CN116013470B

CN116013470B - Method, system and device for dynamically adjusting sleep behavior activity level

Info

Publication number: CN116013470B
Application number: CN202310326415.7A
Authority: CN
Inventors: 何将
Original assignee: Anhui Xingchen Zhiyue Technology Co ltd
Current assignee: Anhui Xingchen Zhiyue Technology Co ltd
Priority date: 2023-03-30
Filing date: 2023-03-30
Publication date: 2023-06-09
Anticipated expiration: 2043-03-30
Also published as: CN116013470A

Abstract

The invention provides a method for dynamically regulating the activity level of sleep behavior, which is used for carrying out real-time acquisition and processing on physiological state signals and behavior state signals of a user in the sleep process, generating real-time characteristics of physiological and behavior states, identifying the state of a sleep time phase in real time and generating a sleep time phase curve; performing real-time behavioral energy quantitative analysis, baseline change analysis and extremum harmonic analysis on the physiological and behavioral state real-time characteristics to generate real-time indexes of myotonia level, central exercise capacity and behavioral action level; and carrying out real-time baseline change analysis and extremum harmonic analysis according to the indexes to generate a sleep behavior activity level real-time curve, predicting and generating a sleep behavior activity level trend curve, generating a sleep behavior activity level dynamic regulation strategy and dynamic regulation in real time, generating a sleep behavior activity level regulation report and establishing a personalized behavior regulation long-term database. The invention realizes scientific detection, assessment and training of the activity or inhibition level of sleep behavior.

Description

Method, system and device for dynamically adjusting sleep behavior activity level

Technical Field

The invention relates to the field of dynamic regulation of sleep behavior activity level, in particular to a method, a system and a device for dynamic regulation of sleep behavior activity level.

Background

In the normal sleeping process of human beings, the central nervous movement capacity representation, the muscle tension level, the limb movement and other action levels are periodically changed along with different sleeping time phases, and the active or inhibitory level of the sleeping action with alternating high and low circulation is presented. However, various factors such as age, aging, disease, pain, tiredness, mental stress, mutation in sleep environment, etc., may cause deterioration of sleep quality and abnormality of active level of sleep behavior or level of behavior suppression of the user.

The prior art scheme CN113995939A discloses a sleep music playing method, device and terminal based on brain electrical signals, wherein the method comprises the following steps: acquiring sleep brain electrical signals of a target user; determining a target brain activity level corresponding to the sleep electroencephalogram signal; and determining target music and target volume according to the target brain activity level, and playing the target music according to the target volume. And, prior art scheme CN113926045a discloses an intelligent control method of home textile products for assisting sleep, wherein the method is applied to an intelligent control system of home textile products, the system comprises a temperature sensing device and a pressure sensing device, the method comprises: acquiring first body parameter characteristics of a first user, and calling a first sleep quality assessment model from a sleep assessment model library according to the first body parameter characteristics; obtaining a body temperature change curve of the first user through the temperature sensing device; the stress curve of the first user at different positions of the first home textile product is obtained through the pressure sensing device, and first pressure distribution change information is generated based on the stress curve; respectively inputting the body temperature change curve and the first pressure distribution change information into the first sleep quality assessment model according to time nodes to obtain sleep quality assessment results of all the time nodes; generating a first sleep quality curve according to the sleep quality evaluation results of the time nodes; obtaining an ideal sleep curve according to the first body parameter characteristic and the first sleep habit of the first user; comparing the first sleep quality curve with the ideal sleep curve to obtain a first sleep quality coefficient; and adjusting and controlling the sleep parameters of the first home textile product according to the first sleep quality coefficient. From the above, the prior art scheme stays on the surface layer characteristic analysis and the general induction processing of neurophysiologic signals, brain states and sleep quality, and lacks of clear quantification, real-time evaluation and dynamic regulation of the active behavior level or the inhibition behavior level in the sleep process; meanwhile, in the prior art, the adjustment is finished immediately, each adjustment is re-analyzed and re-adjusted, continuous and inheritable quantization and adjustment are not performed, and a long-term influence model of quantization-adjustment is not established, so that the intervention process is not scientific and personalized.

