Alzheimer's disease electrical stimulation system based on cerebrospinal fluid rhythms

Abstract

The application belongs to the technical field of nerve regulation and control engineering, and proposes an Alzheimer's disease electric stimulation system based on cerebrospinal fluid rhythm. First, the first frequency range data and the second frequency range data in the brain wave signal are coupled to obtain cerebrospinal fluid rhythm data. Then, the phase difference between the first frequency range data and the cerebrospinal fluid rhythm data is calculated. Finally, the stimulation frequency is dynamically controlled according to the phase difference, and the stimulation current is dynamically controlled according to the brain tissue water volume fraction. The problem that single frequency stimulation cannot match the specific metabolic needs of non-rapid eye movement sleep and rapid eye movement sleep cycles is solved. The dynamic changes of post-traumatic brain edema, as well as the demand and influence of cerebrospinal fluid rhythm on current control, are considered, the real-time adaptive ability to the dynamic changes of post-traumatic brain edema is improved, and the treatment effect of Alzheimer's disease is guaranteed.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of treatment of neurodegenerative diseases, and particularly relates to an Alzheimer's disease electric stimulation system based on cerebrospinal fluid rhythm. BACKGROUND

[0002] Transcranial electric stimulation physical therapy has the advantages of small operation difficulty and good safety, and has been used in the treatment of Alzheimer's disease, sleep disorders and neuropsychiatric diseases. Transcranial electric stimulation forms a half-current loop by applying a bidirectional micro-current to the cerebral cortex. The flow direction of the current is from the anode to the cathode. This current will cause changes in neuronal electrical activity.

[0003] However, the existing transcranial electric stimulation system adopts a fixed parameter mode (frequency 0.5-2Hz, current 1-2mA), which cannot respond to the characteristic abnormality that the theta-gamma coupling strength of Alzheimer's disease (AD) patients is reduced by 30%-50%, and the phase difference between the open-loop control mechanism and the cerebrospinal fluid pulsation rhythm fluctuates by ±0.8π. Traditional single-frequency stimulation (such as 40Hz gamma wave) fails to synchronize with the 0.1-0.01Hz ultra-slow wave rhythm required by the lymphoid system to remove beta amyloid protein, resulting in a 40%-60% reduction in cerebrospinal fluid-interstitial fluid exchange rate in AD model. At the same time, the existing system ignores the key feature that the blood-brain barrier permeability increases by 2-3 times in AD pathology, resulting in a 35%-50% error in stimulation current density (P>0.8 mL / (min·g) at the time), and lacks dynamic feedback on the cerebrospinal fluid Aβ42 / P-tau181 ratio (diagnostic specificity >90%), resulting in a mismatch between stimulation intensity and pathological staging. In addition, the verification scheme based on the trauma model has a structural bias of 28% in hippocampal capillary density compared with the 5xFAD transgenic model specific to AD, and the amplitude of cerebrospinal fluid pulsation in AD patients at night is reduced by 57±12%, with rhythm disorder (positively correlated with beta amyloid deposition r=0.72), further highlighting the inadequacy of the existing technology in regulating the lymphoid clearance mechanism. SUMMARY

[0004] The application proposes an Alzheimer's disease electric stimulation system based on cerebrospinal fluid rhythm to solve the above problems, first coupling first frequency range data and second frequency range data in the brain wave signal to obtain cerebrospinal fluid rhythm data; then calculating the phase difference between the first frequency range data and the cerebrospinal fluid rhythm data; finally, dynamically controlling the stimulation frequency according to the phase difference, and dynamically controlling the stimulation current according to the brain tissue water volume fraction, solving the problem that single frequency stimulation cannot match the specific metabolic needs of non-rapid eye movement sleep and rapid eye movement sleep cycles, considering the dynamic changes of post-traumatic brain edema and the needs and influences of cerebrospinal fluid rhythm on current control, improving the real-time adaptive ability to the dynamic changes of post-traumatic brain edema, and ensuring the treatment effect of Alzheimer's disease.

