Personalized sleep-aiding regulation and control method and system fusing multi-band characteristics and closed-loop feedback

Abstract

The invention relates to a personalized sleep-aiding regulation and control method and system fusing multi-band characteristics and closed-loop feedback, and relates to the technical field of sleep regulation and control. According to the method, a cross-band coupling matrix is constructed, a nerve-physiological synergy index is calculated, environment-brain state interaction potential energy is calculated based on different band powers and environment sound pressure levels, and the nerve-physiological synergy index and the environment-brain state interaction potential energy are combined into a coupling state vector; solving the coupling state vector to judge the stability of the sleep state; on the basis, personalized sleep induction dynamic attractor intensity is calculated, and different sleep regulation and control strategies are generated according to the steady state / unsteady state and the corresponding attractor intensity. According to the invention, the unique sleep dynamics of the user can be understood and self-adaptively regulated.

Description

Technical Field

[0001] This invention relates to the field of sleep regulation technology, and in particular to a personalized sleep-aid regulation method and system that integrates multi-frequency features and closed-loop feedback. Background Technology

[0002] The inherent dynamic variability of EEG signals and the low signal-to-noise ratio in high-noise environments have long conflicted with the stable, precise, and real-time feedback required for personalized sleep aid regulation. This conflict has led to unstable effects of existing sleep aid devices, low user satisfaction, and has hindered the transition from general intervention to precise personalization.

[0003] The drawbacks of existing technologies are as follows: Traditional neurofeedback devices are based on fixed EEG frequency power ratios and induce relaxation through visual or auditory feedback. However, EEG signals are susceptible to noise (such as electromyography artifacts), and fixed thresholds cannot adapt to individual differences. Machine learning-driven sleep monitoring uses accelerometers or simple EEG sensors to analyze movement or coarse EEG patterns through machine learning models, infer sleep stages, and provide generalized feedback. However, accelerometers cannot capture EEG specificity, and the models have poor generalization. Clinical polysomnography uses multichannel physiological signals (EEG, EOG, EMG) combined with expert-annotated sleep stages to provide high-precision sleep assessment and manual feedback intervention. However, the devices are bulky, costly, and the feedback is not real-time, making them unsuitable for everyday home environments.

[0004] Therefore, there is an urgent need to propose a personalized sleep-aid regulation method and system that integrates multi-frequency characteristics and closed-loop feedback. Summary of the Invention

[0005] The main objective of this invention is to provide a personalized sleep aid regulation method that integrates multi-band features and closed-loop feedback, and further to provide a personalized sleep aid regulation system that integrates multi-band features and closed-loop feedback and can run and implement the above method, effectively solving the problems mentioned in the background art.

[0006] The technical solution of the present invention is as follows: Firstly, a personalized sleep-aid regulation method integrating multi-frequency characteristics and closed-loop feedback is proposed, which includes the following steps: S1. Acquire raw EEG signals, simultaneously acquire raw photoplethysmography (PPG) signals and raw environmental audio signals, and perform enhancement processing. S2. Based on the enhanced EEG signal, extract the instantaneous amplitude and instantaneous phase of different frequency bands, calculate the cross-frequency coupling matrix, and calculate the power, pulse wave signal amplitude and ambient sound pressure level of different frequency bands. S3. Calculate the neuro-physiological synergy index based on the cross-band coupling matrix, calculate the environment-brain state interaction potential energy based on the power of different frequency bands and the ambient sound pressure level, combine the neuro-physiological synergy index and the environment-brain state interaction potential energy into a coupled state vector, and solve the coupled state vector to determine the stability of the sleep state. S4. Calculate the strength of the personalized sleep-inducing dynamic attractor and generate a regulation strategy: When the user is a steady-state weak attractor, a low-frequency binaural beat is used, and the frequency difference of the binaural beat is adjusted according to the strength of the personalized sleep-inducing dynamic attractor. The modulation amplitude of the binaural beat is controlled based on the neuro-physiological synergy index. When the user is a steady-state medium attractor, pink noise is used, and the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index, and the modulation depth is set. When the user is a steady-state strong attractor, natural background sound is used, and the modulation amplitude is set. When the user is a non-steady-state weak attractor, high-frequency binaural beats are used, and the frequency difference of the binaural beats is adjusted according to the strength of the personalized sleep-inducing dynamic attractor. The modulation amplitude of the binaural beats is controlled based on the neuro-physiological synergy index. When the user is a non-steady-state medium attractor, amplitude-modulated pink noise is used, and the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index. The modulation depth is set based on the environment-brain state interaction potential energy. When the user is a non-steady-state strong attractor, low-frequency natural sound is combined with amplitude modulation, and the modulation amplitude is set.