How to build a comprehensive and deep assessment method and an adjustment framework, and to perform scientific and efficient dynamic assessment, dynamic training or adjustment on the active level or the suppression level of the sleep behavior in the sleeping process of the user according to the individual needs or specific situations of the user, so as to realize the stabilization of the sleeping process and the normalization of the sleep behavior level of the user which are maintained in different sexes, different ages, different physical and psychological states, different sleeping environments and different sleeping phases, and is still a great problem which is still difficult to solve in the current domestic and foreign sleep health management, neuroscience research and clinical medical practice.

Disclosure of Invention

Aiming at the defects and improvement demands of the existing method, the invention aims to provide a method for dynamically regulating the activity level of sleep behaviors, which is characterized in that physiological state signals and behavior state signals of a user sleep process are collected and monitored in real time, signal processing and time frame feature analysis are carried out, the state of the sleep time phase is identified in real time, a myotonia level real-time index, a central movement capacity real-time index and a behavior action level real-time index are extracted, a sleep activity level real-time index is generated, and a sleep activity level real-time prediction index is obtained through prediction analysis, a sleep activity level dynamic regulation strategy is further generated, the user is subjected to real-time dynamic training or regulation, a sleep activity level regulation report is generated after the cyclic dynamic regulation, and a personalized behavior regulation long-term database is established, so that scientific and efficient dynamic assessment, dynamic training or regulation of the activity or inhibition level of the user in different sexes, different ages, different sleep environments and different sleep phases is realized. The invention also provides a system for dynamically adjusting the sleep behavior activity level, which is used for realizing the method. The invention also provides a device for dynamically adjusting the sleep behavior activity level, which is used for realizing the system.

According to the object of the present invention, the present invention proposes a method for dynamic regulation of the activity level of sleep behaviour, comprising the steps of:

the method comprises the steps of collecting and monitoring physiological state signals and behavior state signals of a user in real time, processing the signals and analyzing time frame characteristics, generating physiological state real-time characteristics and behavior state real-time characteristics, identifying sleep time phase states in real time and generating a sleep time phase curve;

performing real-time behavioral energy quantitative analysis, baseline variation analysis and extremum harmonic analysis on the physiological state real-time characteristics and the behavior state real-time characteristics to generate a myotonia level real-time index, a central exercise capacity real-time index and a behavior action level real-time index;

performing real-time foundation line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index to obtain a sleep behavior activity level real-time index and generate a sleep behavior activity level real-time curve, and predicting and calculating to generate a sleep behavior activity level trend curve;

according to a sleep behavior level optimization knowledge base, the sleep time phase curve, the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve, generating a sleep behavior activity level dynamic regulation strategy in real time, and dynamically regulating the behavior activity level of a user in a sleep process in real time;

Repeating the steps to complete the circulation dynamic regulation of all the sleep behavior activity levels, evaluating the dynamic regulation effect, extracting the phase behavior activity correlation coefficient and the behavior level dynamic regulation effect coefficient, generating a sleep behavior activity level regulation report and establishing a personalized behavior regulation long-term database.

More preferably, the specific steps of collecting and monitoring the physiological state signal and the behavior state signal of the sleeping process of the user in real time, processing the signals and analyzing the characteristics to generate the physiological state real-time characteristic and the behavior state real-time characteristic, identifying the sleeping time phase state in real time and generating the sleeping time phase curve further comprise:

the physiological state and the behavior state of the sleeping process of the user are collected and monitored in real time, and the physiological state real-time signal and the behavior state real-time signal are generated;

performing real-time signal processing on the physiological state real-time signal and the behavior state real-time signal to respectively generate physiological state real-time data and behavior state real-time data;

performing real-time feature analysis on the physiological state real-time data and the behavior state real-time data to generate the physiological state real-time feature and the behavior state real-time feature;

And identifying the sleep time phase real-time state according to the physiological state real-time characteristic and the behavior state real-time characteristic, and obtaining the sleep time phase curve.

More preferably, the physiological status signals include at least central nervous physiological signals, autonomic physiological signals, and muscular system physiological signals; the central nervous physiological signals at least comprise an electroencephalogram signal, a magnetoencephalic signal and an oxygen level dependent signal, the autonomic nervous physiological signals at least comprise an electrocardiosignal, a pulse signal, a respiratory signal, an oxygen level signal, a body temperature signal and a skin electric signal, and the muscle system physiological signals at least comprise an oxygen level dependent signal, an electromyographic signal, a skin electric signal and an acceleration signal.

More preferably, the behavior state signal at least comprises a sleep posture position signal and a limb movement signal.

