Intelligent magnetic field nerve regulation rehabilitation system

Abstract

The invention relates to the technical field of neuromodulation and rehabilitation treatment, and discloses an intelligent magnetic field neuromodulation rehabilitation system, which comprises a signal acquisition and feature extraction module for acquiring brain signals of a patient and extracting multi-scale variables including microscopic neuron membrane potential, mesoscopic brain region synchronism and macroscopic cognitive state; the magnetic field adjusting module is used for generating and outputting magnetic field stimulation signals according to the optimization result, and the magnetic field stimulation signals comprise frequency, intensity and waveform parameters of a magnetic field; the dynamic optimization module is used for calculating an optimal path of the magnetic field stimulation signal based on the input multi-scale variable and the constraint condition; and the closed-loop feedback control module is used for receiving the real-time brain signal to update the system state and adjusting the parameters of the magnetic field stimulation signal based on feedback. Through multi-scale modeling, real-time feedback regulation and control and man-machine interaction technologies, precise regulation and control, efficient adaptability and visual clinical operation effects of the nervous system are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of neuromodulation and rehabilitation therapy, and in particular to an intelligent magnetic field neuromodulation rehabilitation system. Background Art

[0002] In modern society, the incidence of neurological diseases such as cognitive impairment, sequelae of brain injury, depression and anxiety has increased year by year. These diseases have seriously affected the quality of life of patients and even brought heavy burdens to families and society. With the continuous development of neuroscience, people have gradually realized that precise regulation of the brain through external stimulation can effectively improve the symptoms of these diseases. For example, magnetic field neuromodulation technology acts on the patient's brain in a non-invasive way to help restore the functional connection of the neural network, thereby promoting the improvement of cognitive function, emotional regulation and memory. Therefore, the development of a rehabilitation system that can efficiently and individually regulate the patient's neural state has become a technical need that needs to be urgently addressed in clinical treatment.

[0003] In the existing technology, repetitive transcranial magnetic stimulation and low-frequency magnetic field stimulation have become important means of neuromodulation. These technologies use magnetic fields to act on specific brain areas, and provide patients with effective non-invasive treatment by regulating the excitability of neurons or the synchronization between brain areas. Most existing devices use fixed parameter stimulation schemes, which are easy to operate and suitable for large-scale promotion. At the same time, this type of equipment has a certain effect on the regulation of specific neural functions, such as improving the activity level of local brain areas or enhancing patients' short-term memory ability.

[0004] However, existing technologies still have significant limitations in handling the precise regulation of complex neural activities. First, existing devices mostly use a single-scale regulation method, which only stimulates neural signals in specific brain areas or a certain frequency band. Although this design can achieve local functional improvements, it ignores the complexity of the multi-scale dynamic characteristics of the nervous system and makes it difficult to achieve comprehensive regulation of neural activity. Second, traditional devices usually use preset fixed stimulation parameters and lack a real-time feedback adjustment mechanism. When the patient's neural state changes, the device cannot dynamically adjust the output parameters, resulting in limited treatment effects or poor patient experience. In addition, existing technologies have relatively single functions in terms of human-computer interaction and treatment monitoring, lack an intuitive interface and complete data recording capabilities, cannot provide doctors with sufficient information support, and are not conducive to long-term treatment management. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an intelligent magnetic field neuromodulation rehabilitation system, which solves the problem that the prior art cannot achieve comprehensive modeling and precise regulation of the multi-scale dynamic characteristics of the nervous system.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent magnetic field neuromodulation rehabilitation system, comprising:

[0007] The signal acquisition and feature extraction module is used to collect the patient's brain signals and extract multi-scale variables, including micro-neuron membrane potential, meso-brain region synchronization, and macro-cognitive state;

[0008] A magnetic field adjustment module, used to generate and output a magnetic field stimulation signal according to the optimization result, including the frequency, intensity and waveform parameters of the magnetic field;

[0009] A dynamic optimization module for calculating the optimal path of the magnetic field stimulation signal based on the input multi-scale variables and constraints;

[0010] A closed-loop feedback control module for receiving real-time brain signals to update the system status and adjusting the parameters of the magnetic field stimulation signal based on the feedback;

[0011] The human-computer interaction and monitoring module is used to provide a user interface to display the patient's treatment status and support parameter setting and remote control.

[0012] Preferably, the signal acquisition and feature extraction module includes:

[0013] An EEG sensor array for collecting brain signals, including brain wave signals of target brain areas;

[0014] A signal preprocessing unit, used to perform denoising and filtering on brain signals and extract the frequency characteristics and amplitude of the signals;

[0015] The feature extraction unit is used to extract multi-scale variables, including neuronal membrane potential, phase synchronization between brain regions, and representative indicators of the whole brain cognitive state.

