Transcranial electrical stimulation system and storage medium

Abstract

The invention relates to a transcranial electrical stimulation system and a storage medium, and the system comprises a neural signal sensing module which is configured to be used for collecting neural activity signals of the brain of a subject in real time; a transcranial alternating current stimulation module configured to apply transcranial alternating current stimulation to a target area of the brain of the subject; the real-time signal processing and control module is configured to be in communication connection with the neural signal sensing module and the transcranial alternating current stimulation module at the same time; the real-time signal processing and control module is configured to: receive a neural activity signal; extracting features of neural oscillation in a preset target frequency band from the neural activity signal in real time, wherein the features at least comprise an instantaneous phase; predicting a future phase of the neural oscillation based on the instantaneous phase, and generating a stimulation parameter in an anti-phase relationship with the future phase; and based on the stimulation parameters, controlling the transcranial alternating current stimulation module to apply adaptive transcranial alternating current stimulation to the target area of the brain of the patient.

Description

Technical Field

[0001] This application relates to the field of medical device technology, and in particular to a transcranial electrical stimulation system and storage medium for electrically stimulating brain regions of a user. Background Technology

[0002] Depression is a common mental disorder whose pathophysiological mechanisms involve abnormalities in the function and connectivity of multiple brain regions, often manifesting as disturbances in specific frequency neural oscillations. Transcranial electrical stimulation (tES), as a non-invasive neuromodulation technique, has shown potential in the treatment of depression.

[0003] However, current tES systems used for treating depression generally suffer from technical limitations such as inflexible stimulation parameters. This inflexibility manifests primarily in the system's lack of closed-loop feedback adjustment based on the patient's real-time neural activity. The stimulation protocols are often "one-size-fits-all," employing preset, fixed frequencies and intensities, failing to provide personalized treatment and limiting therapeutic efficacy. Due to the fixed parameters and lack of real-time adjustment, some patients may experience poor tolerance to the stimulation intensity, and the treatment effect may not be optimized. Summary of the Invention

[0004] The purpose of this application is to provide a transcranial electrical stimulation system and storage medium for adaptive personalized treatment, which can solve the problems of low target accuracy, inability to perform multi-target synergistic regulation, fixed stimulation parameters, and lack of personalized real-time adjustment capability of existing transcranial electrical stimulation systems in the treatment of diseases such as depression.

[0005] To this end, in a first aspect, this application provides a transcranial electrical stimulation system, comprising: a neural signal sensing module configured to acquire neural activity signals from a user's brain in real time, the neural signal sensing module including a plurality of acquisition electrodes; a transcranial alternating current stimulation module configured to apply transcranial alternating current stimulation to a target region of a subject's brain, the transcranial alternating current stimulation module including a plurality of stimulation electrodes; and a real-time signal processing and control module configured to simultaneously communicate with the neural signal sensing module and the transcranial alternating current stimulation module; the real-time signal processing and control module is configured to: receive the neural activity signals; extract features of neural oscillations within a preset target frequency band from the neural activity signals in real time, the features including at least an instantaneous phase; predict the future phase of the neural oscillations based on the instantaneous phase, and generate stimulation parameters that are inversely related to the future phase; and control the transcranial alternating current stimulation module to apply appropriate transcranial alternating current stimulation to the target region of the user's brain based on the stimulation parameters.

[0006] In some embodiments, the neural signal sensing module includes a high-density electroencephalogram (EEG) electrode cap conforming to the international 10-20 or 10-10 system standard.

[0007] In some embodiments, the real-time signal processing and control module is configured to use a Hilbert transform or a phase-locked loop (PLL) to extract the instantaneous phase of the neural oscillation.

[0008] In some embodiments, the real-time signal processing and control module is configured to use a Kalman filter or a linear extrapolation algorithm to predict the future phase of the neural oscillation and compensate for the inherent delay of the system.

[0009] In some embodiments, the real-time signal processing and control module is further configured to: remove in real-time stimulation artifacts generated in the neural activity signal when transcranial alternating current stimulation is applied by the transcranial alternating current stimulation module.

[0010] In some embodiments, the real-time signal processing and control module uses a template subtraction method or an adaptive filtering method to remove the stimulus artifacts.

[0011] In some embodiments, the plurality of stimulation electrodes are configured to form a multichannel high-density electrode array.

[0012] In some embodiments, the placement of each of the stimulation electrodes is optimized based on the subject's individualized head model and electric field simulation results, in order to focus the stimulation electric field on one or more target brain regions.