More preferably, the signal processing at least comprises a/D digital-to-analog conversion, resampling, re-referencing, noise reduction, artifact removal, power frequency notch, low-pass filtering, high-pass filtering, band-stop filtering, band-pass filtering, correction processing and time frame division; the correction processing specifically comprises signal correction and prediction smoothing processing of signal data fragments containing artifacts or distortion in signals, and the time frame division refers to moving interception processing of target signals according to a preset time window and a preset time step.

More preferably, the feature analysis includes at least numerical feature, physical feature analysis, time-frequency feature analysis, envelope feature, and nonlinear feature analysis; wherein the numerical features include at least mean, root mean square, maximum, minimum, variance, standard deviation, coefficient of variation, kurtosis, and skewness; the physical characteristics at least comprise time information, duration information, amplitude characteristics, intensity characteristics and frequency characteristics; the time-frequency characteristics at least comprise total power, characteristic frequency band power duty ratio, characteristic frequency band central frequency, heart rate and heart rate variability; the envelope features at least comprise an original signal, an envelope signal, a normalized envelope signal, an envelope mean value, an envelope root mean square, an envelope maximum value, an envelope minimum value, an envelope variance, an envelope standard deviation, an envelope variation coefficient, an envelope kurtosis and an envelope skewness; the nonlinear features include at least entropy features, fractal features, and complexity features.

More preferably, the physiological state real-time characteristic comprises at least the numerical characteristic, the time-frequency characteristic, the envelope characteristic, the nonlinear characteristic of the physiological state signal.

More preferably, the behavioral state real-time characteristics include at least the numerical characteristics, the physical characteristics, and the time-frequency characteristics of the behavioral state signals.

More preferably, the extraction method of the sleep phase curve specifically comprises the following steps:

1) Performing learning training and data modeling on the physiological state real-time characteristics, the behavior state real-time characteristics and the corresponding sleep stage data of the scale sleep user sample through a deep learning algorithm to obtain a sleep time phase automatic stage model;

2) Inputting the physiological state real-time characteristics and the behavior state real-time characteristics of the current user into the sleep time phase automatic stage model to obtain corresponding sleep time phase stage values;

3) And acquiring the sleep time phase stage values of the physiological state real-time characteristic and the behavior state real-time characteristic according to a time sequence, and generating the sleep time phase curve.

More preferably, the specific steps of performing the real-time behavioral energy quantization analysis, the baseline variation analysis and the extremum harmonic analysis on the physiological state real-time feature and the behavioral state real-time feature to generate the myotonia level real-time index, the central motor capacity real-time index and the behavioral action level real-time index further include:

performing real-time central movement capacity analysis, baseline variation analysis and extremum harmonic analysis on central nerve physiological characteristics in the physiological state real-time characteristics, and extracting the central movement capacity real-time index;

Performing real-time myotensor level analysis, baseline variation analysis and extremum harmonic analysis on the physiological characteristics of the muscle system in the physiological state real-time characteristics, and extracting the myotensor level real-time index;

and analyzing the behavior state real-time characteristics in real time, and extracting the behavior action level real-time index.

More preferably, the extremum harmonic analysis is a data analysis method which uses the maximum value, the minimum value, the maximum value and the minimum value of the numerical value array as observation base points, uses the mean value, the median, the quantile, the variance, the variation coefficient, the kurtosis, the skewness, the absolute value mean value, the absolute value median, the absolute value quantile, the absolute value variance, the absolute value variation coefficient, the absolute value mean value, the absolute value kurtosis and the absolute value skewness of the numerical value array as main analysis harmonic items to observe and analyze the extremum fluctuation state and the general trend change of the numerical value array.

More preferably, a specific calculation mode of the extremum harmonic analysis is as follows:

for numerical value arrays

For example, the extremum harmonic value is calculated by

；

wherein ,

is the extremum harmonic value of the value array Y, < ->

To take the absolute value operator, M is a positive integer and max is the maximum value.