[0007] A further improvement of this invention is that step S1 specifically involves: acquiring raw EEG signals, including voltage values ​​of Fp1 and Fp2 in the frontal region, C3 and C4 in the central region, O1 and O2 in the occipital region, and reference electrodes A1 and A2; simultaneously acquiring raw photoplethysmography (PPG) signals reflecting changes in peripheral blood flow and raw environmental audio signals reflecting environmental noise intensity; removing eye movement and electromyography (EMG) artifacts from the raw EEG signals through independent component analysis, and performing frequency domain filtering processing with 50Hz notch filtering and 0.5Hz-100Hz bandpass filtering; applying adaptive threshold motion artifact correction to the raw PPG signals and performing 0.5Hz-5Hz bandpass filtering; and performing A-weighted filtering processing on the raw environmental audio signals; and outputting the enhanced EEG signals. Enhanced pulse wave signal Enhanced audio signal Electroencephalogram (EEG) signal quality indicators Pulse wave signal quality indicators and audio signal quality indicators t represents a point in time.

[0008] A further improvement of the present invention is that step S2 includes the following specific steps: S21. The enhanced EEG signal Perform Morlet wavelet transform to extract frequency band frequency band frequency band frequency band and Instantaneous amplitude of frequency band and instantaneous phase ,in, , The frequency band is 0.5Hz-4Hz. The frequency band is 4Hz-8Hz. The frequency band is 8Hz-13Hz. The frequency band is 13Hz-30Hz. The frequency band is 30Hz-100Hz; S22. Calculate the cross-band coupling matrix reflecting the interaction between neural oscillations based on the instantaneous amplitude and instantaneous phase of different frequency bands. , , ;in, The modulation intensity of the low-frequency oscillation f at time point t on the amplitude of the high-frequency oscillation g is expressed by the following formula: ; in, Let g be the instantaneous amplitude of the high-frequency oscillation at time t. Let f be the instantaneous phase of the low-frequency oscillation at time t, where j is a constant and satisfies , This indicates summing over all time points within the sliding time window; S23, calculate separately frequency band frequency band and Frequency band power, the formula is: ; ; Let i be the power of the i-band at time t. Let N be the instantaneous amplitude of frequency band i at time point t, and N be the number of time points within the sliding time window; output the amplitude at time point t. Frequency band power , Frequency band power as well as Frequency band power ; S24, Based on the enhanced pulse wave signal Obtain the amplitude of the pulse wave signal The amplitude of the pulse wave signal is the difference between the maximum and minimum values ​​of the pulse wave signal within the sliding time window; based on the enhanced audio signal. Obtain the ambient sound pressure level The calculation formula is: ; Indicates the enhanced audio signal The root mean square value within the sliding time window.

[0009] A further improvement of this invention is that the formula for calculating the neuro-physiological synergy index in step S3 is: ; in, Let U represent the neurophysiological synergy index at time point t, where U is a 4×4 weight matrix, b is a 4×1 bias term matrix, and W is a 1×4 weight matrix. Cross-band coupling matrix The traces.

[0010] A further improvement of this invention is that the calculation formula for the potential energy of the environment-brain state interaction in step S3 is as follows: ; in, This represents the potential energy of the environment-brain state interaction at time point t. To represent the individualized sleep attractor center vector that characterizes the user's optimal sleep state, The state covariance matrix is ​​of dimension 2x2. The calculation method is as follows , The environmental sensitivity coefficient is set to 0.8. The average deep sleep period for each user Frequency band power, For users within the sliding time window Frequency band power variance.

[0011] A further improvement of the present invention is that step S3 further includes: S31, the neuro-physiological synergy index Interaction potential with the environment-brain state Combined into a coupled state vector The time-varying covariance matrix is ​​estimated based on the coupled state vector within the sliding time window, quantifying the fluctuation characteristics of the coupled state vector within the sliding time window. The calculation formula is as follows: ; in, Let be the time-varying covariance matrix at time point t. The mean vector of the coupled state vectors within the sliding time window; S32. Calculate the Jacobian matrix that reflects local dynamic characteristics. Solve the characteristic equation and output the eigenvalues. , The characteristic equation is: Where I is a 2×2 identity matrix; obtain the real part of the largest eigenvalue. The calculation formula is: ;in, Eigenvalues The real part, Eigenvalues The real part; S33, Based on the real part of the maximum eigenvalue To assess sleep stability, when When sleep is at home, it is considered to be in a stable state; when When sleep is in a state of non-homeostasis, it is judged that sleep is not at home. At that time, sleep is judged to be in a critical state.