[0005] In order to achieve the above purpose, the application provides an Alzheimer's disease electric stimulation system based on cerebrospinal fluid rhythm, which adopts the following technical scheme:

[0006] An Alzheimer's disease electric stimulation system based on cerebrospinal fluid rhythm, comprising:

[0007] A data acquisition module configured to acquire brain wave signals of an Alzheimer's disease patient and brain tissue water volume fraction;

[0008] A decoupling-recombination module configured to determine first frequency range data and second frequency range data in the brain wave signal; coupling the first frequency range data and the second frequency range data to obtain cerebrospinal fluid rhythm data; the frequency of the first frequency range data is lower than the frequency of the second frequency range data;

[0009] A calculation module configured to calculate the phase difference between the first frequency range data and the cerebrospinal fluid rhythm data;

[0010] A control module configured to dynamically control the stimulation frequency according to the phase difference, and dynamically control the stimulation current according to the brain tissue water volume fraction.

[0011] Further, the brain tissue water volume fraction is obtained in real time based on material impedance imaging technology.

[0012] Further, the cerebrospinal fluid rhythm data is:

[0013] ;

[0014] Wherein, is the cerebrospinal fluid rhythm data; is the first frequency range data; is the second frequency range data; and is a preset coefficient.

[0015] Further, the frequency band range of the first frequency band range data is 0.5Hz~4Hz, and the frequency band range of the second frequency band range data is 12Hz~16Hz.

[0016] Further, the stimulation frequency is:

[0017]

[0018] wherein, is the adjusted stimulation frequency; is the initial stimulation frequency; is the phase difference between the first frequency band range data and the cerebrospinal fluid rhythm data.

[0019] Further, the stimulation current is:

[0020]

[0021] wherein, is the adjusted stimulation current; is the initial current; is the brain tissue water volume fraction.

[0022] Further, during the electrical stimulation process, stimulation to the lateral sinus is increased to enhance cerebrospinal fluid reflux.

[0023] Further, during the electrical stimulation process, stimulation to the thalamic reticular nucleus is increased to promote sleep spindle.

[0024] Further, the first frequency band range data is separated by using an empirical mode decomposition algorithm.

[0025] Further, the second frequency band range data is separated by using a multi-scale wavelet packet framework.

[0026] Compared with the prior art, the present application has the following beneficial effects:

[0027] The present application first couples the first frequency band range data and the second frequency band range data in the brain wave signal to obtain cerebrospinal fluid rhythm data; then, calculates the phase difference between the first frequency band range data and the cerebrospinal fluid rhythm data; finally, dynamically controls the stimulation frequency according to the phase difference, and dynamically controls the stimulation current according to the brain tissue water volume fraction; the present application solves the problem that single-frequency stimulation cannot match the specific metabolic needs of non-rapid eye movement sleep and rapid eye movement sleep, considers the dynamic changes of post-traumatic brain edema and the needs and influences of cerebrospinal fluid rhythm on current control, improves the real-time adaptive ability to the dynamic changes of post-traumatic brain edema, and guarantees the treatment effect of Alzheimer's disease. BRIEF DESCRIPTION OF DRAWINGS

[0028] ​​The description and drawings of the specification constituting part of this embodiment serve to provide further understanding of this embodiment, the illustrative embodiments of this embodiment and the description thereof serve to explain this embodiment, and do not constitute undue limitation on this embodiment.

[0029] Figure 1 System framework diagram of embodiment 1 of the present application. DETAILED DESCRIPTION

[0030] The present application will be further described below in conjunction with the drawings and embodiments.

[0031] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0032] Embodiment 1:

[0033] The current post-Alzheimer's electrical stimulation system has the problems of single frequency band stimulation that cannot match the specific metabolic needs of non-rapid eye movement sleep-rapid eye movement sleep (NREM-REM) cycle, traditional electrode interface impedance fluctuation leading to more than 40% decline in stimulation efficiency, and lack of real-time adaptive ability to dynamic changes of post-traumatic brain edema.

[0034] In order to solve at least one of the above problems, the present embodiment provides an Alzheimer's electrical stimulation system based on cerebrospinal fluid rhythm; as shown in Figure 1 The system comprises a data acquisition module, a decoupling-recombination module, a calculation module and a control module; specifically:

[0035] The data acquisition module is configured to acquire the brain wave signal of an Alzheimer's patient and the brain tissue water content (BWC);

[0036] The decoupling-recombination module is configured to determine first frequency band range data and second frequency band range data in the brain wave signal, and to couple the first frequency band range data and the second frequency band range data to obtain cerebrospinal fluid rhythm data; the frequency of the first frequency band range data is lower than the frequency of the second frequency band range data;

[0037] The calculation module is configured to calculate the phase difference between the first frequency band range data and the cerebrospinal fluid rhythm data;

[0038] The control module is configured to dynamically control the stimulation frequency according to the phase difference, and to dynamically control the stimulation current according to the brain tissue water content.