More preferably, the calculation and generation method of the central movement capability real-time index comprises the following steps:

1) Collecting the central nervous physiological real-time signal and the autonomic nervous physiological real-time signal in a resting state when a current user wakes up, and carrying out feature analysis and feature value mean value calculation to obtain a nerve resting movement capacity baseline feature index set;

2) Extracting features corresponding to the central nervous physiological real-time signal and the autonomic nervous physiological real-time signal from the physiological state real-time features to generate central physiological state real-time features;

3) Calculating a baseline variation value of a characteristic value in the central physiological state real-time characteristic and a baseline characteristic index value in the nerve resting motor capacity baseline characteristic index set, namely baseline variation analysis, so as to obtain a central motor capacity characterization characteristic relative variation index set;

4) And carrying out extremum harmonic analysis on all indexes in the central movement capability characterization characteristic relative change index set to obtain extremum harmonic values, namely the current central movement capability real-time index.

More preferably, the calculation and generation method of the myopic level real-time index comprises the following steps:

1) Collecting physiological real-time signals of the muscle system in a resting state when a current user wakes up, and carrying out feature analysis and feature value average calculation to obtain a muscle nerve resting behavior baseline feature index set;

2) Extracting the corresponding characteristics of the physiological signals of the muscle system from the physiological state real-time characteristics to generate the muscle physiological state real-time characteristics;

3) Calculating a baseline variation value of a characteristic value in the real-time characteristic of the muscle physiological state and a baseline characteristic index value in the baseline characteristic index set of the muscle nerve rest behavior, namely baseline variation analysis, and obtaining a relative variation index set of the muscle tension characterization characteristic;

4) And carrying out extremum harmonic analysis on all indexes in the relative change index set of the muscle tension characterization characteristic to obtain extremum harmonic values, namely the current real-time index of the muscle tension level.

More preferably, the calculation and generation method of the behavior action level real-time index comprises the following steps:

1) Acquiring the real-time characteristics of the behavior state, analyzing and quantifying the time distribution, duration, motion amplitude, motion frequency, motion intensity and motion regularity of behavior motion, and generating a behavior action level characterization index set;

2) And carrying out weighted fusion calculation on all indexes in the behavior action level representation index set to obtain the current real-time index of the behavior action level.

More preferably, the specific calculation modes of the baseline variation analysis and the baseline variation value are as follows:

For real-valued variables

And its non-zero base line sequence +.>

For the base line change value thereof is +.>

；

wherein ,

respectively real value variable +.>

The base line change value of (2), the ith base line value and the corresponding weight, and N is a positive integer.

More preferably, the specific steps of performing real-time base line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index to obtain a sleep behavior activity level real-time index and generate a sleep behavior activity level real-time curve, and predicting, calculating and generating a sleep behavior activity level trend curve further include:

performing real-time base line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index to obtain the sleep behavior activity level real-time index, and generating or updating the sleep behavior activity level real-time curve;

and carrying out real-time trend analysis and prediction calculation according to the sleep activity level real-time curve to generate or update the sleep activity level trend curve.

More preferably, the method for generating the sleep behavior activity level real-time index and the sleep behavior activity level real-time curve specifically comprises the following steps:

1) Acquiring the standard behavior active curve characteristic baseline index sets of healthy user groups with different sexes, different age groups and different scale numbers under the state of rest in the awake period and the state of motion task in the awake period;

2) Acquiring the current myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index, and calculating a baseline variation value of a rest baseline value and a task baseline value in the standard behavior activity curve characteristic baseline index set of healthy people of the same age group, namely baseline variation analysis, so as to obtain a sleep behavior activity level characteristic variation set;

3) Carrying out extremum harmonic analysis on all indexes in the sleep behavior activity level characteristic variation set to obtain extremum harmonic values, namely the current real-time index of the sleep behavior activity level;

4) And obtaining the sleep behavior activity level real-time index of the whole process according to the time sequence, and generating or updating to obtain the sleep behavior activity level real-time curve.

More preferably, the method for calculating and generating the sleep activity level trend curve comprises the following steps:

1) Acquiring the current sleep behavior activity level real-time index and the current sleep behavior activity level real-time curve of a user;

2) Trend analysis and index prediction are carried out on the sleep behavior activity level real-time curve to obtain a sleep behavior activity level index of the next time frame, and a sleep behavior activity level real-time prediction index is generated;

3) And incorporating the sleep behavior activity level real-time prediction index according to time sequence, and generating or updating the sleep behavior activity level trend curve.