[0013] In some embodiments, the real-time signal processing and control module is further configured to: monitor the power change of the target frequency band neural oscillation in real time; and adaptively adjust the amplitude of the transcranial alternating current stimulation output by the transcranial AC stimulation module based on the error between the power change and a preset inhibition level using a proportional-integral-derivative control algorithm.

[0014] In a second aspect, this application provides a storage medium storing computer-readable instructions thereon, characterized in that the computer-readable instructions are executed by one or more processors to implement a transcranial electrical stimulation method, the transcranial electrical stimulation method comprising:

[0015] Acquire neural activity signals from the brain of the target individual;

[0016] The neural activity signal is processed, including: real-time extraction of the instantaneous phase of neural oscillations within a preset target frequency band; prediction of the future phase of the neural oscillations; and generation of stimulation parameters that are inversely related to the future phase; and

[0017] Based on the generated stimulation parameters, transcranial alternating current stimulation is applied to the target area of ​​the individual's brain.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: the system uses a neural signal sensing module to collect neural activity signals from the subject's brain in real time, so as to realize real-time monitoring of the user's neural activity and dynamically adjust stimulation parameters according to feedback, thereby realizing personalized and intelligent treatment and is expected to improve the treatment effect. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the circuit principle of a transcranial electrical stimulation system provided in an embodiment of this application.

[0020] Figure 2 for Figure 1 The diagram shows the unit composition of the real-time signal processing and control module.

[0021] Figure 3 This is a schematic diagram illustrating the usage process of the transcranial electrical stimulation system provided in the embodiments of this application. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application 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 in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0023] This application relates to a transcranial electrical stimulation system for the treatment of diseases such as depression. It introduces the concept of "reverse signal" cancellation from active noise cancellation (ANC) into the field of transcranial electrical stimulation. It can perform closed-loop control based on the "active noise cancellation" concept. During operation, it does not apply a fixed stimulus like traditional transcranial electrical stimulation devices or systems. Instead, it collects neural oscillations in the brain of the target individual in real time that are related to specific diseases, predicts their future phase, and applies a transcranial alternating current stimulation (tACS) with the opposite phase (180 degrees out of phase) to actively attenuate or cancel specific pathological neural activity, thereby achieving precise stimulation of the target area of ​​the brain.

[0024] Reference Figure 1The illustration shows an embodiment of the present invention providing a transcranial electrical stimulation system 100. The system 100 includes a neural signal sensing module 101, a real-time signal processing and control module (RPCU) 102, and a transcranial alternating current stimulation (tACS) module 103.

[0025] The neural signal sensing module 101 is responsible for real-time acquisition of neural activity signals from the target individual's brain. The neural signal sensing module 101 includes several acquisition electrodes. In a preferred embodiment, the module 101 employs a high-density electroencephalography (EEG) electrode cap, for example, containing 64 or 128 Ag / AgCl sintered electrodes, arranged according to the international 10-20 or 10-10 standard system. To ensure signal quality, the contact impedance between the electrodes and the scalp must be maintained below 5kΩ, and the signal acquisition sampling rate must be no less than 1kHz to capture neural oscillations up to 100Hz without distortion.

[0026] Combination Figure 2 As shown, the Real-Time Signal Processing and Control Module (RPCU) 102 is the intelligent core of the entire system. It can be further divided into multiple functional units such as the signal preprocessing unit 102A, the feature extraction unit 102B, the phase prediction and parameter generation unit 102C, the stimulus artifact removal unit 102D, and the adaptive control unit 102E.

[0027] The signal preprocessing unit 102A is used to receive the raw EEG signal from the neural signal sensing module 101 and perform preprocessing. The preprocessing mainly includes preliminary digital bandpass filtering (e.g., 0.1Hz-200Hz) and 50 / 60Hz power frequency notch filtering, while performing preliminary detection and labeling of artifacts such as eye movement and electromyography.

[0028] The feature extraction unit 102B performs precise narrowband filtering on the signal preprocessed by the signal preprocessing unit 102A to separate the target frequency band that is preset by the user and related to a specific disease (e.g., the alpha band of the frontal region, 8-12Hz, associated with depression). Subsequently, algorithms such as Hilbert transform or phase-locked loop (PLL) are used to calculate the instantaneous phase and amplitude of the target neural oscillation from the narrowband signal.