[0012] A further improvement of the present invention is that step S4 includes the following specific steps: S41. Calculate the strength of the personalized sleep-inducing dynamic attractor. The calculation formula is as follows: ; in, The intensity of the personalized sleep-inducing dynamic attractor at time point t. The characteristic scale matrix, ;in, The variance of the neurophysiological synergy index within the sliding time window. The variance of the environment-brain state interaction potential energy within the sliding time window; S42, when and When the signal is determined to be a steady-state weak attractor, a low-frequency binaural beat is used, and the frequency difference of the low-frequency binaural beat is: The modulation amplitude is: ; S43, when and When the signal is determined to be a steady-state medium attractor, pink noise is used, and the modulation frequency of the pink noise is: The modulation depth is: ; S44, when and At that time, it was determined to be a steady-state strong attractor, and natural background sound was used with a modulation amplitude of: .

[0013] S45, when and When the condition is determined to be a non-steady-state weak attractor, a high-frequency binaural beat is used, and the frequency difference of the high-frequency binaural beat is: The modulation amplitude is: ; S46, when and When the signal is determined to be a non-steady-state intermediate attractor, amplitude-modulated pink noise is used with a modulation frequency of: The modulation depth is: , The potential threshold for environment-brain state interaction; S47, when and At that time, it was determined to be a non-steady-state strong attractor, and a combination of low-frequency natural sound and amplitude modulation was used, with the modulation amplitude being: .

[0014] Secondly, a personalized sleep aid regulation system integrating multi-band EEG features and closed-loop neural feedback is proposed. This system includes: a signal acquisition and enhancement module, a cross-band coupling module, an interactive processing module, and a regulation module. The signal acquisition and enhancement module is used to acquire raw EEG signals, simultaneously acquire raw photoplethysmography (PPG) signals and raw environmental audio signals, and perform enhancement processing. The cross-band coupling module is used to extract the instantaneous amplitude and instantaneous phase of different frequency bands based on the enhanced EEG signal, calculate the cross-band coupling matrix, and calculate the power, pulse wave signal amplitude and ambient sound pressure level of different frequency bands. The interactive processing module is used to calculate the neuro-physiological synergy index based on the cross-band coupling matrix, calculate the environment-brain state interaction potential energy based on the power of different frequency bands and the ambient sound pressure level, combine the neuro-physiological synergy index and the environment-brain state interaction potential energy into a coupling state vector, and solve the coupling state vector to determine the stability of the sleep state. The control module is used to calculate the strength of the personalized sleep-inducing dynamic attractor and generate a control strategy: when the user is a steady-state weak attractor, a low-frequency binaural beat is used, the frequency difference of the binaural beats is adjusted according to the strength of the personalized sleep-inducing dynamic attractor, and the modulation amplitude of the binaural beats is controlled based on the neuro-physiological synergy index; when the user is a steady-state moderate attractor, pink noise is used, the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index, and the modulation depth is set; when the user is a steady-state strong attractor, natural background sound is used and the modulation amplitude is set; when the user is a non-steady-state weak attractor, a high-frequency binaural beat is used, the frequency difference of the binaural beats is adjusted according to the strength of the personalized sleep-inducing dynamic attractor, and the modulation amplitude of the binaural beats is controlled based on the neuro-physiological synergy index; when the user is a non-steady-state moderate attractor, amplitude-modulated pink noise is used, the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index, and the modulation depth is set based on the environment-brain state interaction potential energy; when the user is a non-steady-state strong attractor, a combination of low-frequency natural sound and amplitude modulation is used and the modulation amplitude is set.

[0015] The technical effects of this invention are as follows: A personalized sleep-aid modulation method integrating multi-band features and closed-loop feedback was constructed. This method simultaneously acquires EEG, PPG, and audio signals, and employs independent component analysis technology to separate and remove physiological artifacts in real time, fundamentally improving signal quality and reliability and overcoming the low signal-to-noise ratio bottleneck. It adopts cross-band coupling features, abandoning the traditional fixed frequency band power ratio, and innovatively introduces a phase-amplitude coupling matrix to capture the complex dynamic interactions between different EEG frequency bands. This method is more robust to noise and can more fundamentally characterize the neural mechanisms of sleep-wake cycles. An individualized dynamic system model was constructed, quantifying the user's unique heart-brain coupling and environmental sensitivity through the neuro-physiological synergy index and the potential energy of the environment-brain state interaction. This resulted in an intelligent system capable of understanding and adaptively regulating the user's unique sleep dynamics. Technically, it solved the classic challenges of EEG signal processing, feature extraction, and model generalization. At the application level, it transformed from clinical equipment to a precise, portable home solution, providing a feasible path to overcome the challenge of precise personalized sleep aids. Attached Figure Description