More preferably, the specific steps of generating the sleep behavior activity level dynamic adjustment strategy in real time according to the sleep behavior level optimization knowledge base, the sleep time phase curve, the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve, and performing real-time dynamic adjustment on the behavior activity level of the sleep process of the user further include:

according to a sleep behavior level optimization knowledge base, the sleep time phase curve, the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve, and combining a sleep behavior activity level dynamic regulation purpose, generating the sleep behavior activity level dynamic regulation strategy;

and dynamically adjusting the behavioral activity level of the sleeping process of the user in real time according to the dynamic sleep behavioral activity level adjusting strategy.

More preferably, the dynamic regulation strategy of the sleep behavior activity level at least comprises a regulation mode, an execution part, a regulation method and a regulation intensity; the adjusting mode at least comprises sound, light, smell, electricity, magnetism, ultrasound and sleeping environment, the executing part comprises a head, a neck, a trunk part, left and right upper limbs, left and right lower limbs and various large sensory organs, the adjusting method at least comprises a constant, an increasing curve, a decreasing curve, an exponential curve, a sinusoidal curve, a periodic square wave and a random curve, and the adjusting intensity is determined by the current real-time index of the activity level of the sleeping behavior and the current real-time prediction index of the activity level of the sleeping behavior.

More preferably, the steps are repeated to complete the cyclic dynamic adjustment of all the sleep behavior activity levels, evaluate the dynamic adjustment effect, extract the phase behavior activity correlation coefficient and the behavior level dynamic adjustment effect coefficient, generate the sleep behavior activity level adjustment report, and build the personalized behavior adjustment long-term database, and the specific steps further include:

completing the cyclic dynamic regulation of all the sleep behavior activity levels to obtain the sleep time phase curve, the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve of all the regulation processes;

Analyzing and calculating relation characteristics of the sleep time phase curve and the sleep behavior activity level real-time curve, and extracting the time phase behavior activity correlation coefficient;

analyzing and calculating relation characteristics of the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve, and extracting the behavior level dynamic adjustment effect coefficient;

according to the sleep time phase curve, the sleep behavior activity level real-time curve, the sleep behavior activity level trend curve, the time phase activity correlation coefficient and the behavior level dynamic regulation effect coefficient, analyzing, calculating and generating the sleep behavior activity level regulation report;

and establishing or updating the personalized behavior adjustment long-term database according to the sleep behavior activity level adjustment report and the current state information of the user, and providing a data analysis inheritance model for the continuous sleep behavior activity level dynamic adjustment of the subsequent user.

More preferably, the method for calculating the phase behavior activity correlation coefficient specifically comprises the following steps:

1) Acquiring the sleep time phase curve and the sleep behavior activity level real-time curve;

2) Analyzing and calculating the relation characteristics of the sleep time phase curve and the sleep behavior activity level real-time curve to obtain a time phase behavior activity level relation characteristic index set;

3) And carrying out weighted fusion calculation on the time phase behavior activity level relation characteristic index set to obtain the time phase behavior activity correlation coefficient.

More preferably, the calculation method of the dynamic adjustment effect coefficient of the behavior level specifically comprises the following steps:

1) Acquiring the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve;

2) Analyzing and calculating relation characteristics of the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve to obtain a behavior level dynamic adjustment effect characteristic index set;

3) And carrying out weighted fusion calculation on the behavior level dynamic adjustment effect characteristic index set to obtain the behavior level dynamic adjustment effect coefficient.

More preferably, the relationship features include at least one of an association feature and a distance feature; wherein the correlation characteristic comprises at least one of a coherence coefficient, a pearson correlation coefficient, a jaccard similarity coefficient, a linear mutual information coefficient, and a linear correlation coefficient, and the distance characteristic comprises at least one of a euclidean distance, a manhattan distance, a chebyshev distance, a minkowski distance, a normalized euclidean distance, a mahalanobis distance, a barbita distance, a hamming distance, and an angle cosine.

More preferably, the sleep activity level adjustment report at least comprises the sleep phase curve, the sleep activity level real-time curve, the sleep activity level trend curve, the phase activity correlation coefficient, the activity level dynamic adjustment effect coefficient, all of the sleep activity level dynamic adjustment strategy, activity level phase distribution statistics, peak activity period summary, low peak activity period summary, abnormal activity period summary and sleep activity level adjustment report summary.