[0016] Preferably, the magnetic field regulation module includes:

[0017] A magnetic field generator, used for generating a controllable magnetic field signal, wherein the magnetic field parameters include frequency, intensity and waveform;

[0018] A multi-channel magnetic coil array is used to output magnetic field signals and act on the patient's target brain area;

[0019] The drive control unit is used to receive the magnetic field signal parameters of the dynamic optimization module and adjust the magnetic field output in real time.

[0020] Preferably, the dynamic optimization module includes:

[0021] A multi-scale modeling unit, used to construct a multi-scale neural dynamics model based on the patient's brain signals, wherein the multi-scale model includes a neuron membrane potential model, a brain region synchronization model, and a whole-brain cognitive state model;

[0022] An optimization calculation unit, used to calculate the optimal path of the magnetic field signal based on the input multi-scale variables and constraints;

[0023] The constraint processing unit is used to limit the frequency, intensity and patient tolerance range of the magnetic field signal.

[0024] Preferably, the multi-scale modeling unit comprises:

[0025] Neuronal membrane potential model, used to describe the dynamic behavior of single neuron membrane potential;

[0026] The brain region synchronization model is used to describe the dynamic behavior of the phase synchronization characteristics between brain regions;

[0027] A whole-brain cognitive state model is used to describe changes in cognitive states based on diffusion tensor calculations.

[0028] Preferably, the optimization calculation unit includes:

[0029] Action function building block, used to define Lagrangian function;

[0030] The solving unit is used to calculate the optimal path of the magnetic field signal based on the action function and the constraint conditions.

[0031] Preferably, the closed-loop feedback control module includes:

[0032] A feedback signal acquisition unit, used to collect the patient's brain signals in real time and detect changes in brain signals after magnetic field stimulation;

[0033] A dynamic adjustment unit, used to adjust the input variables of the multi-scale model and recalculate the optimization path according to the feedback signal;

[0034] The control output unit is used to send the adjusted magnetic field signal parameters to the magnetic field regulation module.

[0035] Preferably, the human-computer interaction and monitoring module includes:

[0036] A display unit, used to display the patient's brain signals, magnetic field signal parameters and treatment status in real time;

[0037] An operating unit, for providing a user interface to support parameter setting and magnetic field regulation control;

[0038] Data storage and remote unit to record patient treatment data and support remote control and monitoring.

[0039] Preferably, the dynamic optimization module calculates the optimal path of magnetic field stimulation based on the patient's real-time brain signals and satisfies the following constraints:

[0040] The frequency of the magnetic field is in the range of 1 to 50 Hz;

[0041] The magnetic field strength is in the range of 0.1 to 3 mT;

[0042] Patient tolerance to magnetic field stimulation is within the safe threshold.

[0043] Preferably, the optimization calculation is optimized based on the multi-scale neurodynamic characteristics of the patient, and the optimization objectives include:

[0044] Maximize the patient's neurological status;

[0045] Minimize patient discomfort;

[0046] Minimize the energy consumption of magnetic field regulation.

[0047] The present invention provides an intelligent magnetic field neuromodulation rehabilitation system. It has the following beneficial effects:

[0048] 1. This invention realizes cross-scale coupled modeling and precise control from neuronal membrane potential to whole-brain cognitive state through multi-scale modeling and dynamic optimization technology. Compared with the traditional single-level control method, it completely solves the problem that the existing technology is difficult to respond to complex neural activities, ensuring that the system can fully capture and process the dynamic changes of neural activities.

[0049] 2. The present invention adopts closed-loop feedback control and real-time correction technology solutions, which significantly improves the flexibility and accuracy of regulation by real-time acquisition of patient brain signals and rapid adjustment of magnetic field output. Compared with static stimulation technology that relies on preset parameters, it breaks through the limitation of delayed response when the patient's state changes, making the system more adaptable to individualized treatment needs.

[0050] 3. The present invention integrates human-computer interaction and remote monitoring technology solutions, combines intuitive operation interface, real-time data transmission and treatment history analysis, and provides a full range of treatment visualization and control capabilities. Compared with the defects of inconvenient interface interaction and untimely data feedback in the prior art, the present invention optimizes the clinical operation experience and provides more intelligent support for long-term treatment management. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a system structure diagram of the present invention;

[0052] Figure 2 This is a module architecture diagram of the signal acquisition and feature extraction module of the present invention;

[0053] Figure 3 This is a module architecture diagram of the magnetic field regulation module of the present invention;

[0054] Figure 4 This is a module architecture diagram of the dynamic optimization module of the present invention;

[0055] Figure 5 It is a module architecture diagram of the closed-loop feedback control module of the present invention;

[0056] Figure 6 This is a module architecture diagram of the human-computer interaction and monitoring module of the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] Please see attached Figure 1 -Attached Figure 6 , an embodiment of the present invention provides an intelligent magnetic field neuromodulation rehabilitation system, comprising:

[0059] The signal acquisition and feature extraction module is used to collect the patient's brain signals and extract multi-scale variables, including micro-neuron membrane potential, meso-brain region synchronization, and macro-cognitive state;

[0060] Through the signal acquisition and feature extraction module, the multi-scale neural activity characteristics of patients can be extracted, covering the membrane potential behavior at the neuron level, the synchronization state of brain regions, and the cognitive function characteristics of the whole brain. These characteristic variables are the core input parameters for subsequent multi-scale modeling and optimization calculations, ensuring that the system can dynamically adapt to changes in the patient's neural state and achieve precise regulation.

