CN-121400846-B - Sleep optimization system based on brain wave regulation

CN121400846BCN 121400846 BCN121400846 BCN 121400846BCN-121400846-B

Abstract

The invention discloses a sleep optimization system based on brain wave regulation, which relates to the technical field of sleep health and brain electricity regulation and control and comprises a hierarchical notch dynamic mapping module, a notch resonance compensation module, a cross-layer interlocking calibration module and a notch characteristic self-adaptive memory module, wherein the four modules form closed-loop cooperative work; according to the invention, by utilizing the closed loop cooperative technology of brain wave hierarchical structure and three-dimensional notch feature dynamic mapping, sleep stage adaptive targeting resonance compensation, cross-layer interlocking calibration and stage notch bi-dimensional clustering self-optimization, the problems that an existing sleep optimization system does not utilize brain wave hierarchical division mechanism, single frequency band adjustment adaptability is poor, multi-frequency band adjustment is easy to generate signal interference and low in efficiency, deep sleep is incoherent, wakefulness is remained and the like are solved, accurate adaptation of sleep stage and individual difference is realized, frequency band interference is avoided, deep sleep continuity and occupation ratio are improved, and safe and efficient personalized sleep optimization effects of wakefulness and shallow sleep loiteness are reduced.

Inventors

WANG XIAODONG
LI XIAOYAN
ZHANG LEI
YANG CHANGJIAO
QIU YI
WU YULIN
XUE JIAO
GUO JIANTAO
YAN SHENGJUAN

Assignees

内蒙古医科大学第二附属医院(内蒙古自治区骨科研究所)

Dates

Publication Date: 20260508
Application Date: 20251222

Claims (7)

1. The sleep optimization system based on brain wave regulation is characterized by comprising a sensor assembly for acquiring brain wave signals and physiological signals and a signal output device, and further comprising a hierarchical notch dynamic mapping module, a notch resonance compensation module, a cross-layer interlocking calibration module and a notch characteristic self-adaptive memory module, wherein the four modules form closed-loop cooperative work; The hierarchical notch dynamic mapping module defines and dynamically captures three-dimensional notch feature vectors based on three-level structures of a detection layer, a calculation layer and a regulation layer formed by brain wave slow1 to slow6 frequency sub-bands; The notch resonance compensation module is internally provided with a sleep stage dynamic adaptation sub-module, and dynamically adjusts a notch threshold value and compensation wave parameters in combination with sleep stage characteristics for targeted resonance compensation; The cross-layer interlocking calibration module constructs a three-layer annular interlocking mechanism, verifies the compensation effect in combination with the sleep stage and executes abnormal calibration; the notch feature self-adaptive memory module constructs a feature library through stage notch bi-dimensional clustering and is used for system personalized self-optimization.
2. The sleep optimization system based on brain wave adjustment, as set forth in claim 1, characterized in that the three-dimensional notch feature vector comprises a response time difference notch, an amplitude complementary notch and a cooperative matching notch, wherein the response time difference notch is a time difference from a detection layer signal peak value to a calculation layer signal peak value, the amplitude complementary notch is a ratio of a calculation layer signal amplitude value to a geometric mean value of the detection layer and a regulation layer signal amplitude value, the cooperative matching notch is a harmonic matching degree of three-level signal frequency, the three-level signal is separated by a level notch dynamic mapping module through a digital filtering algorithm, and the three-dimensional notch feature vector is calculated once per second by adopting a level signal peak value alignment algorithm.
3. A sleep optimization system based on brain wave regulation is characterized in that a sleep stage dynamic adaptation sub-module adopts a brain wave physiological signal fusion identification method, combines a slow5slow6 ratio, a slow1slow2 ratio, heart rate variability and respiratory rate in brain signals to identify sleep stages, wherein the sleep stages comprise a shallow sleep stage, a deep sleep stage, a REM stage and a wake stage, a sub-module establishes a sleep stage notch threshold offset table and a stage compensation parameter correction coefficient table, and a stage notch threshold is obtained by superposing a basic notch threshold and a stage offset, so that effective notch secondary confirmation is carried out on a three-dimensional notch feature vector.
4. A sleep optimization system based on brain wave adjustment is characterized in that the frequency band of compensation waves is determined according to the type of a notch, a compensation calculation layer is used for responding to the abnormal time difference notch, a compensation detection layer and a regulation layer are used for compensating the abnormal time difference notch, three-level synchronous compensation is used for cooperatively matching the abnormal time difference notch, the compensation wave frequency is obtained through calculation of the frequency correction coefficient of the current sleep stage and the signal frequency of a target layer, the compensation wave amplitude is obtained through calculation of the amplitude correction coefficient of the current sleep stage and the resonance gain coefficient, and the resonance gain coefficient is 1.5 minus the amplitude complementary notch value.
5. The sleep optimization system based on brain wave modulation as recited in claim 1, wherein the cross-layer interlock calibration module has interlock verification logic of: The abnormal calibration mechanism comprises a notch reconstruction flow and a non-cooperative notch judgment rule, and the notch is judged to be the non-cooperative notch and is suspended for compensation after three continuous reconstructions, and the compensation is suspended and the effective notch is reconfirmed immediately when the sleep stage is switched.
6. The sleep optimization system based on brain wave adjustment according to claim 1, wherein the feature library of the notch feature adaptive memory module is stored by an SQLite database, three-dimensional notch feature vectors, sleep stages, compensation wave parameters and compensation effect data of each sleep period are recorded, the module divides the notch into six subdivision types by adopting a K-means clustering principle, performs staged notch feature update every 7 sleep periods, and recalibrates basic notch threshold intervals of each stage by counting median.
7. The sleep optimization system based on brain wave regulation of claim 1, wherein the system workflow comprises an initialization stage, a sleep stage identification and gap monitoring stage, an effective gap confirmation and resonance compensation stage, a cross-layer interlocking calibration stage and a self-optimization stage, wherein the initialization stage determines an initial basic gap threshold interval through signal acquisition of a complete sleep period and constructs a blank feature library, and the self-optimization stage calculates a score according to a compensation efficiency scoring formula, wherein the score is weighted by a gap convergence time inverse value, a deep sleep duty ratio, a wake-up frequency inverse value and a stage switching compensation interruption rate inverse value.

