Brain stimulation closed-loop control device and parameter control instruction generation system, method, medium and product

Abstract

This invention discloses a brain stimulation closed-loop control device and parameter control command generation system, method, medium, and product, relating to the field of neuromodulation. It includes: an acquisition module for acquiring electroencephalogram (EEG) data of the target and stimulation circuit impedance; and a processing module for determining whether the target is in a first preset state based on the EEG data; if so, generating brain stimulation parameter control commands based on the EEG data and stimulation circuit impedance. This invention enables on-demand intervention and can counteract the influence of electrode environment fluctuations on electric field strength, ensuring the consistency of stimulation dose at deep target points.

Description

Technical Field

[0001] This invention relates to the field of neuromodulation technology, and in particular to a brain stimulation closed-loop modulation device and parameter control command generation system, method, medium and product. Background Technology

[0002] Existing non-invasive deep brain stimulation (TIS) in the treatment of Parkinson's disease mostly adopts an open-loop stimulation mode with constant frequency or timed activation. It cannot detect the real-time cyclical fluctuations of the beta bursts in the brain. As a result, the system continues to output high-intensity current during non-pathological outbreaks, which leads to premature tolerance of the nervous system and increases unnecessary energy dissipation.

[0003] Moreover, due to individual differences in the dominant frequency of pathological beta waves among different patients and at different stages of the disease, existing technologies using fixed frequency differences (such as a constant 20Hz) cannot accurately couple with the patient's current neural characteristic frequency, resulting in poor suppression of abnormal oscillations and limited treatment precision.

[0004] Furthermore, the impedance of the electrode-skin contact surface fluctuates in real time due to perspiration, displacement, and electrode polarization. Under the premise of constant output voltage, the actual current vector entering deep tissues will decrease as the impedance increases, resulting in unstable interference field strength at deep target points. This makes it difficult to maintain a constant treatment dose and may even cause safety hazards such as contact overheating. Summary of the Invention

[0005] To address the technical problems existing in the background art, the present invention proposes a brain stimulation closed-loop regulation device and a parameter control instruction generation system, method, medium, and product.

[0006] In a first aspect, the present invention proposes a brain stimulation parameter control command generation system based on real-time EEG feedback and impedance compensation, comprising: The acquisition module is used to acquire the target's electroencephalogram (EEG) data and stimulation circuit impedance; The processing module is used to determine whether the target is in a first preset state based on the EEG data; if so, it generates brain stimulation parameter control instructions based on the EEG data and stimulation circuit impedance.

[0007] Preferably, determining whether the target is in a first preset state based on electroencephalogram (EEG) data specifically includes: Based on the EEG data, the power spectral density of the EEG data in the β band was obtained; When the absolute value of the difference between the power spectral density and the target's preset power spectral density benchmark value is greater than a preset threshold and the duration exceeds a preset duration, the target is determined to be in the first preset state.

[0008] Preferably, before determining whether the target is in a first preset state based on the electroencephalogram (EEG) data, the method further includes: The EEG data were preprocessed, including artifact removal and bandpass filtering.

[0009] Preferably, based on EEG data and stimulation circuit impedance, brain stimulation parameter control instructions are generated, specifically including: The feature vector of the EEG signal in the β band is extracted from the power spectral density; wherein, the feature vector of the EEG signal in the β band includes the power increment reflecting the pathological degree and the center frequency reflecting the rhythm characteristics; Based on the stimulation circuit impedance and the preset initial stimulation circuit impedance, the impedance compensation coefficient used to characterize the fluctuation of the electrode environment is calculated. The power increment is input into a preset nonlinear physiological mapping model to obtain the baseline stimulus intensity; The gain of the basic stimulus intensity is corrected by using an impedance compensation coefficient to obtain the output amplitude of the stimulation current used to maintain a constant field strength at the deep target point. Map the center frequency to TIS differential frequency control parameters; The stimulation current output amplitude and TIS difference frequency control parameters are integrated and encapsulated to generate brain stimulation parameter control instructions.

[0010] Preferably, the amplitude of the stimulation current output is: ; ; ; ; In the formula, I out To stimulate the output amplitude of the current, This is the impedance compensation coefficient; To preset the initial stimulation circuit impedance, Let be the impedance of the stimulation circuit at time t; This represents the physiological mapping at time t, i.e., the base stimulus intensity output by the nonlinear physiological mapping model. The preset minimum effective current threshold, This is the preset gain sensitivity coefficient. For power increment; Let be the power spectral density at time t. To preset the power spectral density reference value, Let be the stimulus state regulation factor at time t.