[0061] In general, the signal acquisition and feature extraction module needs to work in conjunction with the magnetic field regulation module and the dynamic optimization module. Specifically, the collected brain signal data is processed in real time and then passed to the dynamic optimization module to calculate the optimal path of the magnetic field stimulation parameters. At the same time, the closed-loop feedback control module will input the updated brain signal features into the module again to ensure the closed-loop working state of the entire system.

[0062] In this embodiment, the signal acquisition and feature extraction module includes the following three main units: EEG sensor array, signal preprocessing unit and feature extraction unit. The specific structure and function of each unit are described in detail below:

[0063] EEG sensor array

[0064] In this embodiment, the EEG sensor array is used to collect real-time bioelectric signals from the patient's brain. Generally, the sensor array covers key brain regions such as the patient's frontal lobe, parietal lobe, and temporal lobe. These brain regions are closely related to cognitive functions, emotional regulation, and other activities.

[0065] As a possible implementation, the sampling frequency of the EEG sensor array is set to 1000Hz. This sampling rate can ensure the timing integrity of the signal while avoiding missing high-frequency neuronal activity. Specifically, the sensitivity of the sensor can reach 1μV, ensuring that weaker brain electrical activity can be detected.

[0066] In some embodiments, the sensor array can be designed as a flexible structure to better fit the patient's scalp. This design can reduce the impedance between the electrode and the scalp and improve the stability of signal acquisition.

[0067] In this embodiment, the signal preprocessing unit is used to perform denoising and filtering on the original signal collected by the EEG sensor. In general, the original brain signal will be interfered by various noises, including power frequency noise (50Hz) and motion artifacts. In order to improve the accuracy of the signal, this unit uses a variety of algorithms to optimize the signal.

[0068] Specifically, the preprocessing unit includes a bandpass filter for filtering out DC offsets below 1 Hz and high-frequency noise above 50 Hz. The passband range of the bandpass filter is 1 Hz to 50 Hz, which can cover important EEG frequency bands such as alpha waves (8-13 Hz) and beta waves (14-30 Hz). As an option, an adaptive filtering algorithm can also be used to dynamically suppress specific noise (such as myoelectric noise).

[0069] In some embodiments, in order to further improve the quality of the signal, the preprocessing unit uses an ICA (independent component analysis) algorithm to separate the main components of the signal. The algorithm can effectively separate the EEG signal from interfering components such as the electro-oculogram and the electrocardiogram, thereby extracting a purer neural signal.

[0070] The feature extraction unit is the core part of the signal acquisition and feature extraction module. This unit extracts multi-scale neural activity features from the preprocessed signal, including variables at the micro, meso and macro levels.

[0071] At the microscopic level, the feature extraction unit uses the Hodgkin-Huxley model to model the membrane potential of neurons. The equation of this model is:

[0072]

[0073] Where: C m : Capacitance of neuronal membrane, used to indicate the energy storage capacity of neuronal membrane; V: Neuronal membrane potential, indicating the potential difference between the inside and outside of the membrane; I stim : External stimulation current, provided by the magnetic field regulation module; I Na , I K , IL : Sodium ion current, potassium ion current and leak current, respectively, which describe the kinetic behavior of ion channels in neuronal membranes.

[0074] The specific expression of ion current is:

[0075] I Na =g Na m 3 h(VE Na )

[0076] I K =g K n 4 (VE K )

[0077] I L =q L (VE L )

[0078] Where: g Na , g K ,q L : The conductance of sodium, potassium and leak ion channels respectively; E Na , E K , E L : are the equilibrium potentials of sodium, potassium and leakage ions respectively; m, h, n: are the ion channel switch variables that depend on the membrane potential.

[0079] At the meso-level, the feature extraction unit analyzes the synchronization of brain regions through the Kuramoto coupling model, whose equation is:

[0080]

[0081] Where: θ i : the phase of the ith brain region, in radians; ω i : the natural frequency of the ith brain region, in Hz; K ij : the coupling strength between brain regions i and j, dimensionless; N: the total number of brain regions.

[0082] At the macro level, the feature extraction unit uses the diffusion tensor model to describe the whole brain cognitive state, and its equation is:

[0083]

[0084] Among them: u: whole-brain cognitive state variable, usually represented by α wave power; D: diffusion coefficient, indicating the speed of information propagation between brain regions; S(U,t): magnetic field stimulation term, determined by the output parameter U of the magnetic field regulation module.

