Intelligent sleep adjusting method and system based on closed-loop acousto-optic brain wave entrainment

Abstract

The invention relates to an intelligent sleep adjusting method and system based on closed-loop acousto-optic brain wave entrainment, electroencephalogram (EEG) signals are collected and preprocessed in real time through a wireless dry electrode, and high-precision sleep staging is achieved through a deep learning model. The system adopts a fuzzy PID (Proportion Integration Differentiation) controller to dynamically adjust acousto-optic synergistic stimulation: an acoustic module generates binaural beat signals, an optical module outputs blue light (470nm) and amber light (590nm) pulses, and the brain wave entrainment efficiency is enhanced through phase synchronization (the phase difference is less than or equal to 10 degrees) and golden section frequency coupling (fL is equal to 1.618 fA). Parameters are updated every 30 seconds through closed-loop feedback, the sound pressure level (30-50 dB) and the light intensity (10-100 lux) are adjusted according to the Weber-Fechner law, and the response delay is lt; the time is 200 ms. A three-level safety mechanism monitors gamma wave abnormity, epilepsy sample discharge and impedance overrun in real time, and triggers graded protection (alarming, cutting off light stimulation and shutdown). Clinical verifications show that the entrainment success rates of the delta wave and the theta wave respectively reach 71% and 68%, the sleep improvement effect is good, and a safe and efficient intervention scheme is provided for sleep disorders.

Description

Technical Field

[0001] This patent is based on the interdisciplinary integration of neuroscience, biomedical engineering and electronic technology, and constructs a closed-loop sleep regulation system with brainwave entrainment as the core and multimodal co-stimulation as the means. Background Art

[0002] The neural regulation mechanism of brainwave entrainment (BLE) uses combined acoustic and optical stimulation to synchronize the cerebral cortical oscillation frequency to a target frequency band (delta waves: 0.5-4Hz, theta waves: 4-8Hz). The acoustic module uses binaural beat technology, delivering pure tones with a frequency difference of Δf to each ear (e.g., 200Hz in the left ear and 204Hz in the right ear induces 4Hz theta waves). Through auditory nerve integration in the superior olivary nucleus, this generates an EEG rhythm in the thalamus that is consistent with Δf. The optical module, through the retinal-hypothalamic pathway, uses periodic pulses of specific wavelengths (470nm blue light to suppress melatonin and 590nm amber light to promote pineal gland regulation) to directly target circadian clock neurons in the suprachiasmatic nucleus. The synergistic effect of these two modalities significantly improves BLE entrainment efficiency, with experimental data showing a 42% increase in synchronization success rate compared to single-modality stimulation (p<0.05).

[0003] This patent uses a multimodal spatiotemporal collaborative optimization model to break through the limitations of the separation of sound and light stimulation in traditional devices and establish a collaborative control model based on phase-frequency coupling. The control model includes time and spectrum dimensions. Time dimension: Phase synchronization technology is used to ensure that the phase difference between the sound pulse and the light flash is ≤10° to avoid multi-sensory conflict; Spectral dimension: The light stimulation frequency f_L and the sound stimulation frequency f_A are set to meet the golden ratio (f_L=1.618f_A), and the resonance characteristics of the auditory and visual pathways are used to enhance the entrainment effect; According to the Weber-Fechner law, the light intensity (10-100 lux) and sound pressure level (30-50dB) are dynamically adjusted to maintain the logarithmic linear relationship between the stimulation intensity and the neural response. At the same time, a closed-loop feedback control system is established, and EEG signals are collected in real time through the prefrontal stem electrode array, combined with an adaptive algorithm to achieve dynamic adjustment. It includes sleep stage recognition: based on the improved AASM standard, millisecond-level stage classification is achieved through the delta wave power ratio (>20% to determine deep sleep) and beta wave burst suppression detection (to identify the wakefulness state); dynamic parameter adjustment: using a fuzzy PID controller, according to the current brain wave frequency error (±0.3Hz accuracy) and phase coherence coefficient (0-1 scale), the stimulation parameters are adjusted in real time, with a response delay of <200ms; a safety protection mechanism automatically performs safety processing when abnormal high-frequency activity is detected to avoid the risk of inducing epilepsy.