[0011] Preferably, the center frequency is mapped to TIS differential frequency control parameters, specifically including: The center frequency at the current moment is weighted and averaged with the center frequencies of a preset number of historical moments, and the weighted average result is mapped to the TIS differential frequency control parameters.

[0012] In a second aspect, the present invention also proposes a brain stimulation closed-loop control device based on real-time EEG feedback and impedance compensation, comprising: a signal acquisition module, an impedance monitoring module, a TIS stimulation module, and a brain stimulation parameter control instruction generation system based on real-time EEG feedback and impedance compensation as described in any one of the first aspects. The signal acquisition module is used to acquire the target's electroencephalogram (EEG) data; the impedance monitoring module is used to obtain the target's stimulation circuit impedance; and the TIS stimulation module is used to perform non-invasive brain stimulation on the target according to the brain stimulation parameter control instructions.

[0013] Thirdly, this invention also proposes a method for generating brain stimulation parameter control commands based on real-time EEG feedback and impedance compensation, comprising: Acquire the target's electroencephalogram (EEG) data and stimulation circuit impedance; Based on the EEG data, determine whether the target is in the first preset state; if so, generate brain stimulation parameter control instructions based on the EEG data and stimulation circuit impedance.

[0014] Preferably, determining whether the target is in a first preset state based on electroencephalogram (EEG) data specifically includes: Based on the EEG data, the power spectral density of the EEG data in the β band was obtained; When the absolute value of the difference between the power spectral density and the target's preset power spectral density benchmark value is greater than a preset threshold and the duration exceeds a preset duration, the target is determined to be in the first preset state.

[0015] Preferably, the brain stimulation parameter control command is generated based on the EEG data and the stimulation circuit impedance, specifically including: extracting the feature vector of the EEG signal in the β band from the power spectral density; wherein, the feature vector of the EEG signal in the β band includes the power increment used to reflect the pathological degree and the center frequency used to reflect the rhythm characteristics; Based on the stimulation circuit impedance and the preset initial stimulation circuit impedance, the impedance compensation coefficient used to characterize the fluctuation of the electrode environment is calculated. The power increment is input into a preset nonlinear physiological mapping model to obtain the baseline stimulus intensity; The gain of the basic stimulus intensity is corrected by using an impedance compensation coefficient to obtain the output amplitude of the stimulation current used to maintain a constant field strength at the deep target point. Map the center frequency to TIS differential frequency control parameters; The stimulation current output amplitude and TIS difference frequency control parameters are integrated and encapsulated to generate brain stimulation parameter control instructions.

[0016] Fourthly, the present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the brain stimulation parameter control instruction generation method based on real-time EEG feedback and impedance compensation as described in any one of the third aspects.

[0017] Fifthly, the present invention also proposes a computer program product, comprising a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the brain stimulation parameter control instruction generation method based on real-time EEG feedback and impedance compensation as described in any one of the third aspects.

[0018] This invention proposes a closed-loop brain stimulation control device and parameter control command generation system, method, medium, and product. By acquiring the target's electroencephalogram (EEG) data and stimulation circuit impedance, and determining whether the target is in a first preset state based on the EEG data, if so, brain stimulation parameter control commands are generated based on the EEG data and stimulation circuit impedance. This invention triggers stimulation only during the pathological signal burst phase (i.e., the first preset state), achieving on-demand intervention and significantly delaying neural tolerance while reducing device power consumption. Furthermore, the introduction of impedance monitoring, generating brain stimulation parameter control commands based on EEG data and stimulation circuit impedance, can counteract the influence of electrode environment fluctuations on electric field strength, ensuring the consistency of deep target stimulation dosage and improving the safety of clinical use. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the parameter control instruction generation system in one embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the brain stimulation closed-loop regulation device in one embodiment of the present invention. Detailed Implementation

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] Firstly, such as Figure 1 As shown, the present invention proposes a brain stimulation parameter control instruction generation system based on real-time EEG feedback and impedance compensation, comprising: an acquisition module and a processing module; The acquisition module is used to acquire the target's electroencephalogram (EEG) data and stimulation circuit impedance; the processing module is used to determine whether the target is in a first preset state based on the EEG data; if so, it generates brain stimulation parameter control instructions based on the EEG data and stimulation circuit impedance.

[0023] It should be understood that the first preset state is a pathological beta burst state. Here, a pathological beta burst refers to the phenomenon that the power of the target in the beta band increases sharply in a short period of time and lasts for a certain period of time.

[0024] This invention acquires the target's electroencephalogram (EEG) data and the corresponding stimulation circuit impedance, and determines whether the target is in a first preset state based on the EEG data. If so, it generates brain stimulation parameter control instructions based on the EEG data and stimulation circuit impedance.