[0016] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart illustrating the personalized sleep aid regulation method that integrates multi-band features and closed-loop feedback according to Embodiment 1 of the present invention. Figure 2 This is a schematic diagram of the structure of the personalized sleep aid regulation system that integrates multi-band EEG features and closed-loop neural feedback according to Embodiment 2 of the present invention. Detailed Implementation

[0017] Example 1 presents a personalized sleep aid regulation method integrating multi-band features and closed-loop feedback. This method simultaneously acquires EEG, PPG, and audio signals, and employs independent component analysis to separate and remove physiological artifacts in real time, fundamentally improving signal quality and reliability and overcoming the low signal-to-noise ratio bottleneck. It utilizes cross-band coupling features, abandoning the traditional fixed frequency band power ratio and innovatively introducing a phase-amplitude coupling matrix. This captures the complex dynamic interactions between different EEG frequency bands, exhibiting stronger robustness to noise and more fundamentally characterizing the neural mechanisms of sleep-wake cycles. An individualized dynamic system model is constructed, quantifying the user's unique heart-brain coupling and environmental sensitivity through neurophysiological synergy indices and environment-brain state interaction potentials. This constructs an intelligent system capable of understanding and adaptively regulating the user's unique sleep dynamics. Technically, it solves the classic challenges of EEG signal processing, feature extraction, and model generalization. At the application level, it achieves a transformation from clinical equipment to a precise, portable home solution, providing a feasible path to overcome the challenge of precise personalized sleep aids.

[0018] Personalized sleep-aid modulation methods that integrate multi-frequency characteristics and closed-loop feedback, such as Figure 1 As shown, the specific steps include the following: S1. Acquire raw EEG signals, simultaneously acquire raw photoplethysmography (PPG) signals and raw environmental audio signals, and perform enhancement processing. S2. Based on the enhanced EEG signal, extract the instantaneous amplitude and instantaneous phase of different frequency bands, calculate the cross-frequency coupling matrix, and calculate the power, pulse wave signal amplitude and ambient sound pressure level of different frequency bands. S3. Calculate the neuro-physiological synergy index based on the cross-band coupling matrix, calculate the environment-brain state interaction potential energy based on different frequency band power and ambient sound pressure level, combine the neuro-physiological synergy index and the environment-brain state interaction potential energy into a coupled state vector, and solve the coupled state vector to determine the stability of sleep state. S4. Calculate the strength of the personalized sleep-inducing dynamic attractor and generate a regulation strategy: When the user is a steady-state weak attractor, a low-frequency binaural beat is used, and the frequency difference of the binaural beat is adjusted according to the strength of the personalized sleep-inducing dynamic attractor. The modulation amplitude of the binaural beat is controlled based on the neuro-physiological synergy index. When the user is a steady-state medium attractor, pink noise is used, and the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index, and the modulation depth is set. When the user is a steady-state strong attractor, natural background sound is used and the modulation amplitude is set. When the user is a non-steady-state weak attractor, high-frequency binaural beats are used, and the frequency difference of binaural beats is adjusted according to the intensity of the personalized sleep-inducing dynamic attractor. The modulation amplitude of binaural beats is controlled based on the neuro-physiological synergy index. When the user is a non-steady-state medium attractor, amplitude-modulated pink noise is used, and the modulation frequency of pink noise is controlled based on the neuro-physiological synergy index. The modulation depth is set based on the potential energy of the environment-brain state interaction. When the user is a non-steady-state strong attractor, low-frequency natural sound is combined with amplitude modulation, and the modulation amplitude is set.

[0019] In this embodiment, step S1 specifically involves: acquiring raw EEG signals, including voltage values ​​at Fp1 and Fp2 in the frontal region, C3 and C4 in the central region, O1 and O2 in the occipital region, and reference electrodes A1 and A2; simultaneously acquiring raw photoplethysmography (PPG) signals reflecting changes in peripheral blood flow and raw environmental audio signals reflecting environmental noise intensity; removing eye movement and electromyography (EMG) artifacts from the raw EEG signals through independent component analysis, and performing frequency domain filtering with 50Hz notch filtering and 0.5Hz-100Hz bandpass filtering; applying adaptive threshold motion artifact correction to the raw PPG signals and performing 0.5Hz-5Hz bandpass filtering; and performing A-weighted filtering on the raw environmental audio signals; and outputting the enhanced EEG signals. Enhanced pulse wave signal Enhanced audio signal Electroencephalogram (EEG) signal quality indicators Pulse wave signal quality indicators and audio signal quality indicators t represents a point in time.