More preferably, the behavioral activity level phase distribution statistics are specifically an average behavioral activity level, a maximum behavioral activity level and a minimum behavioral activity level of different sleep phases; the peak activity time summary is specifically peak time distribution corresponding to a segment exceeding a preset peak threshold value in the sleep behavior activity level real-time curve, and time numerical sum and duty ratio of the peak time distribution; the low peak activity period summary is specifically low peak period distribution corresponding to a segment exceeding a preset low peak threshold value in the sleep behavior activity level real-time curve, and time numerical sum and duty ratio of the low peak period distribution; the abnormal activity period summary is specifically an abnormal period distribution corresponding to an abnormal segment which is separated from a curve baseline trend in the sleep behavior activity level real-time curve, a time value sum and a duty ratio of the abnormal period distribution.

According to the object of the invention, the invention proposes a system for dynamic regulation of the activity level of sleep behaviour, comprising the following modules:

the time phase state analysis module is used for carrying out real-time acquisition monitoring, signal processing and time frame characteristic analysis on physiological state signals and behavior state signals of a sleeping process of a user, generating physiological state real-time characteristics and behavior state real-time characteristics, identifying sleeping time phase states in real time and generating a sleeping time phase curve;

the characterization index extraction module is used for carrying out real-time behavioral energy quantitative analysis, baseline change analysis and extremum harmonic analysis on the physiological state real-time characteristics and the behavioral state real-time characteristics to generate a myotonia level real-time index, a central movement capacity real-time index and a behavioral movement level real-time index;

the behavior level quantification module is used for carrying out real-time foundation line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index to obtain a sleep behavior activity level real-time index and generate a sleep behavior activity level real-time curve, and predicting, calculating and generating a sleep behavior activity level trend curve;

The dynamic strategy adjustment module is used for generating a sleep behavior activity level dynamic adjustment strategy in real time according to a sleep behavior level optimization knowledge base, the sleep time phase curve, the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve, and dynamically adjusting the behavior activity level of a user in real time in the sleep process;

the behavior regulation report module is used for completing the circulation dynamic regulation of all the sleep behavior activity levels, evaluating the dynamic regulation effect, extracting the phase behavior activity correlation coefficient and the behavior level dynamic regulation effect coefficient, generating a sleep behavior activity level regulation report and establishing a personalized behavior regulation long-term database;

and the data management center module is used for visual display and data operation management of all process data in the system.

More preferably, the phase state analysis module further comprises the following functional units:

the signal real-time monitoring unit is used for collecting and monitoring the physiological state and the behavior state of the sleeping process of the user in real time and generating the physiological state real-time signal and the behavior state real-time signal;

the signal real-time processing unit is used for carrying out real-time signal processing on the physiological state real-time signal and the behavior state real-time signal to respectively generate physiological state real-time data and behavior state real-time data;

The feature real-time analysis unit is used for carrying out real-time feature analysis on the physiological state real-time data and the behavior state real-time data to generate the physiological state real-time feature and the behavior state real-time feature;

and the time phase state identification unit is used for identifying the sleep time phase real-time state according to the physiological state real-time characteristic and the behavior state real-time characteristic to obtain the sleep time phase curve.

More preferably, the characterization index extraction module further comprises the following functional units:

the central movement capacity analysis unit is used for carrying out central movement capacity analysis, baseline change analysis and extremum harmonic analysis on the central nerve physiological characteristics in real time in the physiological state real-time characteristics, and extracting the central movement capacity real-time index;

the myotonic level analysis unit is used for carrying out real-time myotonic level analysis, baseline change analysis and extremum harmonic analysis on the physiological characteristics of the muscle system in the physiological state real-time characteristics, and extracting the myotonic level real-time index;

and the behavior operation analysis unit is used for carrying out the behavior action level analysis on the behavior state real-time characteristics in real time and extracting the behavior action level real-time index.

More preferably, the behavior level quantization module further comprises the following functional units:

the behavior level quantification unit is used for carrying out real-time foundation line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index to obtain the sleep behavior activity level real-time index and generating or updating the sleep behavior activity level real-time curve;

and the exponential trend prediction unit is used for carrying out trend analysis and prediction calculation in real time according to the sleep activity level real-time curve to generate or update the sleep activity level trend curve.

More preferably, the dynamic policy adjustment module further comprises the following functional units:

the dynamic strategy generation unit is used for generating the dynamic regulation strategy of the sleep behavior activity level according to the sleep behavior level optimization knowledge base, the sleep time phase curve, the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve and combining the purpose of dynamic regulation of the sleep behavior activity level;

and the dynamic strategy execution unit is used for dynamically regulating the activity level of the sleeping process of the user in real time according to the sleep activity level dynamic regulation strategy.