[0025] This invention triggers stimulation only during the pathological signal burst phase (i.e., the first preset state), enabling on-demand intervention and significantly delaying neural tolerance while reducing device power consumption. Furthermore, by introducing impedance monitoring, it generates brain stimulation parameter control commands based on EEG data and stimulation circuit impedance. This counteracts the impact of electrode environment fluctuations on electric field strength, ensuring consistent deep target stimulation dosage and improving the safety of clinical use.

[0026] It is important to understand that stimulation circuit impedance refers to the resistance between the TIS electrode that performs the stimulation and the target's skin. EEG data and stimulation circuit impedance are synchronized data to accurately determine the target's state.

[0027] In some embodiments, the acquisition of electroencephalogram (EEG) data and stimulation circuit impedance is performed at preset time intervals, which are set according to actual needs.

[0028] In other embodiments, acquiring EEG data specifically involves acquiring EEG data in real time. Therefore, acquiring the target's stimulation circuit impedance specifically involves acquiring the target's stimulation circuit impedance in real time, thereby enabling real-time adjustment of brain stimulation parameter control commands during brain stimulation, and ultimately achieving real-time control of brain stimulation.

[0029] In some embodiments, brain stimulation parameter control instructions include stimulation current output amplitude and TIS difference frequency control parameters.

[0030] The TIS frequency difference control parameter refers to the frequency difference (i.e., 20Hz) between two high-frequency carriers (such as 2000Hz and 2020Hz) in the TIS system. This TIS frequency difference control parameter is the effective treatment frequency that actually acts on the neurons.

[0031] In some embodiments, determining whether a target is in a first preset state based on electroencephalogram (EEG) data specifically includes: Based on the EEG data, the power spectral density of the EEG data in the β band was obtained; The power spectral density is compared with the target's preset power spectral density benchmark value. When the absolute value of the difference between the power spectral density and the preset power spectral density benchmark value is greater than a preset threshold and the duration exceeds a preset duration, the target is determined to be in the first preset state.

[0032] With this setup, this embodiment can accurately determine whether each target is in the first preset state, so that stimulation can be triggered only during the pathological signal outbreak stage, thus achieving on-demand intervention.

[0033] It is important to understand that the power spectral density of the β band is represented by an energy distribution curve within the 13-30Hz frequency range. During the state determination process, characteristic physical quantities (such as peak power spectral density or in-band integrated power) of this curve (the power spectral density of the β band) within the target frequency band are extracted, and a difference calculation is performed between this and a preset power spectral density benchmark value. When the absolute value of the difference calculation result is greater than a preset threshold and meets the time coherence requirement (i.e., the duration exceeds a preset duration), it is determined to be the first preset state.

[0034] In some further embodiments, the preset threshold is three times the standard deviation of a preset power spectral density reference value.

[0035] In some embodiments, a preset power spectral density reference value P base The preset process includes: Multiple EEG data of the target in a second preset state are acquired in advance; Artifact removal, bandpass filtering, and power spectral density calculation were performed on the second EEG data of the target under the second preset state, and the average value of the calculation results was used as the preset power spectral density reference value P of the target in the β band. base。

[0036] The second preset state is the physiological baseline recovery state, which refers to the stable state in which the target's EEG characteristics fall back to the non-pathological range.

[0037] The specific judgment criteria are as follows: when the absolute value of the difference between the power spectral density of the β band and the preset power spectral density benchmark value is less than or equal to the preset threshold, and the duration of this state exceeds the preset resting observation time, the target is judged to be in the second preset state.

[0038] In this embodiment, a preset power spectral density reference value P can be set for different targets. base Individualized settings for precise matching of targets.

[0039] To avoid introducing errors, in some embodiments, before determining whether the target is in a first preset state based on the EEG data, the method further includes: preprocessing the EEG data; wherein, the preprocessing includes artifact removal and bandpass filtering.

[0040] In some embodiments, the process of generating brain stimulation parameter control instructions specifically includes: First, the feature vector of the EEG signal in the β band is extracted from the power spectral density; wherein, the feature vector of the EEG signal in the β band includes the power increment that reflects the pathological degree and the center frequency that reflects the rhythm characteristics. Secondly, based on the obtained stimulation circuit impedance and the preset initial stimulation circuit impedance, the impedance compensation coefficient used to characterize the fluctuation of the electrode environment is calculated. Subsequently, the power increment is input into a preset nonlinear physiological mapping model to obtain the baseline stimulus intensity; Impedance compensation coefficients are used to correct the gain of the basic stimulation intensity, thereby calculating the stimulation current output amplitude that can maintain a constant field strength at the deep target point; at the same time, the center frequency is mapped to the TIS difference frequency control parameter. Finally, the stimulation current output amplitude and TIS difference frequency control parameters are integrated and encapsulated to generate closed-loop brain stimulation parameter control instructions.