[0020] In this embodiment, step S2 includes the following specific steps: S21. Enhanced EEG signals Perform Morlet wavelet transform to extract frequency band frequency band frequency band frequency band and Instantaneous amplitude of frequency band and instantaneous phase ,in, , The frequency band is 0.5Hz-4Hz. The frequency band is 4Hz-8Hz. The frequency band is 8Hz-13Hz. The frequency band is 13Hz-30Hz. The frequency band is 30Hz-100Hz; S22. Calculate the cross-band coupling matrix reflecting the interaction between neural oscillations based on the instantaneous amplitude and instantaneous phase of different frequency bands. , , ;in, The modulation intensity of the low-frequency oscillation f at time point t on the amplitude of the high-frequency oscillation g is expressed by the following formula: ; in, Let g be the instantaneous amplitude of the high-frequency oscillation at time t. Let f be the instantaneous phase of the low-frequency oscillation at time t, where j is a constant and satisfies , This indicates summing over all time points within the sliding time window; S23, calculate separately frequency band frequency band and Frequency band power, the formula is: ; ; Let i be the power of the i-band at time t. Let N be the instantaneous amplitude of frequency band i at time point t, and N be the number of time points within the sliding time window; output the amplitude at time point t. Frequency band power , Frequency band power as well as Frequency band power ; S24, Based on the enhanced pulse wave signal Obtain the amplitude of the pulse wave signal The pulse wave signal amplitude is the difference between the maximum and minimum pulse wave signal values ​​within the sliding time window; based on the enhanced audio signal. Obtain the ambient sound pressure level The calculation formula is: ; Indicates the enhanced audio signal The root mean square value within the sliding time window.

[0021] In this embodiment, the formula for calculating the neuro-physiological synergy index in step S3 is as follows: ; in, Let U represent the neurophysiological synergy index at time point t, where U is a 4×4 weight matrix, b is a 4×1 bias term matrix, and W is a 1×4 weight matrix. Cross-band coupling matrix The traces.

[0022] In this embodiment, the formula for calculating the potential energy of the environment-brain state interaction in step S3 is as follows: ; in, This represents the potential energy of the environment-brain state interaction at time point t. To represent the individualized sleep attractor center vector that characterizes the user's optimal sleep state, The state covariance matrix is ​​of dimension 2x2. The calculation method is as follows , The environmental sensitivity coefficient is set to 0.8. The average deep sleep period for each user Frequency band power, For users within the sliding time window Frequency band power variance.

[0023] In this embodiment, step S3 further includes: S31, Neurophysiological Coordination Index Potential for interaction between environment and brain state Combined into a coupled state vector The time-varying covariance matrix is ​​estimated based on the coupled state vector within the sliding time window, quantifying the fluctuation characteristics of the coupled state vector within the sliding time window. The calculation formula is as follows: ; in, Let be the time-varying covariance matrix at time point t. The mean vector of the coupled state vectors within the sliding time window; S32. Calculate the Jacobian matrix that reflects local dynamic characteristics. Solve the characteristic equation and output the eigenvalues. , The characteristic equation is: Where I is a 2×2 identity matrix; obtain the real part of the largest eigenvalue. The calculation formula is: ;in, Eigenvalues The real part, Eigenvalues The real part; S33, Based on the real part of the maximum eigenvalue To assess sleep stability, when When sleep is at home, it is considered to be in a stable state; when When sleep is in a state of non-homeostasis, it is judged that sleep is not at home. At that time, sleep is judged to be in a critical state.

[0024] In this embodiment, step S4 includes the following specific steps: S41. Calculate the strength of the personalized sleep-inducing dynamic attractor. The calculation formula is as follows: ; in, The intensity of the personalized sleep-inducing dynamic attractor at time point t. The characteristic scale matrix, ;in, The variance of the neurophysiological synergy index within the sliding time window. The variance of the environment-brain state interaction potential energy within the sliding time window; S42, when and When the signal is determined to be a steady-state weak attractor, a low-frequency binaural beat is used, and the frequency difference of the low-frequency binaural beat is: The modulation amplitude is: ; S43, when and When the signal is detected, it is determined to be a steady-state medium attractor. Pink noise is used, and the modulation frequency of the pink noise is: The modulation depth is: ; S44, when and At that time, it was determined to be a steady-state strong attractor, and natural background sound was used with a modulation amplitude of: .

[0025] S45, when and When the signal is determined to be a non-steady-state weak attractor, a high-frequency binaural beat is used, and the frequency difference of the high-frequency binaural beat is: The modulation amplitude is: ; S46, when and When the signal is determined to be a non-steady-state intermediate attractor, amplitude-modulated pink noise is used with a modulation frequency of: The modulation depth is: , The potential threshold for environment-brain state interaction; S47, when and At that time, it was determined to be a non-steady-state strong attractor, and a combination of low-frequency natural sound and amplitude modulation was used, with the modulation amplitude being: .