[0029] Take the application of the Hilbert transform to the prefrontal alpha oscillation as an example. The Hilbert transform is mainly used to convert a real signal into a complex signal, thereby directly extracting its instantaneous features. Specifically, a high-precision digital bandpass filter is applied to the preprocessed EEG signal to separate the alpha band (8-12Hz) of the prefrontal cortex.

[0030] Then, the Hilbert transform is applied to the filtered alpha band signal x(t) to construct an analytic signal, obtaining its 90-degree phase-shifted version H{x(t)}. Taking the original signal x(t) as the real part and its Hilbert transform H{x(t)} as the imaginary part, a complex analytic signal z(t) is constructed.

[0031] z(t) = x(t) + j·H{x(t)}

[0032] The instantaneous amplitude A(t) is obtained by calculating the absolute value of the analytic signal z(t) at each time point, which reflects the real-time change in the intensity of the alpha oscillation.

[0033]

[0034] The instantaneous phase φ(t) is obtained by calculating the argument of the analytic signal z(t) at each time point, which precisely describes the position of the alpha oscillation within its period.

[0035]

[0036] The phase prediction and parameter generation unit 102C is used to predict the phase of neural oscillations at future moments. Due to the inherent time delay in the entire closed-loop system (from signal acquisition to stimulus output), the phase prediction and parameter generation unit 102C is necessary. In one embodiment, the phase prediction and parameter generation unit 102C uses a Kalman filter to establish a state-space model to predict phase evolution. After prediction, the phase prediction and parameter generation unit 102C generates a tACS stimulation parameter (mainly including phase, frequency, and amplitude) that is 180 degrees out of phase with the predicted phase and sends this parameter to the tACS module 103.

[0037] In the phase prediction and parameter generation unit 102C, to compensate for the inherent time delay of the system, it is necessary to predict the future phase of the neural oscillation signal. The Kalman filter is a powerful estimation algorithm capable of optimally estimating the state of a dynamic system from a series of noisy observations. Its core process is a recursive loop comprising prediction and update phases.

[0038] Step 1: Establish a state-space model

[0039] To apply the Kalman filter, the observed system must first be described mathematically. This is done by establishing a state-space model containing two core equations: a process model and an observation model.

[0040] The process model describes the evolution of the system state over time. For neural oscillations, the core states we are interested in are their instantaneous phase φ and instantaneous angular frequency ω. We define this state as a vector x_t = [φ_t, ω_t]^T.

[0041] x_t=F*x_{t-1}+w_{t-1}

[0042] Where x_t is the state vector at time t, and F is the state transition matrix. φ_t = φ_{t-1} + ω_{t-1} * Δt, and ω_t is approximately equal to ω_{t-1}, therefore the matrix F can be set as [[1,Δt],[0,1]], where Δt is the time step. w_{t-1} is the process noise, which is assumed to be Gaussian white noise with zero mean.

[0043] The observation model describes the relationship between the observed values ​​and the system state. Our observed value z_t comes from the instantaneous phase calculated by the feature extraction unit 102B using methods such as Hilbert transform.

[0044] z_t=H*x_t+v_t

[0045] z_t is the observation value at time t. H is the observation matrix. H is initially set to [1,0]. v_t is the observation noise, representing the error generated during the measurement process, and is assumed to be Gaussian white noise with zero mean.

[0046] Step 2: Recursive loop of the Kalman filter

[0047] After the model is built, the filter enters a real-time recursive loop, executing the following two phases at each time point t:

[0048] Phase 1: Based on the posterior optimal estimate at time t-1, the filter predicts the state at time t.

[0049] x_{t|t-1}=F*x_{t-1|t-1}

[0050] This formula uses the state transition matrix F and the best state estimate x_{t-1|t-1} from the previous time step to predict the state x_{t|t-1} at the current time step ("t|t-1" is the estimate of time step t given the data at time step t-1).

[0051] At the same time, the filter also updates the uncertainty of the state estimate (represented by the error covariance matrix P), reflecting that the uncertainty of the prediction increases due to the presence of process noise.

[0052] Phase 2: When the actual observed value z_t arrives at time t, the filter uses this new information to correct the predicted value and obtain a more accurate posterior estimate.

[0053] x_{t|t}=x_{t|t-1}+K_t*(z_t-H*x_{t|t-1})

[0054] K_t is the Kalman gain. This formula multiplies the prediction residual (i.e., the difference between the actual observation z_t and the predicted observation H*x^_{t|t-1}) by the Kalman gain to correct the predicted state, thereby obtaining the optimal estimate x^_{t|t} at the current time.