[0041] This embodiment monitors the stimulation circuit impedance in real time, automatically calculates the impedance compensation coefficient, and uses this coefficient to correct the gain of the base stimulation intensity. This allows for the calculation of a stimulation current output amplitude that maintains a constant field strength at the deep target point, thus offsetting impedance fluctuations caused by skin perspiration and displacement during subsequent stimulation. This ensures the effective interference field strength at the deep target point remains at a constant therapeutic dose, improving the safety of clinical use. Furthermore, this embodiment sets the TIS difference frequency control parameter to the center frequency f. peak This allows the TIS difference frequency Δf of the two high-frequency carriers generated by the TIS stimulation module at the back end to be adjusted to the center frequency, achieving frequency following. It not only adjusts the "intensity" but also dynamically aligns the "frequency," so that the TIS difference frequency Δf changes synchronously with the patient's real-time pathological main frequency, thereby achieving efficient and resonant suppression of abnormal oscillations and significantly improving the suppression efficiency of abnormal nerve oscillations.

[0042] It should be understood that the process of determining the center frequency includes: searching for the maximum point of the power spectral density in the β band (13-30Hz), and the frequency corresponding to this maximum point is determined as the center frequency f. peak This is used to lock onto the dominant frequency of pathological oscillations in real time.

[0043] In some further embodiments, the process of obtaining the feature vector of the EEG signal in the β band is as follows: first, the EEG data is transformed in the frequency domain by the processing module to obtain the power spectral density; then, the power spectral density is identified to obtain the center frequency and power increment of the EEG signal in the β band, and the center frequency and power increment of the EEG signal in the β band are used as the feature vector of the EEG signal in the β band.

[0044] Of course, in other embodiments, the power spectral density during the first preset state judgment process is directly extracted as a feature vector to quickly obtain the feature vector of the EEG signal in the β band.

[0045] Specifically, the power increment is as follows: ; In the formula, ΔP is the power increment, which represents the power deviation of the β band, that is, the dynamic change of the real-time power spectral density relative to the preset power spectral density reference value. Let be the power spectral density at time t. This is the preset power spectral density reference value.

[0046] Therefore, in some embodiments, the processing module includes a preprocessing unit, a β feature mapping unit, and a control unit; The preprocessing unit is used to preprocess the EEG data; the preprocessing includes artifact removal and bandpass filtering. The β feature mapping unit is used to calculate the power spectral density of the preprocessed EEG data to obtain the power spectral density of the EEG data in the β band. The control unit is used to determine whether the target is in a first preset state based on the power spectral density; if so, it extracts the feature vector of the EEG signal in the β band from the power spectral density; wherein, the feature vector of the EEG signal in the β band includes the power increment and the center frequency; The power increment is input into a preset nonlinear physiological mapping model to obtain the baseline stimulus intensity; The impedance compensation coefficient is calculated based on the stimulation circuit impedance and the preset initial stimulation circuit impedance. The gain of the basic stimulus intensity is corrected by using an impedance compensation coefficient to obtain the stimulus current output amplitude. Map the center frequency to TIS differential frequency control parameters; The stimulation current output amplitude and TIS difference frequency control parameters are integrated and encapsulated to generate brain stimulation parameter control instructions.

[0047] To improve the smoothness of center frequency positioning, in some further embodiments, a weighted average can be performed on the center frequency at the current moment and the center frequencies at a preset number of historical moments, and the weighted average result can be mapped to the TIS differential frequency control parameter rate.

[0048] This embodiment introduces historical frequency points as constraints to eliminate frequency jumps caused by instantaneous noise, thereby obtaining smoothly evolving characteristic frequency values ​​and mapping them to TIS difference frequency control parameters to ensure dynamic synchronization between intervention frequency and pathological rhythm.

[0049] It's important to understand that TIS outputs a high-intensity electric field at the kilohertz level, which has a strong drowning effect on microsecond-level EEG signals. The key technical bottleneck in achieving closed-loop modulation is how to remove electrical interference and extract β-oscillation features in real time without distortion.

[0050] Since TIS stimulation is a continuous sine wave, its artifacts manifest as extremely strong narrowband high-frequency peaks in the frequency domain. To address this issue, in some embodiments, a band-stop filter module is connected between the acquisition module and the preprocessing module. The band-stop filter module is used to suppress common-mode and differential-mode interference caused by the high-intensity carrier electric field before the EEG data enters the preprocessing module, preventing signal saturation in the front-end analog link and providing an effective linear dynamic range for artifact removal in the subsequent preprocessing module.