[0085] In practical applications, these multi-scale features will be extracted simultaneously and fed into the dynamic optimization module to ensure the comprehensiveness and accuracy of the optimization calculation;

[0086] Through the signal acquisition and feature extraction module, the system can obtain complete patient brain signal characteristics. These characteristic data provide key support for the subsequent optimization of magnetic field stimulation parameters, and also lay the foundation for closed-loop feedback regulation.

[0087] A magnetic field adjustment module, used to generate and output a magnetic field stimulation signal according to the optimization result, including the frequency, intensity and waveform parameters of the magnetic field;

[0088] The magnetic field regulation module receives the optimal stimulation parameter instructions from the dynamic optimization module and accurately adjusts the frequency, intensity and waveform of the magnetic field according to these instructions. Through this module, the system can non-invasively regulate the patient's neural activity, providing key support for accurate neural rehabilitation.

[0089] In general, the magnetic field regulation module works closely with the signal acquisition and feature extraction module, the dynamic optimization module and the closed-loop feedback control module. Specifically, the brain signals acquired by the signal acquisition module are passed to the dynamic optimization module for calculation after feature extraction, and the optimization results are then generated by the magnetic field regulation module to act on the patient's brain area. The design of this module fully considers the needs of various treatment scenarios, such as stimulation of deep brain areas, efficient signal transmission and synchronous regulation of multiple target brain areas.

[0090] In this embodiment, the magnetic field regulation module includes the following main components: a magnetic field generator, a multi-channel magnetic coil array and a drive control unit.

[0091] In this embodiment, the magnetic field generator is responsible for generating a controllable electromagnetic signal to provide adaptive magnetic field stimulation to the patient's target brain area.

[0092] Generally, a magnetic field generator can generate magnetic field signals of different frequencies, intensities, and waveforms. The frequency range is 1Hz to 50Hz, which can cover the common neuromodulation frequency bands and is suitable for stimulation needs of superficial and deep brain areas. As an option, the magnetic field strength can be adjusted between 0.1mT and 3mT, a range that can effectively stimulate neural activity while ensuring the safety of the patient. The magnetic field waveform can be selected according to actual needs, such as a sine wave, square wave, or a custom waveform.

[0093] Specifically, the magnetic field generator controls the output of the magnetic field by adjusting the magnitude, frequency and waveform of the input current. The calculation formula for setting the magnetic field strength B is:

[0094]

[0095] Where: B: magnetic field intensity; μ0: vacuum magnetic permeability; n: number of turns of the coil, dimensionless; I: current passing through the coil; I: coil length.

[0096] In some embodiments, in order to improve the accuracy of magnetic field output, the magnetic field generator can change the time-varying characteristics of the magnetic field by adjusting the current waveform function I(t). For example, the current expression of the sinusoidal magnetic field waveform is:

[0097] I(t)=I0sin(2πft)

[0098] Where: I(t): current that changes with time, in A; I0: peak value of current, in A; f: frequency of magnetic field, in Hz; t: time, in seconds.

[0099] Through this design, the magnetic field generator can generate high-precision time-varying magnetic field signals to meet different treatment needs.

[0100] In this embodiment, a multi-channel magnetic coil array is used to transmit magnetic field signals to a target brain region of a patient. As an implementation method, the array is composed of multiple independent magnetic coils, each of which can be controlled individually.

[0101] In general, each magnetic coil is designed as a planar solenoid structure. The magnetic field distribution formula of the planar solenoid is:

[0102]

[0103] Where: B: magnetic field intensity at a point on the axis of the solenoid, unit is T; μ0: vacuum magnetic permeability, unit is the same as above; n: number of turns per unit length of the coil; I: current passing through the coil, unit is A; R: coil radius, unit is m; z: distance from a point on the coil axis to the center of the coil, unit is m.

[0104] Specifically, the multi-channel magnetic coil array can adjust the magnetic field stimulation of the target brain area in real time according to the electrical signal output by the magnetic field generator. For example, when stimulating the patient's frontal lobe, the system will select the front coil in the array and ensure that the magnetic field strength and frequency are adapted to the target brain area by increasing the current I and adjusting the frequency f.

[0105] In some embodiments, the array can also activate multiple coils at the same time to synchronously regulate multiple brain regions. For example, when treating cognitive impairment, the prefrontal lobe and hippocampus can be stimulated at the same time to improve the synchronization of brain regions to enhance the treatment effect.

[0106] The main function of the drive control unit is to receive the magnetic field parameter instructions from the dynamic optimization module and adjust the output of the magnetic field generator and coil array in real time according to these instructions.

[0107] Specifically, the drive control unit includes a high-speed digital signal processor (DSP) for real-time analysis of the parameter instructions output by the optimization module. These parameters include the target frequency, intensity and waveform of the magnetic field.