[0055] Finally, the system updates and reduces the value of the error covariance matrix P based on the new observation information, because the new measurement data reduces the uncertainty of the state estimation.

[0056] Step 3: In order to compensate for the inherent delay Δt_delay (e.g., 20 milliseconds) in the entire closed-loop system, unit 102C will use the obtained optimal state estimate and the process model again to predict the future phase after Δt_delay.

[0057] φ_{future}=φ_t+ω_t*Δt_delay

[0058] φ{future} is a prediction of the phase of the neural oscillation at the arrival of the future stimulus. Unit 102C then generates a stimulation parameter that is 180 degrees out of phase with this prediction and sends it to the tACS module, thereby achieving precise and effective phase intervention. Stimulation artifact removal unit 102D: While the transcranial AC stimulation module 103 applies stimulation, the electric field it generates can contaminate the EEG signal. To ensure the accuracy of closed-loop feedback, the stimulation artifact removal unit 102D uses a template subtraction method or adaptive filtering to subtract this stimulation artifact from the signal acquired by the neural signal sensing module 101 in real time.

[0059] The adaptive control unit 102E can monitor the power of the target frequency band neural oscillation in real time and compare it with the baseline or a preset inhibition target. Based on this error, a proportional-integral-derivative (PID) controller automatically adjusts the stimulation amplitude sent to the transcranial alternating current stimulation module 103, thereby using the lowest effective stimulation dose while ensuring therapeutic efficacy, improving the personalization of treatment and user tolerance.

[0060] The transcranial alternating current (TAC) stimulation module 103 is the execution terminal of the entire system. It is configured to apply TAC stimulation to target regions of the subject's brain. The TAC stimulation module 103 may include several stimulation electrodes. The stimulation electrodes are preferably disposable sponge electrodes or other conductive gel electrodes. The stimulation electrodes are preferably configured into a multi-channel stimulator to optimize the stimulation effect, for example, the stimulation electrodes form an HD-tACS electrode array with a central electrode and surrounding electrodes (such as a 4x1 ring array). In a key implementation step, the placement of the stimulation electrodes is not fixed, but optimized through individualized electric field simulation. Specifically, a personal finite element model of the head is first constructed using the user's T1-weighted MRI data, and then simulation software such as ROAST or SimNIBS is used to simulate the intracranial electric field distribution under different electrode configurations. Finally, the electrode scheme that can most effectively and accurately focus the electric field on the target brain region (e.g., the dorsolateral prefrontal cortex associated with depression) is selected.

[0061] When patients use the system described in this application for treatment of depression, they need to first have a treatment or intervention plan developed by a doctor at a medical institution, such as... Figure 3 As shown, the specific usage process is as follows:

[0062] Step 1. Individualized modeling: Before the first treatment, the doctor uses the patient's MRI data to build a head model and optimize the electrode layout.

[0063] Step 2. System Initialization: Place the acquisition and stimulation electrodes, ensuring good contact between the acquisition and stimulation electrodes and the corresponding brain regions of the patient. Check the impedance and calibrate the system delay.

[0064] Step 3. Baseline recording: Record EEG for 5-10 minutes in a non-stimulated state as a reference.

[0065] Step 4. Closed-loop control: The system is started, and the real-time signal processing and control module 102 begins to execute the complete closed-loop process described above, from signal acquisition, processing, prediction to the generation of inverse stimulation parameters. The transcranial AC stimulation module 103 performs precise stimulation according to the instructions. The adaptive control unit 102E and the artifact removal unit 102D operate online throughout the entire process.

[0066] Step 5. End and Assessment: After treatment, record EEG again to assess the aftereffects.

[0067] Furthermore, this application provides a storage medium storing computer-readable instructions that are executed by one or more processors to implement a transcranial electrical stimulation method. The transcranial electrical stimulation method includes a sensing step, a processing step, and a stimulation step.

[0068] The sensing step includes using a neural signal sensing module to collect neural activity signals from the user's brain in real time.

[0069] The processing steps include processing the neural activity signals acquired in the sensing steps. This processing includes: extracting the instantaneous phase of neural oscillations within a preset target frequency band in real time; predicting the future phase of the neural oscillations; and generating stimulation parameters that are inversely related to the future phase.