[0026] Example 2: This example proposes a personalized sleep aid regulation system that integrates multi-band EEG features with closed-loop neurofeedback, such as... Figure 2 As shown, it includes: a signal acquisition and enhancement module, a cross-frequency band coupling module, an interactive processing module, and a control module; The signal acquisition and enhancement module is used to acquire raw EEG signals, simultaneously acquire raw photoplethysmography (PPG) signals and raw environmental audio signals, and perform enhancement processing. The cross-band coupling module is used to extract the instantaneous amplitude and instantaneous phase of different frequency bands based on the enhanced EEG signal, calculate the cross-band coupling matrix, and calculate the power, pulse wave signal amplitude and ambient sound pressure level of different frequency bands. The interactive processing module is used to calculate the neuro-physiological synergy index based on the cross-band coupling matrix, calculate the environment-brain state interaction potential energy based on different frequency band power and ambient sound pressure level, combine the neuro-physiological synergy index and the environment-brain state interaction potential energy into a coupled state vector, and solve the coupled state vector to determine the stability of the sleep state. The control module is used to calculate the strength of the personalized sleep-inducing dynamic attractor and generate control strategies: When the user is a steady-state weak attractor, low-frequency binaural beats are used, and the frequency difference of the binaural beats is adjusted according to the strength of the personalized sleep-inducing dynamic attractor, with the modulation amplitude of the binaural beats controlled based on the neuro-physiological synergy index; when the user is a steady-state medium attractor, pink noise is used, and the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index, with the modulation depth set; when the user is a steady-state strong attractor, natural background sound is used, with the modulation amplitude set; when the user is a non-steady-state weak attractor, high-frequency binaural beats are used, and the frequency difference of the binaural beats is adjusted according to the strength of the personalized sleep-inducing dynamic attractor, with the modulation amplitude of the binaural beats controlled based on the neuro-physiological synergy index; when the user is a non-steady-state medium attractor, amplitude-modulated pink noise is used, and the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index, with the modulation depth set based on the environment-brain state interaction potential energy; when the user is a non-steady-state strong attractor, a combination of low-frequency natural sound and amplitude modulation is used, with the modulation amplitude set.

[0027] The parameters and steps of each unit module in the personalized sleep aid regulation system integrating multi-band EEG features and closed-loop neurofeedback of the present invention, as described above, can be referred to in the embodiments of the personalized sleep aid regulation method integrating multi-band features and closed-loop feedback in Example 1 above.

[0028] Example 3: This example provides an electronic device, including a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the above-mentioned personalized sleep aid regulation method that integrates multi-band features and closed-loop feedback by calling the computer program stored in the memory.

[0029] The electronic device can vary considerably depending on its configuration or performance. It may include one or more Central Processing Units (CPUs) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the personalized sleep-aid modulation method integrating multi-band features and closed-loop feedback provided in the above-described method embodiments. The electronic device may also include other components for implementing its functions; for example, it may have wired or wireless network interfaces and input / output interfaces for data input and output. Details will not be elaborated upon in this embodiment.

[0030] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this disclosure can be embodied in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0031] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0032] This invention is described with reference to flowchart illustrations and block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and block diagrams, as well as combinations of blocks in the flowchart illustrations and block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0033] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and boxes Figure 1 The steps of the function specified in one or more boxes.

[0034] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A personalized sleep-aid regulation method integrating multi-frequency characteristics and closed-loop feedback, characterized in that, Includes the following steps: S1. Acquire raw EEG signals, simultaneously acquire raw photoplethysmography (PPG) signals and raw environmental audio signals, and perform enhancement processing. S2. Based on the enhanced EEG signal, extract the instantaneous amplitude and instantaneous phase of different frequency bands, calculate the cross-frequency coupling matrix, and calculate the power, pulse wave signal amplitude and ambient sound pressure level of different frequency bands. S3. Calculate the neuro-physiological synergy index based on the cross-band coupling matrix, calculate the environment-brain state interaction potential energy based on different frequency band power and ambient sound pressure level, combine the neuro-physiological synergy index and the environment-brain state interaction potential energy into a coupled state vector, and solve the coupled state vector to determine the stability of the sleep state. S4. Calculate the strength of the personalized sleep-inducing dynamic attractor and generate a regulation strategy: When the user is a steady-state weak attractor, a low-frequency binaural beat is used, and the frequency difference of the binaural beat is adjusted according to the strength of the personalized sleep-inducing dynamic attractor. The modulation amplitude of the binaural beat is controlled based on the neuro-physiological synergy index. When the user is a steady-state medium attractor, pink noise is used, and the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index, and the modulation depth is set. When the user is a steady-state strong attractor, natural background sound is used and the modulation amplitude is set. When the user is a non-steady weak attractor, a high-frequency binaural beat is used. The frequency difference of the binaural beat is adjusted according to the intensity of the personalized sleep-inducing dynamic attractor, and the modulation amplitude of the binaural beat is controlled based on the neuro-physiological synergy index. When the user is a non-steady-state medium attractor, amplitude-modulated pink noise is used, the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index, and the modulation depth is set based on the environment-brain state interaction potential energy; when the user is a non-steady-state strong attractor, low-frequency natural sound is combined with amplitude modulation and the modulation amplitude is set.

