

5. The epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction according to claim 1, characterized in that, The neural effect coupling analysis module includes a target generation unit, a target neural dynamic parameter extraction unit, a second calculation unit, and a second analysis unit. The target generation unit is used to construct an electrical stimulation response distribution model by performing cluster analysis and spatial projection reconstruction on neural response patterns under different stimulation parameters based on the comprehensive parameter set Mtarget combined with the brain region functional mapping matrix Mbrain, using multidimensional feature clustering and spatial correlation modeling techniques. Through the model, the response characteristics of the amygdala and its functional connectivity regions are spatially mapped and path-weighted analyzed to generate an electroacupuncture target candidate set Tcand, and the amygdala functional connectivity region, neural conduction path and response intensity index information corresponding to each target are recorded.

6. The epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction according to claim 5, characterized in that, The target neurodynamic parameter extraction unit is used to calculate key neural effect parameters for each target t in the electroacupuncture target candidate set Tcand, based on the temporal electrophysiological characteristics and functional magnetic resonance dynamic signal change information in the standardized neural response dataset. Specifically, this includes: based on brain region potential change signals and EEG frequency band energy distribution information, using cross-correlation time delay analysis and phase synchronization tracking methods, estimating the time delay and aligning the phase of the temporal characteristics of brain region potential change signals, extracting the average signal transmission time difference between the amygdala and the target brain region, and obtaining the neural response propagation delay. Based on local neural potential changes, synaptic response timing, and conduction pathway characteristics, and combined with the brain region functional mapping matrix Mbrain, this study employs dynamic causal modeling (DCM) and partial minimum correlation analysis (PLSC) to jointly model and perform directional coupling analysis on local neural potential changes and the functional connectivity weights shown in the Mbrain mapping matrix. This quantifies the functional connectivity strength and directional coupling characteristics between the amygdala, thalamus, and cortex, obtaining the amygdala-thalamus-cortex co-coupling coefficient. Based on a standardized neural response dataset, a cross-modal time calibration and abnormal potential event detection method was used to synchronously label brain region potential change signals and local neural potential changes in the standardized neural response dataset. The amygdala abnormal discharge detection time (tabn) and epileptic seizure onset time (tonset) were obtained, and the epileptic wavefront retrospective time was further obtained. .

7. The epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction according to claim 5, characterized in that, The second computing unit is used to obtain the propagation delay of the neural response. amygdala-thalamus-cortex co-coupling coefficient Epilepsy wavefront retrospective time After dimensionless normalization, the comprehensive target evaluation coefficient is calculated and obtained. ; The second analysis unit is used to pass a preset target effectiveness threshold Tth and integrate the comprehensive target evaluation coefficient. A comparative analysis was performed with the target effectiveness threshold Tth to obtain the second evaluation results, including: When the comprehensive target evaluation coefficient When the value is greater than or equal to the target effectiveness threshold Tth, it indicates that the current target's neural effect is qualified, and the current target is determined to be an effective target and included in the effective target set Teff for continuous monitoring. When the comprehensive target evaluation coefficient When the value is less than the target effectiveness threshold Tth, it indicates that the current target's neural effect is unqualified. The current target is determined to be an invalid target, triggering a second warning instruction and generating a second strategy: remove the current target, update the electroacupuncture target candidate set Tcand, and re-execute the target generation and coupling analysis process until all target evaluations are completed.

8. The epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction according to claim 1, characterized in that, The neural response feedback adjustment module includes a comprehensive neural feedback acquisition unit, a third calculation unit, and a third analysis unit; The integrated neurofeedback acquisition unit is used for the implementation of treatment plans based on the effective target set Teff, and for real-time monitoring of neurofeedback data during continuous treatment cycles. It also acquires electrical activity signals from the patient's cerebral cortex in real time using an EEG electrode array, and employs time-series analysis and signal transformation techniques to obtain the rate of change in electrical activity in the amygdala region. Epilepsy waveform signals were acquired in real time using an EEG electrode array, and the power change rate of the epilepsy waves was obtained using wavelet transform and power spectrum analysis methods. Combining fMRI and PET scans, we monitored blood flow changes and metabolic activity in the anterior cingulate cortex and amygdala. Using dynamic causal modeling (DCM) and correlation modeling methods, we obtained the rate of change in the strength of the emotion regulation network. .

9. The epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction according to claim 8, characterized in that, The third calculation unit is used to obtain the rate of change of electrical activity through data acquisition. The rate of change of power of epileptic waves and the rate of change in the strength of the emotion regulation network After dimensionless normalization, the neural plasticity correction index (NPI) is calculated and obtained.

10. The epilepsy electroacupuncture treatment target assessment system based on amygdala neural response prediction according to claim 8, characterized in that, The third analysis unit is used to preset the plasticity stabilization threshold Nth, and compare the neural plasticity correction index NPI with the plasticity stabilization threshold Nth to obtain the third evaluation results, including: When the neural plasticity correction index NPI is greater than or equal to the plasticity stability threshold Nth, it indicates that neural plasticity is stable under the current treatment state. The current treatment plan should be maintained and continuous monitoring should be carried out. When the neural plasticity correction index NPI is less than the plasticity stability threshold Nth, it indicates that the neural plasticity is unstable under the current treatment state, triggering the third warning instruction and generating the third strategy: automatically executing parameter correction, increasing the intensity of low-frequency stimulation; adjusting the waveform to a biphasic decreasing mode; extending the stimulation period and reducing the overlap rate of adjacent target points; and simultaneously updating the neural response database and the prediction model weight matrix to form an adaptive feedback closed loop.

Images