[0070] The stimulation process involves applying transcranial alternating current (TAC) stimulation to a target area of ​​the user's brain using a transcranial AC stimulation module, based on the generated stimulation parameters.

[0071] The method also includes: constructing an individualized head model using the user's magnetic resonance imaging (MRI) data prior to the sensing step; and optimizing the electrode placement of the neural signal sensing module and the transcranial alternating current stimulation module through electric field simulation to achieve precise targeting of the target brain region.

[0072] This application provides a computer program product including computer-readable instructions stored in a storage medium. One or more processors of one or more electronic devices read the computer-readable instructions from the storage medium, load and execute the computer-readable instructions, causing one or more electronic devices to implement the transcranial electrical stimulation method as described above.

[0073] While exemplary embodiments of the present invention have been shown and described above, those skilled in the art will understand that various changes and modifications can be made, and equivalent forms can be substituted for elements therein without departing from the actual scope of the invention. Furthermore, many modifications can be made to adapt to specific situations and the teachings of the invention without departing from its central scope. Therefore, all embodiments falling within the scope of the claims of this invention are within the protection scope of this invention.

Claims

1. A transcranial electrical stimulation system, characterized in that, include: A neural signal sensing module is configured to collect neural activity signals from a user's brain in real time, and the neural signal sensing module includes several acquisition electrodes. A transcranial alternating current stimulation (TAC) module, configured to apply TAC to a target region of a subject's brain, the TAC module comprising a plurality of stimulation electrodes; and A real-time signal processing and control module is configured to communicate simultaneously with both the neural signal sensing module and the transcranial alternating current stimulation module; wherein the real-time signal processing and control module is configured to: Receive the aforementioned neural activity signals; Features of neural oscillations within a preset target frequency band are extracted in real time from the neural activity signals, and the features include at least the instantaneous phase. Based on the instantaneous phase, the future phase of the neural oscillation is predicted, and stimulation parameters that are inversely related to the future phase are generated. as well as Based on the stimulation parameters, the transcranial alternating current stimulation module is controlled to apply appropriate transcranial alternating current stimulation to the target area of ​​the user's brain.

2. The transcranial electrical stimulation system according to claim 1, characterized in that, The neural signal sensing module includes a high-density EEG electrode cap that conforms to the international 10-20 or 10-10 system standard.

3. The transcranial electrical stimulation system according to claim 1, characterized in that, The real-time signal processing and control module is configured to use Hilbert transform or phase-locked loop to extract the instantaneous phase of the neural oscillation.

4. The transcranial electrical stimulation system according to claim 1, characterized in that, The real-time signal processing and control module is configured to use a Kalman filter or a linear extrapolation algorithm to predict the future phase of the neural oscillation and to compensate for the inherent delay of the system.

5. The transcranial electrical stimulation system according to claim 1, characterized in that, The real-time signal processing and control module is also configured to: remove stimulation artifacts generated in the neural activity signal in real time when transcranial alternating current stimulation is applied by the transcranial alternating current stimulation module.

6. The transcranial electrical stimulation system according to claim 5, characterized in that, The real-time signal processing and control module uses template subtraction or adaptive filtering to remove the stimulus artifacts.

7. The transcranial electrical stimulation system according to claim 1, characterized in that, The aforementioned stimulation electrodes are configured to form a multi-channel high-density electrode array.

8. The transcranial electrical stimulation system according to claim 7, characterized in that, The placement of each of the stimulation electrodes is determined based on the individualized head model of the subject and the results of electric field simulation, so as to focus the stimulation electric field on one or more target brain regions.

9. The transcranial electrical stimulation system according to claim 1, characterized in that, The real-time signal processing and control module is also configured to: monitor the power change of the target frequency band neural oscillation in real time; and adaptively adjust the amplitude of the transcranial alternating current stimulation module output stimulation based on the error between the power change and the preset inhibition level using a proportional-integral-derivative control algorithm.

10. A storage medium having computer-readable instructions stored thereon, characterized in that, The computer-readable instructions are executed by one or more processors to implement a transcranial stimulation method, the transcranial stimulation method comprising: Acquire neural activity signals from the brain of the target individual; The neural activity signal is processed, including: real-time extraction of the instantaneous phase of neural oscillations within a preset target frequency band; prediction of the future phase of the neural oscillations; and generation of stimulation parameters that are inversely related to the future phase; and Based on the generated stimulation parameters, transcranial alternating current stimulation is applied to the target area of ​​the individual's brain.

Images