[0039] The embodiment considers the dynamic changes of post-traumatic brain edema and the requirements and influences of cerebrospinal fluid rhythm on current control, improves the real-time adaptive ability to the dynamic changes of post-traumatic brain edema, guarantees the treatment effect of Alzheimer's disease, sleep disorders and craniocerebral trauma, and solves the problem that the single frequency band stimulation cannot match the specific metabolic requirements of non-rapid eye movement sleep and rapid eye movement sleep.

[0040] In some embodiments, the first frequency range data δ (0.5-4Hz), the second frequency range data σ (12-16Hz) and the third frequency range data γ (30-80Hz) in the Electroencephalogram (EEG) signal are separated in real time.

[0041] ;

[0042] Among them, is the cerebrospinal fluid rhythm data; is the first frequency range data; is the second frequency range data; and is a preset coefficient, which is 0.7 and 0.3, respectively.

[0043] In some embodiments, the brain tissue water volume fraction is obtained in real time based on material impedance imaging. The specific impedance imaging method is to apply a micro-current to the brain tissue through a conductive hydrogel electrode (impedance self-adjusting range 50-5000 ), and to detect the difference in current impedance of brain tissue with different water content. The selection of electrode material and the optimization of contact pressure (real-time adjustment using a genetic algorithm with a step size of 0.1N) ensure the stability of signal acquisition. Multi-band signal processing: 0.5-4Hz( δ ), 12-16Hz( σ ) and other specific frequency band Electroencephalogram signals are used to realize signal separation through empirical mode decomposition and multi-scale wavelet packet architecture. Among them, the coupling formula of cerebrospinal fluid rhythm data F = 0.7δ + 0.3σ establishes a correlation model between Electroencephalogram activity and body fluid circulation. Dynamic modeling calculation: based on the exponential relationship, the brain tissue water volume fraction is inversely calculated through impedance change. The model establishes the impedance-water content mapping relationship through real-time calculation of phase difference ΔΦ = δ - F . Closed-loop feedback regulation: impedance data is updated every 200ms, and dynamic adjustment is realized in combination with the stimulation frequency formula to ensure sub-second response capability to brain edema changes.

[0044] In some embodiments, an adaptive interface module is provided, which includes a conductive hydrogel electrode (impedance self-adjusting range 50-5000 The system also includes a material-based genetic algorithm for real-time optimization of contact pressure (in 0.1N steps). Specifically, the core objective of the adaptive interface module is to ensure the stability and accuracy of impedance imaging data by optimizing the electrode-tissue interface characteristics in real time. This includes two key functions: dynamic impedance matching: employing conductive hydrogel electrodes (impedance self-adjustment range 50-5000). This module compensates for impedance fluctuations caused by tissue edema by adjusting the conductivity of the material itself, controlling the contact impedance error within ±15%. Contact pressure optimization: Based on a genetic algorithm, electrode pressure is adjusted in real-time with 0.1N steps to address poor contact caused by skull deformation or patient movement. Experimental data shows that this algorithm can improve pressure distribution uniformity by 62% and significantly reduce motion artifacts. This module simultaneously optimizes electrical characteristics and mechanical contact, enabling the impedance imaging system to maintain a signal fidelity above 0.95 even under dynamic physiological conditions.

[0045] In some embodiments, the stimulation frequency is:

[0046] ;

[0047] in, The adjusted stimulation frequency; This is the initial stimulation frequency; This represents the phase difference between the data in the first frequency band and the cerebrospinal fluid rhythm data.

[0048] The stimulation current is:

[0049] ;

[0050] in, The adjusted stimulation current; The initial current is 2mA; The brain tissue water volume is used; optionally, the brain tissue water volume is detected independently by impedance imaging, directly correlated with the tissue conductivity, to determine the intensity of the stimulation current.

[0051] In conclusion, when δ Waves and F Increased rhythmic phase difference indicates intensified neuronal cluster desynchronization. Reducing the stimulation frequency (down to a minimum of 0.8f0) can promote slow-wave oscillation renormalization, and experimental data show that it can increase the power of sleep spindle waves by 43%. Current regulation: For every 10% increase in brain tissue water volume, conductivity increases by 2.3 times. An exponential decay model is used to compensate for changes in conductivity, ensuring that the stimulation current density remains stable at 0.2-0.5. Within a safe range, animal experiments have verified that it can reduce edema volume by 58%.