[0108] As an option, the drive control unit can also monitor real-time brain signals. For example, when the system detects that the patient's brain area activity has not reached the target state, the drive control unit will dynamically adjust the magnetic field parameters, such as increasing the current I or changing the frequency f, to enhance the stimulation effect.

[0109] In some embodiments, the drive control unit also has an automatic magnetic field strength calibration function. When the system is started, the unit measures the actual output magnetic field strength through the built-in magnetic field sensor and compares it with the target value. If the deviation exceeds the preset threshold, the unit will automatically adjust the output current to ensure the accuracy of the magnetic field signal;

[0110] A dynamic optimization module for calculating the optimal path of the magnetic field stimulation signal based on the input multi-scale variables and constraints;

[0111] The dynamic optimization module uses multi-scale modeling and optimization methods, combined with the Hamiltonian principle, to ensure the precise coupling of magnetic field parameters (such as frequency, intensity, and waveform) with the patient's neural state. In actual applications, the calculation results of the dynamic optimization module are directly transmitted to the magnetic field regulation module to guide the output of the magnetic field, and at the same time, it is linked with the closed-loop feedback control module to dynamically update the optimization path to adapt to the patient's real-time changing neural state.

[0112] In general, the dynamic optimization module receives multi-scale feature variables from the signal acquisition and feature extraction module, including neuronal membrane potential, brain region synchronization, and whole-brain cognitive state. These variables are processed by the modeling and optimization units within the module to generate optimal parameters for magnetic field stimulation. As a core module, its design needs to take into account computational efficiency, parameter accuracy, and applicability.

[0113] In this embodiment, the dynamic optimization module includes the following main components: a multi-scale modeling unit, an optimization calculation unit, and a constraint processing unit;

[0114] Among them, the multi-scale modeling unit is used to build a multi-scale neural dynamics model from micro to macro based on the patient's brain signal characteristics. Through this unit, neural activities at the neuron level, brain region synchronization level and whole brain cognitive state level can be integrated to ensure the comprehensiveness of the optimization calculation.

[0115] In one possible implementation, the Hodgkin-Huxley model is used at the microscopic level to describe the dynamic behavior of neuronal membrane potential.

[0116] In this embodiment, the optimization calculation unit calculates the optimal path of the magnetic field parameters based on the Hamiltonian principle by constructing an action function. As an implementation method, the action function S is defined as:

[0117]

[0118] Where: S: action, which represents the total energy consumption from the initial state to the target state; Lagrangian, defined as the difference between the kinetic energy and potential energy of the system; Q = {q1, q2, q3}: multi-scale variables, including microscopic neuronal membrane potential q1 = V, mesoscopic brain region phase synchronization q2 = {θ i} and the macroscopic whole-brain cognitive state q3=u; The time derivative of a multiscale variable represents the rate of change of the variable.

[0119] As an option, the Lagrangian L can be further expressed as:

[0120]

[0121] in: Kinetic energy term, which describes the dynamic changes of multi-scale variables; V(Q,U(t)): Potential energy term, which describes the regulatory effect of magnetic field parameters on neural state.

[0122] In one possible implementation, the unit solves the optimization problem by Lagrange multiplier method. The optimization goal is:

[0123]

[0124] Among them, U(t): magnetic field stimulation parameter, is the main variable of dynamic optimization, represents the output characteristics of magnetic field changing with time t, where intensity U: magnetic field strength, unit is mT (milli-Tesla), frequency f: magnetic field oscillation frequency, unit is Hz (Hertz); waveform parameters: such as sine wave, square wave or other waveform shape control parameters; Q: multi-scale variables; t1, t2: optimized time interval, represents the duration of magnetic field regulation. Usually depends on the design of the treatment plan, such as the period from the start to the end of stimulation;

[0125] At the same time, the following constraints are met:

[0126] g1(U)=U min ≤U(t)≤U max

[0127] g2(U)=f min ≤f(t)≤f max

[0128] g3(U)=patient’s subjective discomfort≤threshold

[0129] Among them: U min : The minimum value of magnetic field strength, in mT (milli-Tesla). Used to avoid the magnetic field strength being too low to produce effective stimulation; U max : The maximum value of the magnetic field strength, in mT (milli-Tesla). It is set to 3mT to ensure that the magnetic field strength is within the patient's safety range; U(t): The function of the change of magnetic field strength over time; f min : The minimum value of the magnetic field frequency, in Hz (Hertz). Set to 1Hz to adjust the low-frequency stimulation of deep brain areas; f max : The maximum value of the magnetic field frequency, in Hz (Hertz). Set to 50Hz, used to adjust the high-frequency stimulation of the superficial brain area; f(t): The function of the change of magnetic field frequency over time;

[0130] The constraint processing unit is used to ensure that the optimization calculation results meet the physical and physiological limitations of the system. For example, the upper and lower limits of the magnetic field intensity U(t) are determined by the patient's tolerance range, and the frequency f(t) limit is determined by the neurophysiological characteristics. This unit dynamically adjusts the Lagrange multiplier λ i To update the constraints in real time to ensure that the calculation results meet the safety and effectiveness requirements;

[0131] Through the multi-scale modeling, optimization calculation and constraint processing of the dynamic optimization module, reliable parameter support is provided for magnetic field regulation, which can achieve precise regulation of the patient's neurological state.