[0051] In some embodiments, the artifact removal process includes: acquiring the stimulation-driven signal in the brain stimulation parameter control command and using it as a reference channel, and using an adaptive filter to remove artifacts from the electroencephalogram data based on the reference channel.

[0052] This embodiment uses the stimulation drive signal (measured current feedback at the output of the stimulation source) in the brain stimulation parameter control command as a reference channel. The coherent components are subtracted from the EEG data in real time through an adaptive filter, thereby ensuring that the changes in the β band wave can still be accurately monitored during stimulation.

[0053] It is important to understand that the stimulus-driven signal refers to the original control signal used to excite the output of the constant current stimulus source. Since this signal has a high temporal coherence with the electric field interference present in the EEG data, using it as the reference input of the adaptive filter can achieve precise removal of stimulus artifacts, thus solving the bottleneck of traditional algorithms being unable to handle physical nonlinear artifacts. In a further embodiment, the adaptive filter employs a step size update strategy to achieve dynamic response optimization, ensuring that it can track and adaptively adjust the convergence speed in real time during the dynamic switching of the TIS carrier frequency, thus overcoming the lag in response of the fixed step size algorithm.

[0054] The variable step size update strategy is specifically applied to the iterative update loop of the filter weights of the adaptive filter: by monitoring the instantaneous deviation between the current signal center frequency and the preset stimulus frequency, the step size factor is dynamically adjusted; a large step size is used at the moment of frequency switching to achieve sub-second rapid convergence, and a very small step size is used during the steady period to ensure the complete preservation of weak pathological signals in the β band, thereby achieving dynamic adaptive suppression of physical nonlinear artifacts.

[0055] During the iteration of the filter weights in the adaptive filter, a cost function weighting strategy based on the target frequency band is adopted to achieve precise frequency protection. This constructs a protection range for the core physiological characteristic range of the β band, preventing weak pathological outbreak signals from being misjudged as interference and filtered out during the strong artifact cancellation process.

[0056] Specifically, the power density characteristics of the residual signal in the target β frequency band are extracted in real time. This power density characteristic is mapped to a penalty factor in the cost function. When the attenuation of this frequency band exceeds the physiological safety threshold, the penalty term is increased to limit further updates of the filter weights. This allows the adaptive filter to form targeted gating protection in the frequency domain, ensuring that the high-intensity artifact cancellation action only acts on non-physiological feature areas, thereby completely preserving the weak pathological outbreak signal and providing accurate triggering criteria for subsequent closed-loop intervention.

[0057] In some embodiments, the amplitude of the stimulation current output is: ; ; ; ; In other words, ; In the formula, I out To stimulate the output amplitude of the current, This is the impedance compensation coefficient. When the electrode contact deteriorates (impedance increases), this term decreases, and the stimulation current output amplitude I is maintained by increasing the driving voltage. out Stable, thereby ensuring the interference field strength of deep target points. Constant; The initial impedance of the stimulation circuit is denoted as . The real-time impedance of the stimulation circuit at time t; For the physiological mapping at time t, a logarithmic function is used. The nonlinear sensing characteristics of neurons to electrical signals are simulated to simulate the response characteristics of neurons, thereby improving the system sensitivity while avoiding the safety risks of overstimulation, i.e., the basic stimulation intensity output by the nonlinear physiological mapping model. The preset minimum effective current threshold, This is the preset gain sensitivity coefficient. Let be the power spectral density at time t. To preset the power spectral density reference value, It is a stimulus state regulator used to characterize the trigger state and output envelope of non-invasive brain stimulation commands.

[0058] In this embodiment, the amplitude of the stimulation current output does not increase linearly with the pathological signal, but rather uses a logarithmic mapping relationship for gain control. This ensures sensitive intervention at low power while also guaranteeing electrophysiological safety for clinical use through "high-level voltage limit".

[0059] It should be understood that the gain sensitivity coefficient It is preset based on the basic physiological characteristics of the target individual. Different values ​​can be manually configured for different patients, but it is fixed for the same patient in one treatment.

[0060] Specifically, when the processing module determines that the target is in the first preset state, S(t) is the enabled state value, and amplitude smoothing is performed according to the preset rise and fall slope function to avoid physiological discomfort caused by sudden current changes; when the target is determined to leave the first preset state, S(t) returns to zero and the non-invasive brain stimulation output stops.

[0061] Among them, impedance compensation In the calculation formula It includes a real part (resistance) and an imaginary part (capacitance). The compensation logic mainly adjusts the voltage drop caused by the increase in real impedance in real time.