[0004] This patent addresses issues such as traditional open-loop preset parameters that ignore individual physiological differences, asynchronous acoustic and visual signals, a mere 19-32% brainwave entrainment efficiency, reliance on offline analysis, and latency exceeding 5 minutes. This patent utilizes a closed-loop feedback system that updates stimulation parameters every 30 seconds to achieve personalized interventions. A spatiotemporal coupling algorithm achieves multimodal phase locking, increasing entrainment success rates to an average of 67%. Furthermore, an edge computing architecture achieves a 200ms real-time response, ensuring real-time intervention.

[0005] This patent introduces closed-loop control theory into the field of multimodal brainwave entrainment for the first time, by establishing a real-time control loop of "perception-decision-execution" (such as Figure 1 This system addresses the three major challenges of traditional technologies: rigid stimulation parameters, insufficient modality coordination, and delayed intervention. Clinically validated, this system can shorten sleep onset latency in insomniacs from 34.2±12.1 minutes to 21.5±8.7 minutes (p=0.008), improving sleep efficiency by 19.3%, providing an innovative technical approach for non-drug interventions for sleep disorders. Summary of the Invention

[0006] In response to the above technical problems, the present invention proposes an intelligent sleep regulation method and system based on closed-loop sound and light brain wave entrainment. This method uses EEG (Electroencephalogram) sleep staging to identify sleep stages (Wake / N1 / N2 / N3 / REM staging) and performs sleep induction for different sleep stages; to ensure safety, a real-time monitoring bypass system is introduced to monitor brain wave status in real time. When the energy changes in the γ frequency band (30-100Hz) are monitored in real time, a primary alarm is triggered when it exceeds 3 standard deviations of the baseline value for 5 seconds. When epileptic discharges are detected (spike frequency>3Hz and amplitude>200μV), secondary intervention is initiated and the light stimulation output is cut off. When the impedance abnormality (>50kΩ) lasts for 10 seconds, the third level protection is activated and the system automatically shuts down; the control component uses a fuzzy PID controller to control the photoelectric frequency output, and the phase synchronization subsystem and the frequency coupling subsystem perform brainwave entrainment (Brainwave Entrainment) neural regulation according to the input.

[0007] The technical solution provided by the present invention includes the following contents.

[0008] The present application provides an intelligent sleep regulation method based on closed-loop acousto-optic brainwave entrainment, comprising: extracting human brain waves; determining sleep stage staging based on the acquired brainwave signals; determining in real time whether a monitoring target appears based on the acquired brainwave signals; determining a fuzzy PID (Proportional-Integral-Derivative) controller to perform acousto-optic frequency control based on the determined sleep stage staging; and determining phase synchronization and frequency coupling based on the determined fuzzy PID to complete brainwave entrainment.

[0009] In some implementation examples, human brain waves are extracted; the original EEG signal is determined based on an 8-channel wireless dry electrode array; the preprocessed EEG signal is determined based on the original signal; and the EEG frequency band decomposition and the EEG energy spectrum of each frequency band are determined based on the preprocessed EEG signal.

[0010] In some implementation examples, the acquired brainwave signals are classified into sleep stages; and the sleep stages are classified according to brainwave frequency bands.

[0011] In some embodiments, the acquired brain wave signals are used to determine in real time whether a monitoring target appears; energy changes in the gamma frequency band (30-100 Hz) are determined based on the acquired brain wave signals; epileptiform discharges are determined based on the acquired brain wave signals; impedance abnormalities are determined based on the acquired brain wave signals;

[0012] In some embodiments, a fuzzy PID controller is used to control the sound and light frequency according to the sleep stage classification; and an audio frequency adjustment value Δf is determined according to a fuzzy rule base. A (domain: ±0.5Hz); Determine the light intensity correction coefficient K based on the fuzzy rule base L (domain: 0.5-1.5);

[0013] In some implementation examples, the phase synchronization and frequency coupling are determined based on the determined fuzzy PID to complete the brainwave entrainment;

[0014] Phase synchronization algorithm ensures that the phase difference between sound and light stimulation and brain waves is ≤10°, enhancing the entrainment effect; frequency coupling is determined according to the golden ratio method;

[0015] Intensity collaborative optimization is performed according to the determined phase synchronization and frequency coupling; and the acousto-optic output is determined according to the determined intensity collaborative optimization result.