[0132] A closed-loop feedback control module for receiving real-time brain signals to update the system status and adjusting the parameters of the magnetic field stimulation signal based on the feedback;

[0133] Through the closed-loop feedback control module, the system can collect feedback data of the patient's brain signals in real time, and correct the magnetic field output parameters in combination with the optimization results provided by the dynamic optimization module, so as to adapt to the patient's individual needs and dynamic changes in neural activity during the treatment process.

[0134] In general, the closed-loop feedback control module works closely with the signal acquisition and feature extraction module, the dynamic optimization module, and the magnetic field regulation module. Specifically, the real-time brain signal provided by the signal acquisition module is analyzed by the closed-loop feedback module to generate a correction signal corresponding to the patient's current neural state. The correction signal is transmitted to the dynamic optimization module for recalculation, and the output signal of the magnetic field regulation module is further adjusted to form a closed-loop control. This design not only improves the accuracy and efficiency of treatment, but also enhances the adaptability and robustness of the system.

[0135] In this embodiment, the closed-loop feedback control module includes a feedback signal acquisition unit, a dynamic adjustment unit and a control output unit.

[0136] The feedback signal acquisition unit is responsible for real-time monitoring of the patient's brain signal changes and converting these signals into feature variables for analysis. As a possible implementation method, the feedback signal acquisition unit collects the patient's EEG signals, including alpha, beta and gamma waves, through the EEG sensor array shared with the signal acquisition and feature extraction module.

[0137] Generally, the collected signals will be filtered and preprocessed to remove external interference. Specifically, the feedback signal acquisition unit has a built-in fast Fourier transform (FFT) module to extract the frequency domain characteristics of the EEG signal. In addition, it also includes an independent component analysis (ICA) algorithm module to separate non-target components in brain signals (such as electrooculogram and electrocardiogram noise).

[0138] In some embodiments, the unit also integrates biofeedback sensors, such as skin conductance and heart rate variability (HRV) sensors, to monitor the patient's physiological response to magnetic field stimulation. These biofeedback signals can be used as auxiliary variables and analyzed together with the EEG signals.

[0139] In this embodiment, the main function of the dynamic adjustment unit is to update the input variables of the dynamic optimization module according to the real-time feedback signal and generate new magnetic field adjustment parameters. The dynamic adjustment unit uses the real-time collected brain signal characteristics Q(t) and time derivatives to The deviation between the current neural state and the target state is calculated and converted into the adjustment amount of the magnetic field parameters through the correction algorithm.

[0140] Specifically, the dynamic adjustment unit performs calculations based on the following correction formula:

[0141]

[0142] Where: ΔU(t): adjustment of magnetic field parameters, unit is mT; K p , K d , K i : Proportional, differential and integral gain coefficients, dimensionless; e(t): real-time neural state Q(t) and target state Q target The deviation is defined as:

[0143] e(t)=Q target -Q(t)

[0144] The time derivative of the deviation indicates the rate of change of the deviation; ∫e(t)dt: the integral value of the deviation, which is used to reflect the cumulative deviation.

[0145] In a possible implementation, the dynamic adjustment unit also combines constraints to ensure that the adjusted magnetic field parameters meet safety and physiological adaptability requirements. For example:

[0146] U min ≤U(t)+ΔU(t)≤U max

[0147] f min ≤f(t)+Δf(t)≤f max

[0148] In this embodiment, the main function of the control output unit is to send the magnetic field parameter adjustment amount ΔU(t) generated by the dynamic adjustment unit to the magnetic field regulation module and update the magnetic field output signal in real time.

[0149] Generally, the control output unit has a built-in high-precision digital signal processor (DSP) to convert the magnetic field parameter adjustment ΔU(t) into a control signal for the magnetic field generator. These control signals include current adjustment instructions, voltage control instructions, etc., which are used to adjust the magnetic field strength and frequency in real time.

[0150] Specifically, when the dynamic adjustment unit detects the brain area synchronization q2 = {θ i} decreases, the control output unit will increase the magnetic field strength U(t) of the target brain area and reduce the frequency f(t) to improve the synchronization effect. Similarly, when the whole brain cognitive state variable reaches the set target value, the control output unit will appropriately reduce the magnetic field strength U(t) to reduce the patient's discomfort.

[0151] As an option, the control output unit can also dynamically adjust the magnetic field waveform parameters based on the patient's real-time feedback. For example, when the patient reports a feeling of dizziness, the system will switch the magnetic field waveform from a sine wave to a square wave to reduce the stimulation effect caused by the waveform change.