2. The personalized sleep aid regulation method integrating multi-band features and closed-loop feedback according to claim 1, characterized in that: Step S1 specifically involves: acquiring raw EEG signals, including voltage values ​​at Fp1 and Fp2 in the frontal region, C3 and C4 in the central region, O1 and O2 in the occipital region, and reference electrodes A1 and A2; simultaneously acquiring raw photoplethysmography (PPG) signals reflecting changes in peripheral blood flow and raw environmental audio signals reflecting environmental noise intensity; removing eye movement and electromyography (EMG) artifacts from the raw EEG signals through independent component analysis, and performing frequency domain filtering with 50Hz notch filtering and 0.5Hz-100Hz bandpass filtering; applying adaptive threshold motion artifact correction to the raw PPG signals and performing 0.5Hz-5Hz bandpass filtering; and performing A-weighted filtering on the raw environmental audio signals; and outputting the enhanced EEG signals. Enhanced pulse wave signal Enhanced audio signal Electroencephalogram (EEG) signal quality indicators Pulse wave signal quality indicators and audio signal quality indicators t represents a point in time.

3. The personalized sleep aid regulation method integrating multi-band features and closed-loop feedback according to claim 2, characterized in that: Step S2 includes the following specific steps: S21. The enhanced EEG signal Perform Morlet wavelet transform to extract frequency band frequency band frequency band frequency band and Instantaneous amplitude of frequency band and instantaneous phase ,in, , The frequency band is 0.5Hz-4Hz. The frequency band is 4Hz-8Hz. The frequency band is 8Hz-13Hz. The frequency band is 13Hz-30Hz. The frequency band is 30Hz-100Hz; S22, calculate the cross-band coupling matrix reflecting the interaction between neural oscillations based on the instantaneous amplitude and instantaneous phase of different frequency bands. , , ;in, The modulation intensity of the low-frequency oscillation f at time point t on the amplitude of the high-frequency oscillation g is expressed by the following formula: ; in, Let g be the instantaneous amplitude of the high-frequency oscillation at time t. Let f be the instantaneous phase of the low-frequency oscillation at time t, where j is a constant and satisfies , This indicates summing over all time points within the sliding time window; S23, calculate separately frequency band frequency band and Frequency band power, the formula is: ; ; Let i be the power of the i-band at time t. Let N be the instantaneous amplitude of frequency band i at time point t, and N be the number of time points within the sliding time window; output the amplitude at time point t. Frequency band power , Frequency band power as well as Frequency band power ; S24, Based on the enhanced pulse wave signal Obtain the amplitude of the pulse wave signal The amplitude of the pulse wave signal is the difference between the maximum and minimum values ​​of the pulse wave signal within the sliding time window; based on the enhanced audio signal. Obtain the ambient sound pressure level The calculation formula is: ; Indicates the enhanced audio signal The root mean square value within the sliding time window.

4. The personalized sleep aid regulation method integrating multi-band features and closed-loop feedback according to claim 3, characterized in that: The formula for calculating the neurophysiological synergy index in step S3 is as follows: ; in, Let U represent the neurophysiological synergy index at time point t, where U is a 4×4 weight matrix, b is a 4×1 bias term matrix, and W is a 1×4 weight matrix. Cross-band coupling matrix The traces.

5. The personalized sleep aid regulation method integrating multi-band features and closed-loop feedback according to claim 4, characterized in that, The formula for calculating the potential energy of the environment-brain state interaction in step S3 is: ; in, This represents the potential energy of the environment-brain state interaction at time point t. To represent the individualized sleep attractor center vector that characterizes the user's optimal sleep state, The state covariance matrix is ​​of dimension 2x2. The calculation method is as follows , The environmental sensitivity coefficient is set to 0.

8. The average deep sleep period for each user Frequency band power, For users within the sliding time window Frequency band power variance.