[0062] Secondly, such as Figure 2 As shown, the present invention also proposes a brain stimulation closed-loop control device based on real-time EEG feedback and impedance compensation, comprising: a signal acquisition module, an impedance monitoring module, a TIS stimulation module, and a brain stimulation parameter control instruction generation system based on real-time EEG feedback and impedance compensation as described in any one of the first aspects. The signal acquisition module is used to acquire the target's electroencephalogram (EEG) data; the impedance monitoring module is used to obtain the target's stimulation circuit impedance; and the TIS stimulation module is used to perform non-invasive brain stimulation on the target according to the brain stimulation parameter control instructions.

[0063] In some embodiments, the TIS stimulation module includes a dual-channel high-frequency constant-current stimulation source and a TIS stimulation electrode pair. The dual-channel high-frequency constant-current stimulation source is electrically connected to the processing module; the TIS stimulation electrode pair is electrically connected to the dual-channel high-frequency constant-current stimulation source. The dual-channel high-frequency constant-current stimulation source is used to generate dual-channel high-frequency sine waves according to brain stimulation parameter control commands, and the TIS stimulation electrode pair is used to apply the dual-channel high-frequency sine waves to the target to synthesize an interference electric field at the deep target points.

[0064] In some embodiments, the TIS stimulation electrode pair includes a pair of electrodes.

[0065] In some embodiments, the TIS stimulation electrode pair includes multiple pairs of electrodes.

[0066] It should be understood that the multiple pairs in this embodiment include two or more pairs. The number of electrode pairs is set as needed, and the spatial intersection point is moved by phase control after adjusting the number of electrode pairs.

[0067] In some embodiments, a swept carrier is used instead of a fixed high-frequency carrier to further reduce the nervous system’s adaptation to the carrier itself.

[0068] It should be understood that high-frequency carrier refers to the dual-channel high-frequency sinusoidal current signal generated by dual-channel high-frequency constant current stimulation sources. Frequency sweep carrier achieves the frequency sweep function by shifting the fundamental frequency of the dual-channel high-frequency sinusoidal current signal within a preset frequency band in real time. Its purpose is to eliminate the physical tolerance of neural tissue to a single carrier frequency while maintaining the stability of the TIS difference frequency.

[0069] In some embodiments, the impedance monitoring module acquires the waveform of the induced current generated by the stimulation carrier in the circuit in real time by connecting a precision sampling resistor in series in the constant current output circuit of the dual high-frequency constant current stimulation source of the TIS stimulation module; the processing module uses the same source detection technology to extract the amplitude and phase information of the current feedback, and then calculates the stimulation circuit impedance including the electrode-skin contact impedance, so as to calculate the impedance compensation coefficient based on the stimulation circuit impedance and the preset initial stimulation circuit impedance; and uses the impedance compensation coefficient to perform gain correction on the basic stimulation intensity.

[0070] At the end of stimulation, the stimulation energy gradient of the TIS stimulation electrode pair is controlled to decrease smoothly to prevent transient inductive effects or patient discomfort caused by sudden power outage.

[0071] In summary, the brain stimulation closed-loop control device based on real-time EEG feedback and impedance compensation proposed in this invention establishes a three-in-one closed-loop mapping mechanism of physiological signal, physical field strength, and environmental impedance: EEG data is acquired through a signal acquisition module, and the power spectral density of the β band is extracted from the power spectral density through a processing module; when the absolute value of the difference between the power spectral density and the preset power spectral density benchmark value is greater than a preset threshold and the duration exceeds a preset duration, the target is determined to be in a pathological β burst state; TIS is triggered only when a pathological β burst state is detected, and a frequency following algorithm is used to dynamically correct the difference frequency, that is, the center frequency of the β band is used as the TIS difference frequency control parameter to achieve precise intervention for abnormal oscillations; moreover, in order to address the problem of inaccurate dosage caused by impedance fluctuations, this invention introduces a real-time impedance feedback loop, which automatically adjusts the voltage gain of the stimulation current by calculating the impedance compensation coefficient; when the contact impedance increases, the output drive is increased proportionally, thereby ensuring that the effective interference electric field strength of the deep target point is always maintained at a preset constant treatment level.

[0072] Thirdly, this invention also proposes a method for generating brain stimulation parameter control commands based on real-time EEG feedback and impedance compensation, comprising: Obtain the target's electroencephalogram (EEG) data; Obtain the stimulation circuit impedance of the target; Based on the EEG data, determine whether the target is in the first preset state; if so, generate brain stimulation parameter control instructions based on the EEG data and stimulation circuit impedance.