Description

Sleep optimization system based on brain wave regulation Technical Field The invention relates to the technical field of sleep health and electroencephalogram regulation, in particular to a sleep optimization system based on brain wave regulation. Background Along with the improvement of the requirements of people on sleep quality, a sleep optimization system based on brain wave regulation is widely focused due to the advantages of noninvasive and accurate. The core principle of the system is that by detecting sleep stage brain wave signals, adjusting signals are output in a targeted mode, so that slow wave oscillation is optimized, and sleep repair efficiency is improved. At present, the prior art is mainly divided into two types, namely a single frequency band adjusting scheme, focusing on a specific brain wave frequency band related to sleep, improving sleep by strengthening amplitude or frequency stability of the frequency band, wherein the schemes ignore multi-dimensional function requirements of brain waves in sleep stage, are difficult to adapt to a complex sleep adjusting mechanism of a brain only aiming at adjustment of the single frequency band, and easily cause problems of incoherence of deep sleep, wakefulness, and the like, and a multi-frequency band parallel adjusting scheme, simultaneously carries out parameter adjustment on a plurality of slow frequency bands, but does not consider internal association among the frequency bands, only adopts an irregular parameter superposition mode, and is likely to cause signal interference among the frequency bands and low adjusting efficiency due to lack of pertinence. In recent years, the demonstration study of research institutions such as the North Master and the like clearly proves that the slow oscillation of sleep phase brain waves is not an unordered signal set, but can be divided into 6 sub-bands from slow-1 to slow-6, and further comprises 3 functional layers, namely a detection layer (slow-1/2/3) is responsible for processing somatosensory signals, a calculation layer (slow-4) is responsible for integrating information, and a regulation layer (slow-5/6) dominates sleep state switching and stabilization. The research reveals the hierarchical physiological structure of brain waves in the sleep stage, but the existing sleep optimization system based on brain wave regulation is not aware of the regulation value of the hierarchical relationship, still stays at the level of standing in isolation to be seen as each frequency band or irregular multi-frequency band superposition, fails to realize accurate and natural sleep optimization by utilizing the inherent hierarchical division mechanism of the brain, and causes a great gap between the regulation effect and the user demand of the existing system, so that a brand new regulation scheme based on the physiological hierarchy characteristics of the brain waves is needed. In view of the above, a sleep optimization system based on brain wave modulation is provided to overcome the above-mentioned problems. Disclosure of Invention The invention aims to provide a sleep optimization system based on brain wave regulation, which aims to solve the problems in the background technology. In order to solve the technical problems, the sleep optimization system based on brain wave regulation is characterized by comprising a sensor assembly for acquiring brain wave signals and physiological signals, a signal output device, a hierarchical notch dynamic mapping module, a notch resonance compensation module, a cross-layer interlocking calibration module and a notch characteristic self-adaptive memory module, wherein the four modules form closed-loop cooperative work; The hierarchical notch dynamic mapping module defines and dynamically captures three-dimensional notch feature vectors based on three-level structures of a detection layer, a calculation layer and a regulation layer formed by brain wave slow1 to slow6 frequency sub-bands; The notch resonance compensation module is internally provided with a sleep stage dynamic adaptation sub-module, and dynamically adjusts a notch threshold value and compensation wave parameters in combination with sleep stage characteristics for targeted resonance compensation; The cross-layer interlocking calibration module constructs a three-layer annular interlocking mechanism, verifies the compensation effect in combination with the sleep stage and executes abnormal calibration; the notch feature self-adaptive memory module constructs a feature library through stage notch bi-dimensional clustering and is used for system personalized self-optimization. The three-dimensional notch feature vector comprises a response time difference notch, an amplitude complementation notch and a cooperative matching notch, wherein the response time difference notch is the time difference from a detection layer signal peak value to a calculation layer signal peak value, the amplitude complementation notch is