[0152] The above-mentioned closed-loop feedback control module can realize real-time dynamic regulation of the patient's neural state, thereby improving the system's adaptability to complex neural activities and enhancing the accuracy and safety of the treatment process.

[0153] Human-computer interaction and monitoring module, used to provide a user interface to display the patient's treatment status and support parameter setting and remote control;

[0154] The human-computer interaction and monitoring module works in collaboration with other modules (such as the signal acquisition module, dynamic optimization module, and closed-loop feedback control module) to present the patient's treatment status, magnetic field adjustment parameters, and neural feedback data to users (including doctors and patients) in an intuitive form. At the same time, users can use this module to adjust treatment parameters in real time to ensure personalized and efficient treatment.

[0155] Generally speaking, the functions of the human-computer interaction and monitoring module are mainly reflected in three aspects: visualization of treatment data, interactive setting of parameters, and real-time monitoring of system operation status. Specifically, through this module, doctors can view the dynamic change trend of the patient's brain signal in real time, understand the real-time parameters of magnetic field regulation, and make appropriate parameter adjustments based on the patient's treatment response, thereby optimizing the treatment effect.

[0156] In this embodiment, the human-computer interaction and monitoring module includes a display unit, an operation unit and a remote monitoring unit.

[0157] The display unit is used to present the patient's treatment status and the system's operating parameters in real time. As a possible implementation method, the display unit uses a high-resolution LCD screen and combines it with an intuitive graphical interface to display the following information:

[0158] Brain signal characteristic curve: a trend curve showing the changes of the patient's EEG signals (such as alpha wave power, beta wave frequency and gamma wave amplitude) over time. Generally, these signal characteristics are marked with different colors to distinguish EEG activities in different frequency bands.

[0159] Magnetic field adjustment parameters: including real-time magnetic field strength U(t), frequency f(t) and waveform type (such as sine wave or square wave). Specifically, the unit of magnetic field strength is mT, the unit of frequency is Hz, and the waveform type is intuitively displayed through icons.

[0160] Neurological status feedback: Displays the deviation between the patient's real-time neurological status and the target status. The deviation value e(t) is displayed in the form of numerical value and bar graph to help users intuitively understand the treatment effect.

[0161] In some embodiments, the display unit can also present the synchronization distribution of the patient's brain regions in the form of a heat map. For example, different colors are used to indicate the strength of synchronization between brain regions, helping doctors identify brain regions that need to be focused on.

[0162] In this embodiment, the operation unit is used to support the user's interactive operation of the system, including the adjustment of treatment parameters and the setting of treatment targets. Generally, the operation unit includes physical buttons, a touch screen, and a voice control interface.

[0163] Specifically, users can set the following parameters through the operating unit:

[0164] Magnetic field strength range: Set the maximum magnetic field strength U through the slider or digital input method max and minimum strength U min For example, the user can set the intensity range from 0.5mT to 2mT to suit the tolerance of a specific patient.

[0165] Frequency Range: Users can select a predefined frequency range (such as 10Hz to 30Hz) or a custom range to meet different treatment needs.

[0166] Treatment duration: By selecting the treatment time (such as 10 minutes, 20 minutes, etc.), users can flexibly adjust the duration of the treatment plan.

[0167] As an option, the operating unit can also allow the user to define treatment goals. For example, the user can set the alpha wave power increase amplitude of the target brain area (such as a 20% increase), or set the brain area synchronization target (such as a synchronization index of 0.8). The system will automatically adjust the optimization path and magnetic field output parameters according to these goals.

[0168] In a possible implementation, the operation unit also supports quick operation modes, such as "cognitive enhancement mode" or "emotion regulation mode." Each mode corresponds to different magnetic field parameter settings and optimization targets, which is convenient for users to quickly select.

[0169] In this embodiment, the remote monitoring unit is used to realize the doctor's remote access and real-time monitoring function of the system. The remote monitoring unit is connected to the remote server via a wireless network (such as Wi-Fi or 5G) and supports the following functions:

[0170] Real-time data transmission: The patient's brain signal characteristics, magnetic field parameters and treatment status data are uploaded to a remote server in real time for doctors to view on mobile devices (such as mobile phones or tablets).

[0171] Remote parameter adjustment: Doctors can modify system parameters through the remote access interface, such as adjusting the magnetic field intensity range or treatment time.

[0172] Historical data analysis: By accessing the database on the server, doctors can view the patient's treatment history and analyze the effects of different treatment plans.

[0173] As a possible implementation method, the remote monitoring unit also integrates an alarm function. For example, when the patient's subjective discomfort exceeds the safety threshold, the system will send an alarm message to the doctor through the remote monitoring unit. The doctor can intervene in time and adjust the treatment plan.