[0052] In one embodiment, the principle of regulating brain tissue water volume fraction E when determining the stimulation current size is based on its nonlinear relationship with conductivity: the water content increases by 10%, and the conductivity increases by 2.3 times, resulting in an abnormal increase in current density under the same voltage. For this purpose, exponential decay is used for dynamic compensation: when E = 0.3 (30% water content), the current decays to 74% of the initial value, accurately matching the conductivity change curve (E = 0.98), ensuring that the stimulation current density is stable in the range of 0.2-0.5 E

[0053] Through effect verification (TBI rat model): brain edema volume is reduced by 58% (T2 weighted MRI); deposition around the trauma focus is reduced by 72% (immunofluorescence). Aβ

[0054] In some embodiments, stimulation of the lateral sinus is increased to enhance cerebrospinal fluid reflux, and stimulation of the reticular nucleus of the thalamus is increased to promote sleep spindle waves. Optionally, when stimulating the lateral sinus, a 1 Hz square wave (duty cycle 30%) is used to enhance cerebrospinal fluid reflux; when stimulating the reticular nucleus of the thalamus, a 40 Hz gamma pulse train (pulse width 200 μs) is used to promote sleep spindle waves.

[0055] In some embodiments, dynamic matching is performed: in NREM period, 0.8-1.2 Hz sweep stimulation (sweep speed 0.05 Hz / s) is used to lock the negative phase of δ wave; in REM period, 5 Hz carrier + 80 Hz envelope modulation is used to enhance θ γ cross-frequency coupling; optionally, θ the brain wave is 4 Hz~8 Hz.

[0056] In some embodiments, calcium signals are detected to indicate neuronal activity (GCaMP6s, sampling rate 30 Hz), and endogenous NADH fluorescence is detected to indicate metabolic state (excitation wavelength 340 nm). When the metabolic stress index is triggered; the stability of mitochondrial membrane potential is increased by 3.8 times (JC-1 flow detection).

[0057] Embodiment 2:

[0058] This embodiment provides an Alzheimer's disease electrical stimulation system based on cerebrospinal fluid rhythm, which is based on embodiment 1:

[0059] ​​​​​Independent component analysis (ICA) is combined with prior lead constraints to supplement existing frequency-domain decomposition methods. By constructing a lead topology matrix, electromyographic artifacts and γ oscillatory components can be effectively separated. This algorithm can be integrated into the preprocessing stage of the multi-dimensional feature extraction unit 521.

[0060] The empirical mode decomposition (EMD) algorithm is introduced to separate δ waves, addressing δ the non-stationary nature of the waves; optionally, the IMF list is initialized, and the decomposition is performed using a while loop until the residual energy is below a threshold 1, in each iteration, a new IMF 15 is extracted by the sift function, an adaptive stopping condition (0.3 times the residual standard deviation) is used to control the number of sifting times 13, the newly extracted IMF is added to the list and the residual signal is updated; this algorithm can complement existing power spectral density analysis, and is particularly suitable for separating abnormal δ waves in Alzheimer's patients. An 8-layer wavelet packet decomposition tree is designed to optimize α (12-16 Hz) and γ (30-80 Hz) frequency bands, which can replace existing fixed bandwidth filters, and the resolution of the α frequency band is improved to ±0.5 Hz. A dual-flow CNN-LSTM hybrid network is constructed to replace traditional frequency spectrum analysis, which automatically learns the nonlinear combination of time-frequency features, and forms a cascade structure with existing hierarchical neural networks.

[0061] The weighted phase-lag index (wPLI) is introduced to improve the existing synchronization index, which is not sensitive to volume conduction effects and can more accurately evaluate the cross-frequency coupling strength, supporting α the calculation of the coherence square term.

[0062] Through the original signal → blind source separation → time-frequency decomposition → deep feature extraction, the final output parameters are used by the stimulation encoding unit. When implemented in FPGA, modular design is required, and the processing delay of each stage is controlled within 20 ms to meet the real-time requirements.