[0016] Beneficial effects of the present invention: This closed-loop acoustic and optical brainwave entrainment method effectively improves the success rate of brainwave entrainment. Experiments have shown that the success rate of brainwave entrainment is significantly higher than that of pure sound stimulation and pure light stimulation. Specific comparative data are shown in Table 1.

[0017]

[0018] Table 1

[0019] At the same time, by real-time monitoring of brainwave activity, it is possible to provide personalized and targeted brainwave entrainment neural regulation solutions, solving the problem that the open-loop preset parameters of traditional solutions ignore individual physiological differences. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The scope of the present disclosure may be better understood by reading the following detailed description of exemplary embodiments in conjunction with the accompanying drawings, which include:

[0021] Figure 1 This is an overall flow chart of an intelligent sleep regulation method based on closed-loop sound and light brainwave entrainment provided by an embodiment of the present invention;

[0022] Figure 2 A flowchart of an EEG signal processing method provided by an embodiment of the present invention;

[0023] Figure 3 EEG acquisition location diagram (international 10-20 system standard) provided in an embodiment of the present invention;

[0024] Figure 4 A real-time EEG monitoring flow chart provided by an embodiment of the present invention;

[0025] Figure 5 A flow chart of a fuzzy PID control of acousto-optic frequency provided by an embodiment of the present invention;

[0026] Figure 6 A flowchart of a method for performing brainwave entrainment neural regulation using phase synchronization and frequency coupling provided by an embodiment of the present invention; DETAILED DESCRIPTION

[0027] In practice, traditional open-loop preset parameter schemes ignore individual physiological differences, resulting in a low success rate for EEG neuromodulation. To improve this success rate, the present invention designs a closed-loop EEG neuromodulation method. This method monitors EEG changes in real time and adaptively adjusts the scheme parameters, thereby improving the success rate of EEG neuromodulation. Furthermore, while maintaining safety, a safety alarm bypass system has been designed to promptly alert and automatically address potential safety issues. This solution improves the success rate of EEG neuromodulation while maintaining safety, and can be deployed on embedded computers and dedicated neural network accelerators.

[0028] The brainwave entrainment mind control method comprises:

[0029] Extract human brain waves; determine sleep stage based on the acquired brain wave signals; determine whether the monitored target appears based on the acquired brain wave signals; determine the fuzzy PID controller to perform sound and light frequency control based on the determined sleep stage; determine phase synchronization and frequency coupling based on the determined fuzzy PID to complete brain wave entrainment.

[0030] Step S1:

[0031] In some implementation examples, human brain waves are extracted; the original EEG signal is determined based on an 8-channel wireless dry electrode array; the pre-processed EEG signal is determined based on the original signal; and the EEG frequency band decomposition and the EEG energy spectrum of each frequency band are determined based on the pre-processed EEG signal. Figure 2 .

[0032] Step S11: Extract EEG through dry electrodes at Fp1, Fp2, F3, F4, C3, C4, P3, and P4. The electrode positions are shown in Figure 3 .

[0033] Step S12: Determine a pre-processed EEG signal based on the original signal.

[0034] Step S121: Filtering is performed using a Butterworth filter. The transfer function H(z) can be converted to Step S122: Use a notch filter to eliminate power frequency interference (50Hz or 60Hz) and its harmonics. Use a narrow band filter based on IIR (Infinite Impulse Response). Its parameter center frequency f notch =50Hz, control the narrowness of the impedance bandwidth (quality factor Q=30, bandwidth The transfer function (second-order IIR notch filter) formula is (f s is the sampling rate), r is the pole radius (the stopband is narrower when it is close to 1, such as r = 0.99).

[0035] Step S123: To eliminate myoelectric artifacts or motion interference noise, it is necessary to dynamically track and eliminate the noise. The LMS algorithm formula, the error signal is e[n] = d[n] - w T [n]x[n], d[n] is the noisy signal, x[n] is the reference noise signal, and w[n] is the adaptive weight vector. The weight update formula is w[n+1]=w[n]+μ·e[n]·x[n], where μ is the step size factor (which needs to be adjusted to avoid divergence).