[0174] The display, operation and remote functions of the above-mentioned human-computer interaction and monitoring module provide the system with powerful operability and flexibility, allowing doctors and patients to easily adjust treatment parameters and monitor treatment effects in real time, thus forming efficient interaction and comprehensive monitoring between users and the system.

[0175] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Intelligent magnetic field neuromodulation rehabilitation system, characterized in that: include: The signal acquisition and feature extraction module is used to collect the patient's brain signals and extract multi-scale variables, including micro-neuron membrane potential, meso-brain region synchronization, and macro-cognitive state; A magnetic field adjustment module, used to generate and output a magnetic field stimulation signal according to the optimization result, including the frequency, intensity and waveform parameters of the magnetic field; A dynamic optimization module for calculating the optimal path of the magnetic field stimulation signal based on the input multi-scale variables and constraints; A closed-loop feedback control module for receiving real-time brain signals to update the system status and adjusting the parameters of the magnetic field stimulation signal based on the feedback; The human-computer interaction and monitoring module is used to provide a user interface to display the patient's treatment status and support parameter setting and remote control.

2. The intelligent magnetic field neuromodulation rehabilitation system according to claim 1, characterized in that: The signal acquisition and feature extraction module includes: An EEG sensor array for collecting brain signals, including brain wave signals of target brain areas; A signal preprocessing unit, used to perform denoising and filtering on brain signals and extract the frequency characteristics and amplitude of the signals; The feature extraction unit is used to extract multi-scale variables, including neuronal membrane potential, phase synchronization between brain regions, and representative indicators of the whole brain cognitive state.

3. The intelligent magnetic field neuromodulation rehabilitation system according to claim 1, characterized in that: The magnetic field regulation module comprises: A magnetic field generator, used for generating a controllable magnetic field signal, wherein the magnetic field parameters include frequency, intensity and waveform; A multi-channel magnetic coil array is used to output magnetic field signals and act on the patient's target brain area; The drive control unit is used to receive the magnetic field signal parameters of the dynamic optimization module and adjust the magnetic field output in real time.

4. The intelligent magnetic field neuromodulation rehabilitation system according to claim 1, characterized in that: The dynamic optimization module includes: A multi-scale modeling unit, used to construct a multi-scale neural dynamics model based on the patient's brain signals, wherein the multi-scale model includes a neuron membrane potential model, a brain region synchronization model, and a whole-brain cognitive state model; An optimization calculation unit, used to calculate the optimal path of the magnetic field signal based on the input multi-scale variables and constraints; The constraint processing unit is used to limit the frequency, intensity and patient tolerance range of the magnetic field signal.

5. The intelligent magnetic field neuromodulation rehabilitation system according to claim 1, characterized in that: The multi-scale modeling unit comprises: Neuronal membrane potential model, used to describe the dynamic behavior of single neuron membrane potential; The brain region synchronization model is used to describe the dynamic behavior of the phase synchronization characteristics between brain regions; A whole-brain cognitive state model is used to describe changes in cognitive states based on diffusion tensor calculations.

6. The intelligent magnetic field neuromodulation rehabilitation system according to claim 1, characterized in that: The optimization calculation unit comprises: Action function building block, used to define Lagrangian function; The solving unit is used to calculate the optimal path of the magnetic field signal based on the action function and the constraint conditions.

7. The intelligent magnetic field neuromodulation rehabilitation system according to claim 1, characterized in that: The closed-loop feedback control module comprises: A feedback signal acquisition unit, used to collect the patient's brain signals in real time and detect changes in brain signals after magnetic field stimulation; A dynamic adjustment unit, used to adjust the input variables of the multi-scale model and recalculate the optimization path according to the feedback signal; The control output unit is used to send the adjusted magnetic field signal parameters to the magnetic field regulation module.

8. The intelligent magnetic field neuromodulation rehabilitation system according to claim 1, characterized in that: The human-computer interaction and monitoring module includes: A display unit, used to display the patient's brain signals, magnetic field signal parameters and treatment status in real time; An operating unit, for providing a user interface to support parameter setting and magnetic field regulation control; Data storage and remote unit to record patient treatment data and support remote control and monitoring.

9. The intelligent magnetic field neuromodulation rehabilitation system according to claim 1, characterized in that: The dynamic optimization module calculates the optimal path of magnetic field stimulation based on the patient's real-time brain signals and satisfies the following constraints: The frequency of the magnetic field is in the range of 1 to 50 Hz; The magnetic field strength is in the range of 0.1 to 3 mT; Patient tolerance to magnetic field stimulation is within the safe threshold.

10. The intelligent magnetic field neuromodulation rehabilitation system according to claim 1, characterized in that: The optimization calculation is optimized based on the multi-scale neural dynamic characteristics of the patient, and the optimization objectives include: Maximize the patient's neurological status; Minimize patient discomfort; Minimize the energy consumption of magnetic field regulation.

Images