More preferably, the behavior adjustment reporting module further comprises the following functional units:

the circulation dynamic adjusting unit is used for completing circulation dynamic adjustment of all the sleep behavior active levels to obtain the sleep time phase curve, the sleep behavior active level real-time curve and the sleep behavior active level trend curve in all the adjusting process;

the activity correlation analysis unit is used for analyzing and calculating the relation characteristics of the sleep time phase curve and the sleep behavior activity level real-time curve and extracting the time phase behavior activity correlation coefficient;

the adjusting effect analysis unit is used for analyzing and calculating the relation characteristics of the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve and extracting the behavior level dynamic adjusting effect coefficient;

the regulation report generation unit is used for analyzing, calculating and generating the sleep behavior activity level regulation report according to the sleep time phase curve, the sleep behavior activity level real-time curve, the sleep behavior activity level trend curve, the time phase activity correlation coefficient and the behavior level dynamic regulation effect coefficient;

and the behavior adjustment inheritance unit is used for establishing or updating the personalized behavior adjustment long-term database according to the sleep behavior activity level adjustment report and the current state information of the user, and providing a data analysis inheritance model for the continuous sleep behavior activity level dynamic adjustment of the subsequent user.

More preferably, the data management center module further comprises the following specific functional units:

a user information management unit for registering input, editing, inquiry, output and deletion of user basic information;

the data visual management unit is used for visual display management of all data in the system;

and the data operation management unit is used for storing, backing up, migrating and exporting all data in the system.

According to the purpose of the invention, the invention provides a device for dynamically adjusting the activity level of sleep behaviors, which comprises the following modules:

the characterization index extraction module is used for carrying out real-time behavioral energy quantitative analysis, baseline change analysis and extremum harmonic analysis on the physiological state real-time characteristics and the behavioral state real-time characteristics to generate a myotonia level real-time index, a central motor capacity real-time index and a behavioral motion level real-time index;

the behavior adjustment report module is used for completing the circulation dynamic adjustment of all the sleep behavior activity levels, evaluating the dynamic adjustment effect, extracting the phase behavior activity correlation coefficient and the behavior level dynamic adjustment effect coefficient, generating a sleep behavior activity level adjustment report and establishing a personalized behavior adjustment long-term database;

the data visualization tube module is used for performing visualization display management on all data in the device;

And the data operation management module is used for storing, backing up, migrating and exporting all data in the device.

According to the method, the system and the device for dynamically adjusting the sleep behavior activity level, the physiological state signals and the behavior state signals of the sleep process of the user are collected and monitored in real time, the signal processing and the time frame characteristic analysis are carried out, the sleep time phase state is identified in real time, the myotonia level real-time index, the central movement capacity real-time index and the behavior action level real-time index are extracted, the sleep behavior activity level real-time index is generated and is subjected to prediction analysis to obtain the sleep behavior activity level real-time prediction index, the sleep behavior activity level dynamic adjustment strategy is further generated, the user is subjected to real-time dynamic training or adjustment, the sleep behavior activity level adjustment report is generated after the cyclic dynamic adjustment, and the personalized behavior adjustment long-term database is built, so that scientific and efficient dynamic evaluation, dynamic training or adjustment of the sleep activity or inhibition levels of the users with different sexes, different ages, different physical and different sleep environments and different sleep time phases are realized, and the health management and the physiological analysis are assisted.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate and do not limit the invention.

FIG. 1 is a flowchart illustrating a method for dynamically adjusting sleep activity level according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of the module composition of a system for dynamic adjustment of sleep activity level according to an embodiment of the present invention;

fig. 3 is a schematic diagram of a module structure of an apparatus for dynamically adjusting sleep activity level according to an embodiment of the invention.

Detailed Description

In order to more clearly illustrate the objects and technical solutions of the present invention, the present invention will be further described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the embodiments described below are only some, but not all, embodiments of the invention. Other embodiments, which are derived from the embodiments of the invention by a person skilled in the art without creative efforts, shall fall within the protection scope of the invention. It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be arbitrarily combined with each other.