[0063] In some embodiments, multi-dimensional parameter optimization is achieved through triple gene encoding: the chromosome structure includes three dimensions of pressure (0.1N step / 0-10N range), frequency (0.1Hz precision / 0.5-100Hz range), and phase shift (5° step / -180~180° range), and 30-bit Gray code encoding is used to improve the local search efficiency by 23%. The fitness function integrates impedance matching (weight 0.4), cerebrospinal fluid flow gain (weight 0.5), and patient comfort (weight 0.1) as three objectives, where F_discomfort is a 0-10 subjective discomfort score. In the dynamic adjustment mechanism, the cerebrospinal fluid flow equation is adjusted by the metabolic pressure index MPI ( The two-factor coupled regulation is achieved through the driving force (MPI) and the edema index: when MPI > 2.5 α Increase tanh by 0.2 (MPI - 2.5) to enhance δ The CSF clearance function of the wave (0.5-4Hz) is also effective, and for every 0.1 increase in the edema index... β The compensation is increased by 0.01 to offset the change in conductivity. Phase synchronization adjustment is achieved through... α _adj=0.7·[1+(wPLI( δ - γ [0.4) / 0.5] is implemented when δ Waves and γ When the weighted phase hysteresis index (wPLI) of the wave (30-80Hz) is greater than 0.4, the cross-frequency coupling strength is increased, resulting in a 45% increase in the density of slow-wave sleep spindle waves. This innovative approach combines metabolic state with changes in tissue electrical properties, and in animal experiments, it increased CSF clearance by 38% while maintaining an impedance matching error of less than 7%.

[0064] The cerebrospinal fluid rhythm-based Alzheimer's disease electrical stimulation system achieves precise monitoring through a multimodal data acquisition module, employing conductive hydrogel electrodes (impedance self-adjusting range 50-5000Ω). The contact pressure was optimized using a genetic algorithm (step size 0.1N) to ensure that the electrode-tissue interface impedance error was controlled within ±15%. The implantable microelectrode array simultaneously detected the Aβ42 / P-tau181 ratio in cerebrospinal fluid (sensitivity 0.1 pg / mL) and... θ - γ Phase amplitude coupling strength (5kHz sampling rate). The system uses empirical mode decomposition to extract the θ wave (4-8Hz) and wavelet packet transform to analyze the γ wave (30-100Hz), constructing an θ-γ coupling model: F_AD(t)=0.6θ(t)+0.4γ(t), and dynamically optimizes the weight coefficients through an LSTM network. The dynamic control algorithm adjusts the parameters in real time according to the phase difference Δφ=θ_phase-F_phase (accuracy ±0.1π): when the θ-γ coupling strength <0.35, the γ enhancement mode is activated: f_γ=40Hz+5·(0.5-PAC_θγ), and during N3 sleep, a 0.05Hz ultra-slow wave (pulse width 200ms, 1.2mA) is switched to increase the cerebrospinal fluid-in-tissue fluid (CSF-ISF) exchange rate to (8.7±1.2)×10. -4 min -1 The brain tissue water volume fraction E was inverted by multi-band impedance imaging (0.5-100Hz sweep) and dynamically compensated according to I_adj=I_0·e^{-2.3(E-0.2)}. =0.98), combined with ultrasound Doppler tracking carotid artery pulse phase (Δt_stim=T_cardiac·arctan(φ_CSF / φ_artery) / 2π) to achieve fluid dynamics synchronization optimization. The closed-loop verification system updates parameters every 200 ms, increases frequency by 0.8 Hz when Δφ>0.5π, and automatically enhances the γ stimulation intensity (I_hotspot=0.8+0.1·SUV_PiB) when the Aβ clearance rate monitored by micro PET-CT decreases by <15% / week. A safety mechanism establishes a conductivity-temperature coupling model: T_max=39.5-2.3·log 10 (σ / 0.5), and switches to safety mode (1 Hz @ 0.2 mA) when the local temperature >38.5℃ or the impedance mutation >20%. Verified by a 5xFAD animal model, this system reduces hippocampal Aβ plaque density by 55.3% (p<0.01), increases θ-γ coupling strength by 75%, and restores CSF pulse amplitude to normal levels at night by 82±7%, forming a closed-loop regulation system for Alzheimer's disease characteristics.