6. The personalized sleep aid regulation method integrating multi-band features and closed-loop feedback according to claim 5, characterized in that, Step S3 also includes: S31, the neuro-physiological synergy index Interaction potential with the environment-brain state Combined into a coupled state vector The time-varying covariance matrix is ​​estimated based on the coupled state vector within the sliding time window, quantifying the fluctuation characteristics of the coupled state vector within the sliding time window. The calculation formula is as follows: ; in, Let be the time-varying covariance matrix at time point t. The mean vector of the coupled state vectors within the sliding time window; S32. Calculate the Jacobian matrix that reflects local dynamic characteristics. Solve the characteristic equation and output the eigenvalues. , The characteristic equation is: Where I is a 2×2 identity matrix; obtain the real part of the largest eigenvalue. The calculation formula is: ;in, Eigenvalues The real part, Eigenvalues The real part; S33, Based on the real part of the maximum eigenvalue To assess sleep stability, when When sleep is at home, it is considered to be in a stable state; when When sleep is in a state of non-homeostasis, it is judged that sleep is not at home. At that time, sleep is judged to be in a critical state.

7. The personalized sleep aid regulation method integrating multi-band features and closed-loop feedback according to claim 6, characterized in that: Step S4 includes the following specific steps: S41. Calculate the strength of the personalized sleep-inducing dynamic attractor. The calculation formula is as follows: ; in, The intensity of the personalized sleep-inducing dynamic attractor at time point t. The characteristic scale matrix, ;in, The variance of the neurophysiological synergy index within the sliding time window. The variance of the environment-brain state interaction potential energy within the sliding time window; S42, when and When the signal is determined to be a steady-state weak attractor, a low-frequency binaural beat is used, and the frequency difference of the low-frequency binaural beat is: The modulation amplitude is: ; S43, when and When the signal is determined to be a steady-state medium attractor, pink noise is used, and the modulation frequency of the pink noise is: The modulation depth is: ; S44, when and At that time, it was determined to be a steady-state strong attractor, and natural background sound was used with a modulation amplitude of: ; S45, when and When the condition is determined to be a non-steady-state weak attractor, a high-frequency binaural beat is used, and the frequency difference of the high-frequency binaural beat is: The modulation amplitude is: ; S46, when and When the signal is determined to be a non-steady-state intermediate attractor, amplitude-modulated pink noise is used with a modulation frequency of: The modulation depth is: , The potential threshold for environment-brain state interaction; S47, when and At that time, it was determined to be a non-steady-state strong attractor, and a combination of low-frequency natural sound and amplitude modulation was used, with the modulation amplitude being: .

8. A personalized sleep aid regulation system integrating multi-frequency features and closed-loop feedback, implemented based on the personalized sleep aid regulation method integrating multi-frequency features and closed-loop feedback as described in any one of claims 1-7, characterized in that... The system includes: a signal acquisition and enhancement module, a cross-frequency band coupling module, an interactive processing module, and a control module; The signal acquisition and enhancement module is used to acquire raw EEG signals, simultaneously acquire raw photoplethysmography (PPG) signals and raw environmental audio signals, and perform enhancement processing. The cross-band coupling module is used to extract the instantaneous amplitude and instantaneous phase of different frequency bands based on the enhanced EEG signal, calculate the cross-band coupling matrix, and calculate the power, pulse wave signal amplitude and ambient sound pressure level of different frequency bands. The interactive processing module is used to calculate the neuro-physiological synergy index based on the cross-band coupling matrix, calculate the environment-brain state interaction potential energy based on the power of different frequency bands and the ambient sound pressure level, combine the neuro-physiological synergy index and the environment-brain state interaction potential energy into a coupling state vector, and solve the coupling state vector to determine the stability of the sleep state. The control module is used to calculate the strength of the personalized sleep-inducing dynamic attractor and generate a control strategy: when the user is a steady-state weak attractor, a low-frequency binaural beat is used, the frequency difference of the binaural beats is adjusted according to the strength of the personalized sleep-inducing dynamic attractor, and the modulation amplitude of the binaural beats is controlled based on the neuro-physiological synergy index; when the user is a steady-state moderate attractor, pink noise is used, the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index, and the modulation depth is set; when the user is a steady-state strong attractor, natural background sound is used and the modulation amplitude is set; when the user is a non-steady-state weak attractor, a high-frequency binaural beat is used, the frequency difference of the binaural beats is adjusted according to the strength of the personalized sleep-inducing dynamic attractor, and the modulation amplitude of the binaural beats is controlled based on the neuro-physiological synergy index; when the user is a non-steady-state moderate attractor, amplitude-modulated pink noise is used, the modulation frequency of the pink noise is controlled based on the neuro-physiological synergy index, and the modulation depth is set based on the environment-brain state interaction potential energy; when the user is a non-steady-state strong attractor, a combination of low-frequency natural sound and amplitude modulation is used and the modulation amplitude is set.

Images