[0036] Step S13: Determine the brainwave frequency band decomposition and the brainwave energy spectrum of each frequency band based on the preprocessed EEG signal. Real-time feature extraction, using short-time Fourier transform (STFT), window length 2 seconds (Hamming window), overlap rate 50%; key parameter extraction, delta wave (0.5-4Hz), power ratio Theta / beta wave power ratio Extracting instantaneous phase using Hilbert transform The formula for calculating phase coherence is: Step S2:

[0037] In some implementation examples, the acquired brainwave signals are classified into sleep stages; and the sleep stages are classified according to brainwave frequency bands.

[0038] Step S21: Sleep staging is performed based on brainwave frequency bands. Deep learning is used for sleep staging. A hybrid model (CNN+LSTM) is used to train staging using spatiotemporal features. Here is an example of the CNN layer formula: x t+k is the input signal (time step), y t is the value of the output feature map, and b is the bias, which is the data offset of the decision surface relative to the origin. Classification accuracy: Wake / N1 / N2 / N3 / REM. Experimental results show that the five-stage classification accuracy reaches 87.3%.

[0039] Step S3:

[0040] In some implementation examples, the acquired brain wave signal is used to determine in real time whether the monitoring target appears. Figure 4 ; Determine the energy change of the gamma frequency band (30-100 Hz) according to the acquired brain wave signal; Determine the epileptic discharge according to the acquired brain wave signal; Determine the impedance abnormality according to the acquired brain wave signal;

[0041] Step S31: Real-time monitoring of the energy change in the γ band (30-100 Hz), triggering a primary alarm when the energy exceeds the baseline value by 3 standard deviations for 5 seconds.

[0042] Step S32: When epileptiform discharges (spike frequency > 3 Hz and amplitude > 200 μV) are detected, secondary intervention is initiated and the light stimulation output is cut off.

[0043] Step S33: When the impedance abnormality (>50kΩ) lasts for 10 seconds, the third level protection is activated and the brain wave entrainment neural regulation is turned off.

[0044] Step S4:

[0045] In some implementation examples, the fuzzy PID controller is determined to perform sound and light frequency control according to the determined sleep stage. The process is shown in FIG. Figure 5. ; Determine the audio frequency adjustment value Δf based on the fuzzy rule base A (domain: ±0.5Hz); Determine the light intensity correction coefficient K based on the fuzzy rule base L (domain: 0.5-1.5);

[0046] Step S41: Determine the fuzzy PID controller for sound and light frequency control based on the determined sleep stage classification, establish a fuzzy rule base, and calculate the matching degree (activation strength) of its premise conditions. Assuming that the premise conditions are connected by logical "AND", the activation strength is the minimum value or product of the membership of the input variables.

[0047] Example:

[0048] Rule 1: IFe_fis Positive(μ1) AND Cis Low(μ2) activation strength w1=min(μ1,μ2)

[0049] Rule 2: IFe_f is Zero(μ3) AND C is High(μ4) Active intensity w2=min(μ3,μ4)

[0050] Rule S42: Determine the audio frequency adjustment value Δf based on the fuzzy rule base A (domain: ±0.5Hz), for the output variable (Δf A ), multiply the activation strength of each rule by its corresponding output value to obtain the weighted output Δf A The weighted sum of The sum of activation strengths of all rules ∑w i =w1+w2; calculate the final output value through weighted average

[0051] Rule S43: Determine the light intensity correction coefficient K based on the fuzzy rule base L (domain: 0.5-1.5), for the output variable (K L ), multiply the activation strength of each rule by its corresponding output value to obtain the weighted output Δf A The weighted sum of The sum of activation strengths of all rules ∑w i =w1+w2; calculate the final output value through weighted average

[0052] Step S5:

[0053] In some implementation examples, the phase synchronization and frequency coupling are determined based on the determined fuzzy PID to complete the brainwave entrainment neural regulation. Figure 6; Determine the phase synchronization algorithm based on fuzzy PID to ensure that the phase difference between the sound and light stimulation and the brain wave is ≤10°, thereby enhancing the entrainment effect; determine the frequency coupling based on the golden ratio method; perform intensity collaborative optimization based on the determined phase synchronization and frequency coupling; determine the sound and light output based on the determined intensity collaborative optimization results.

[0054] S51: The phase synchronization algorithm is determined based on fuzzy PID to ensure that the phase difference between sound and light stimulation and brain waves is ≤10°, avoiding sensory conflict and enhancing the brainwave entrainment effect.

[0055] S52: Determine frequency coupling according to the golden ratio method, and set the optical frequency and the acoustic frequency according to the golden ratio (f_L=1.618f_A) to enhance the resonance effect.