[0073] In some embodiments, determining whether a target is in a first preset state based on electroencephalogram (EEG) data specifically includes: Based on the EEG data, the power spectral density of the EEG data in the β band was obtained; The power spectral density is compared with a preset power spectral density benchmark value. When the absolute value of the difference between the power spectral density and the preset power spectral density benchmark value is greater than a preset threshold and the duration exceeds a preset duration, the target is determined to be in the first preset state.

[0074] In some embodiments, before determining whether the target is in a first preset state based on the electroencephalogram (EEG) data, the method further includes: preprocessing the EEG data; wherein the preprocessing includes artifact removal and bandpass filtering.

[0075] In some embodiments, brain stimulation parameter control instructions include stimulation current output amplitude and TIS difference frequency control parameters.

[0076] In some embodiments, brain stimulation parameter control instructions are generated based on electroencephalogram (EEG) data and stimulation circuit impedance, specifically including: extracting the feature vector of the EEG signal in the β band from the power spectral density; wherein the feature vector of the EEG signal in the β band includes power increment and center frequency; The power increment is input into a preset nonlinear physiological mapping model to obtain the baseline stimulus intensity; The impedance compensation coefficient is calculated based on the stimulation circuit impedance and the preset initial stimulation circuit impedance. The gain of the basic stimulus intensity is corrected by using an impedance compensation coefficient to obtain the stimulus current output amplitude. Map the center frequency to TIS differential frequency control parameters; The stimulation current output amplitude and TIS difference frequency control parameters are integrated and encapsulated to generate brain stimulation parameter control instructions.

[0077] In some embodiments, mapping the center frequency to TIS differential frequency control parameters specifically includes: performing a weighted average of the center frequency at the current moment and the center frequencies at a preset number of historical moments, and mapping the weighted average result to TIS differential frequency control parameters.

[0078] Fourthly, the present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the brain stimulation parameter control instruction generation method based on real-time EEG feedback and impedance compensation as described in any one of the third aspects.

[0079] Fifthly, the present invention also proposes a computer program product, comprising a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the brain stimulation parameter control instruction generation method based on real-time EEG feedback and impedance compensation as described in any one of the third aspects.

[0080] The present invention will now be described in conjunction with specific embodiments.

[0081] Example 1 This embodiment discloses a closed-loop control device for brain stimulation parameters based on real-time EEG feedback and impedance compensation, including: a signal acquisition module, an impedance monitoring module, an acquisition module, a processing module, an impedance monitoring module, and a TIS stimulation module. The signal acquisition module is used to acquire the target's EEG data in real time; The impedance monitoring module is used to synchronously acquire the stimulation circuit impedance of the electrode circuit of the TIS-stimulation module in real time. ; The processing module is used to acquire EEG data and stimulation circuit impedance in real time through the acquisition module. Artifact removal and bandpass filtering were performed on the EEG data; power spectral density was calculated on the artifact-removed and bandpass-filtered EEG data to obtain the power spectral density of the EEG data in the β band. According to power spectral density To determine whether the target is in a pathological β burst state; When power spectral density Compared with the preset power spectral density reference value P base The absolute value of the difference is greater than the preset power spectral density reference value P. base When the value is more than three times the standard deviation and the duration exceeds the preset duration, the target is determined to be in a pathological β-burst state. When the target is determined to be in a pathological beta burst state, the feature vector of the EEG signal in the beta band is extracted from the power spectral density; wherein, the feature vector of the EEG signal in the beta band includes the power increment. and center frequency f peak ; The power increment ΔP is input into a preset nonlinear physiological mapping model to obtain the baseline stimulus intensity. ;in, ; Based on the impedance of the stimulation circuit With preset initial stimulation circuit impedance Calculate the impedance compensation coefficient ;in, ; Using impedance compensation factor Basic stimulus intensity Gain correction is performed to obtain the amplitude of the stimulation current output. ;in, ; The center frequency at the current moment is weighted and averaged with the center frequencies of a preset number of historical moments, and the weighted average result is mapped to the TIS differential frequency control parameters. The amplitude of the stimulation current output Integrate and encapsulate with TIS difference frequency control parameters to generate brain stimulation parameter control instructions; The TIS stimulation module is used to output dual high-frequency sine waves according to brain stimulation parameters and control commands using dual high-frequency constant current stimulation sources. These waves are then applied to the target through TIS stimulation electrodes to synthesize an interference electric field at the deep target point.

[0082] A simulation experiment was conducted using the closed-loop control device for brain stimulation parameters based on real-time EEG feedback and impedance compensation proposed in Example 1. The simulation results showed that, under the premise of achieving the same tremor suppression effect, the total electrical input was reduced by about 40% to 60% due to the "on-demand triggering" mechanism, which significantly reduced the skin thermal effect and effectively avoided the rapid compensation of neurons to the electric field intensity (i.e. tolerance phenomenon).