[0065] The embodiment is based on the cerebrospinal fluid rhythm of Alzheimer's disease electric stimulation system on the basis of primary traumatic brain injury, and is subjected to multidimensional pathological adaptation modification; the system adds an electrochemical sensor to detect the cerebrospinal fluid Aβ42 / P-tau181 ratio in real time (sensitivity 0.1 pg / mL), and adopts 5 kHz sampling Magnetoencephalography (MEG) to monitor the θ-γ phase amplitude coupling strength, and reconstructs the original δ / β rhythm into a θ / γ coupling model, and introduces 0.05 Hz ultra-slow wave stimulation (I_glymphatic=1.2 mA·sin(2πf_slow t)) to enhance the lymphoid clearance function. In terms of dynamic regulation algorithm, when the θ-γ coupling strength is less than 0.35, the γ band enhancement mode is automatically activated, and the blood brain barrier permeability Ktrans measured by DCE-MRI is dynamically adjusted (I_adjusted=I_0·[1+2.5(Ktrans-0.15)]^-1). The system innovatively realizes sleep cycle synchronous stimulation: 0.05 Hz ultra-slow wave is applied in N3 period to make the CSF-ISF exchange rate increase by 58%; 40 Hz γ wave stimulation is switched in REM period to enhance the phagocytic activity of microglia cells by 37%, and the cerebrospinal fluid dynamics is optimized through ultrasonic Doppler synchronization of carotid artery pulsation phase (Δt_stim=T_cardiac·(φ_CSF-φ_artery) / 2π). Clinical verification shows that the system makes the hippocampal Aβ plaque density decrease by 55.3%, the θ-γ coupling strength increase by 75%, and the night CSF pulsation amplitude recover to 82±7% of the normal level. The system adds Aβ hotspot positioning stimulation (I_hotspot=0.8+0.1·SUV_PiB) of PiB-PET fusion and APOEε4 genotype adaptation algorithm (γ_weight=0.4+0.1·ε4 allele number), and finally forms a closed-loop regulation system for the pathological characteristics of Alzheimer's disease.

[0066] The above only describes the preferred embodiments of the present embodiment and is not intended to limit the present embodiment. Those skilled in the art can make various modifications and changes to the present embodiment. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present embodiment shall be included in the protection scope of the present embodiment.

Claims

1. An Alzheimer's disease electrical stimulation system based on cerebrospinal fluid rhythm, characterized in that, include: The data acquisition module is configured to acquire brain wave signals and brain tissue water volume ratio of Alzheimer's patients. The decoupling-recombination module is configured to: determine a first frequency band range data and a second frequency band range data in the electroencephalogram signal; couple the first frequency band range data and the second frequency band range data to obtain cerebrospinal fluid rhythm data; wherein the frequency of the first frequency band range data is lower than the frequency of the second frequency band range data; The calculation module is configured to calculate the phase difference between the data in the first frequency band range and the cerebrospinal fluid rhythm data; The control module is configured to: dynamically control the stimulation frequency based on the phase difference, and dynamically control the stimulation current based on the brain tissue water volume ratio; The cerebrospinal fluid rhythm data are as follows: ； in, For cerebrospinal fluid rhythm data; This refers to data within the first frequency band. Second frequency band range data; and These are preset coefficients; The stimulation frequency is: ； in, The adjusted stimulation frequency; This is the initial stimulation frequency; The phase difference between the data in the first frequency band and the cerebrospinal fluid rhythm data; The stimulation current is: ； in, The adjusted stimulation current; This is the initial current; This refers to the water volume ratio of brain tissue.

2. The Alzheimer's disease electrical stimulation system based on cerebrospinal fluid rhythm as described in claim 1, characterized in that, The brain tissue water volume ratio was obtained in real time using impedance imaging technology.

3. The Alzheimer's disease electrical stimulation system based on cerebrospinal fluid rhythm as described in claim 2, characterized in that, The first frequency band range data has a frequency range of 0.5Hz to 4Hz, and the second frequency band range data has a frequency range of 12Hz to 16Hz.

4. The Alzheimer's disease electrical stimulation system based on cerebrospinal fluid rhythm as described in claim 1, characterized in that, During electrical stimulation, increased stimulation of the parasagittal sinus is used to enhance cerebrospinal fluid return.

5. The Alzheimer's disease electrical stimulation system based on cerebrospinal fluid rhythm as described in claim 4, characterized in that, During electrical stimulation, increased stimulation of the thalamic reticular nucleus is used to promote sleep spindle waves.

6. The Alzheimer's disease electrical stimulation system based on cerebrospinal fluid rhythm as described in claim 1, characterized in that, The data in the first frequency band range is separated using an empirical mode decomposition algorithm.

7. The Alzheimer's disease electrical stimulation system based on cerebrospinal fluid rhythm as described in claim 1, characterized in that, The data in the second frequency band range is separated using a multi-scale wavelet packet architecture.

Images