[0056] S53: Based on the determined phase synchronization and frequency coupling, intensity co-optimization is performed. According to the Weber-Fechner law, the sound pressure level formula is applied.

[0057] Dynamic adjustment is performed using the formula SPL (dB) = 20log10(P / P0) + k × ln(EEG_amplitude). Light intensity is dynamically adjusted using the formula I(lux) = I0 × e^(α × δ_power) (α is the attenuation coefficient, and δ_power is the delta wave power fraction). The adjustment strategy involves calculating the energy distribution of the current EEG frequency band every 30 seconds. If the delta wave fraction is >20%, the light intensity is exponentially reduced (to avoid over-suppression). If the theta wave coherence is <0.6, the sound pressure level is linearly increased (maximum 50 dB).

[0058] S54: Determine the acoustic and optical output based on the determined intensity collaborative optimization results. Acoustic module: Binaural beat synthesis, real-time sine wave generation, frequency resolution 0.1Hz, Dynamic Range Control (DRC): Automatic gain adjustment based on ambient noise (microphone feedback); Optical module: LED driver constant current source (0-50mA), PWM dimming frequency 1kHz, pulse waveform rise time <1ms, duty cycle adjustable from 10% to 30%.

Claims

1. An intelligent sleep regulation system based on closed-loop sound and light brainwave entrainment, characterized in that: Includes the following modules: EEG signal acquisition module: configured to acquire EEG signals (EEG) in real time through an 8-channel wireless dry electrode array (Fp1, Fp2, F3, F4, C3, C4, P3, P4); Signal processing module: configured to pre-process the raw EEG signals, including Butterworth bandpass filtering (0.5-30 Hz), notch filtering (center frequency 50 Hz, Q = 30), and LMS adaptive filtering to eliminate myoelectric artifacts; Feature extraction module: configured to decompose brain wave frequency bands (delta wave, theta wave, beta wave, gamma wave) through short-time Fourier transform (STFT) and calculate the power proportion and phase coherence coefficient of each frequency band; Sleep staging module: configured to classify sleep stages (Wake / N1 / N2 / N3 / REM) of brainwave signals based on a hybrid deep learning model (CNN+LSTM); Dynamic Control Module: This module is configured to adjust the sound and light stimulation parameters using a fuzzy PID controller based on sleep stages and real-time brainwave characteristics, including: Acoustic module: Generates binaural beat signals (the frequency difference Δf is the target frequency band, and the frequency resolution is 0.1Hz); Optical module: output periodic light pulses (470nm blue light, 590nm amber light, light intensity 10-100lux, duty cycle 10%-30%); Phase synchronization unit: ensures that the phase difference between sound and light stimulation and brain waves is ≤10°; Frequency coupling unit: sets the sound and light frequency according to the golden ratio (f_L=1.618f_A); Safety protection module: It is configured to monitor gamma wave energy, epileptic discharges and impedance abnormalities in real time, triggering a three-level protection mechanism: Primary alarm: gamma wave energy exceeds the baseline value by 3 standard deviations for 5 seconds; Secondary intervention: light stimulation was cut off when epileptiform discharges were detected (spike frequency > 3 Hz and amplitude > 200 μV); Level 3 protection: The system shuts down when impedance abnormality (>50kΩ) lasts for 10 seconds.

2. The intelligent sleep regulation system according to claim 1, characterized in that: The transfer function of the notch filter is: r = 0.99, where f s is the sampling rate, f notch =50Hz.

3. The intelligent sleep regulation system according to claim 1, characterized in that: The input variables of the fuzzy PID controller include the brain wave frequency error (e f =f target -f EEG , domain ±0.5Hz) and phase coherence coefficient (C, domain 0-1), the output variable is the audio frequency adjustment amount (Δf A ) and light intensity correction factor (K L ), the control rule base calculates the final output through the weighted average method.

4. The intelligent sleep regulation system according to claim 1, characterized in that: The dynamic control module dynamically adjusts the sound pressure level (30-50dB) and light intensity (10-100lux) according to the Weber-Fechner law. The specific formula is: The sound pressure level formula is SPL (dB) = 20log10 (P / P0) + k × ln (EEG_amplitude). The light intensity formula is I (lux) = I0 × e^(α × δ_power).

Images