[0083] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A brain stimulation parameter control command generation system based on real-time EEG feedback and impedance compensation, characterized in that, include: The acquisition module is used to acquire the target's electroencephalogram (EEG) data and stimulation circuit impedance; The processing module is used to obtain the power spectral density of EEG data in the β band based on the EEG data; determine whether the target is in a first preset state based on the power spectral density; if so, extract the feature vector of the EEG signal in the β band from the power spectral density; wherein the feature vector of the EEG signal in the β band includes the power increment and the center frequency; input the power increment into a preset nonlinear physiological mapping model to obtain the basic stimulation intensity; calculate the impedance compensation coefficient based on the stimulation circuit impedance and the preset initial stimulation circuit impedance; use the impedance compensation coefficient to perform gain correction on the basic stimulation intensity to obtain the stimulation current output amplitude; map the center frequency to the TIS difference frequency control parameter; integrate and encapsulate the stimulation current output amplitude and the TIS difference frequency control parameter to generate brain stimulation parameter control instructions; During the judgment process, when the absolute value of the difference between the power spectral density and the target's preset power spectral density benchmark value is greater than the preset threshold and the duration exceeds the preset duration, the target is determined to be in the first preset state.

2. The brain stimulation parameter control command generation system based on real-time EEG feedback and impedance compensation according to claim 1, characterized in that, The amplitude of the stimulation current output is: ； ； ； ； In the formula, I out To stimulate the output amplitude of the current, This is the impedance compensation coefficient; To preset the initial stimulation circuit impedance, Let be the impedance of the stimulation circuit at time t; This represents the physiological mapping at time t, i.e., the base stimulus intensity output by the nonlinear physiological mapping model. The preset minimum effective current threshold, This is the preset gain sensitivity coefficient. For power increment; Let be the power spectral density at time t. To preset the power spectral density reference value, Let be the stimulus state regulation factor at time t.

3. The brain stimulation parameter control command generation system based on real-time EEG feedback and impedance compensation according to claim 2, characterized in that, Mapping the center frequency to TIS differential frequency control parameters specifically includes: The center frequency at the current moment is weighted and averaged with the center frequencies of a preset number of historical moments, and the weighted average result is mapped to the TIS differential frequency control parameters.

4. The brain stimulation parameter control command generation system based on real-time EEG feedback and impedance compensation according to claim 1, characterized in that, Before determining whether the target is in the first preset state based on the EEG data, the process also includes: The EEG data were preprocessed, including artifact removal and bandpass filtering.

5. A closed-loop modulation device for brain stimulation based on real-time EEG feedback and impedance compensation, characterized in that, include: The signal acquisition module, the impedance monitoring module, the TIS stimulation module, and the brain stimulation parameter control instruction generation system based on real-time EEG feedback and impedance compensation as described in any one of claims 1-4; The signal acquisition module is used to acquire the target's electroencephalogram (EEG) data; the impedance monitoring module is used to obtain the target's stimulation circuit impedance; and the TIS stimulation module is used to perform non-invasive brain stimulation on the target according to the brain stimulation parameter control instructions.

6. A method for generating brain stimulation parameter control commands based on real-time EEG feedback and impedance compensation, characterized in that, include: Acquire the target's electroencephalogram (EEG) data and stimulation circuit impedance; Based on the EEG data, the power spectral density of the EEG data in the β band was obtained; Based on the power spectral density, determine whether the target is in a first preset state; if so, extract the feature vector of the EEG signal in the β band from the power spectral density; the feature vector of the EEG signal in the β band includes the power increment and the center frequency; input the power increment into a preset nonlinear physiological mapping model to obtain the basic stimulation intensity; calculate the impedance compensation coefficient based on the stimulation circuit impedance and the preset initial stimulation circuit impedance; use the impedance compensation coefficient to perform gain correction on the basic stimulation intensity to obtain the stimulation current output amplitude; map the center frequency to the TIS difference frequency control parameter; integrate and encapsulate the stimulation current output amplitude and the TIS difference frequency control parameter to generate brain stimulation parameter control instructions; During the judgment process, when the absolute value of the difference between the power spectral density and the target's preset power spectral density benchmark value is greater than the preset threshold and the duration exceeds the preset duration, the target is determined to be in the first preset state.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the brain stimulation parameter control instruction generation method based on real-time EEG feedback and impedance compensation as described in claim 6.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the brain stimulation parameter control instruction generation method based on real-time EEG feedback and impedance compensation as described in claim 6.

Images