The method, the system and the device for dynamically adjusting the activity level of the sleep behavior, provided by the invention, build a comprehensive and deep evaluation method and an adjustment framework, can scientifically and efficiently dynamically evaluate, dynamically train or adjust the activity or the inhibition level of the sleep behavior of the user in the sleep process according to individual demands or specific situations of the user, and realize the stabilization of the sleep process and the normalization of the sleep behavior level of the user in the sleep process, which are maintained in different sexes, different ages, different physical and psychological states, different sleep environments and different sleep phases.

As shown in fig. 1, a method for dynamically adjusting sleep activity level according to an embodiment of the present invention includes the following steps:

p100: the physiological state signals and the behavior state signals of the sleeping process of the user are collected and monitored in real time, signal processing and time frame feature analysis are carried out, physiological state real-time features and behavior state real-time features are generated, the sleeping time phase state is recognized in real time, and a sleeping time phase curve is generated.

The method comprises the first step of collecting and monitoring physiological states and behavior states of a sleeping process of a user in real time to generate physiological state real-time signals and behavior state real-time signals.

In this embodiment, the physiological status signals include at least a central nervous physiological signal, an autonomic physiological signal, and a muscular system physiological signal; the central nervous physiological signals at least comprise an electroencephalogram signal, a magnetoencephalic signal and an oxygen level dependent signal, the autonomic nervous physiological signals at least comprise an electrocardiosignal, a pulse signal, a respiratory signal, an oxygen signal, a body temperature signal and a skin electric signal, and the muscle system physiological signals at least comprise an oxygen level dependent signal, an electromyographic signal, a skin electric signal and an acceleration signal. The behavioral state signals include at least sleep posture position signals and limb movement signals.

In this embodiment, an electroencephalogram signal is used as a central nerve physiological signal, an electrocardiosignal, a respiratory signal and a blood oxygen signal are used as autonomic nerve physiological signals, and the central nerve physiological signal and the autonomic nerve physiological signal are collected and monitored by a polysomnography recorder. The sampling rate of the electroencephalogram signal and the electrocardiosignal is 1024Hz, the recording electrodes of the electroencephalogram signal are F3, F4, C3, C4 and Cz, the reference electrodes are M1 and M2, and the electrocardiosignal is collected as a left chest V6 lead. The sampling rate of the respiratory signal and the blood oxygen signal is 64Hz, the respiratory signal is from the chest and abdomen belt, and the blood oxygen signal is from the fingertip.

In this embodiment, the myoelectric signal is used as a physiological signal of the muscular system, the limb movement signal is used as a behavior state signal, the myoelectric signal and the limb movement signal are collected and recorded by using a myoelectric and triaxial acceleration composite sensor with multiple positions, the sampling rate of the myoelectric signal is 1024Hz, the sampling rate of the triaxial acceleration signal is 128Hz, and the collecting positions are respectively the outer middle parts of the lower arms and the upper arms of the left hand and the right hand, the outer middle parts of the lower legs and the thighs of the left leg and the right chest.

In an actual use scene, the polysomnography or the polysomnography monitor can be selected according to the specific situation of a user, and the polysomnography monitor is matched with the polysomnography machine and various behavior monitoring devices to select multiple leads of the frontal lobe pre-exercise area, the multiple parts of the central area and the cardiothoracic electrocardio which are closely related to the behavior capacity level to perform the collection and the monitoring of the central nervous state and the autonomic nervous device, and a plurality of different parts of the body and the limbs to perform the collection and the monitoring of the muscle activity and the behavior movement. It is worth to say that the behavior state signal can also be collected and monitored by means of video, radar, bluetooth, WIFI and the like.

And secondly, performing real-time signal processing on the physiological state real-time signal and the behavior state real-time signal to respectively generate physiological state real-time data and behavior state real-time data.

In this embodiment, the signal processing at least includes a/D digital-to-analog conversion, resampling, re-referencing, noise reduction, artifact removal, power frequency notch, low-pass filtering, high-pass filtering, band-stop filtering, band-pass filtering, correction processing, and time frame division; the correction processing specifically comprises signal correction and prediction smoothing processing of signal data fragments containing artifacts or distortion in the signal, and the time frame division refers to moving interception processing of the target signal according to a preset time window and a preset time step.

In this embodiment, the physiological status signal and the behavioral status signal are first preprocessed: removing artifacts from the electroencephalogram signals, correcting, reducing wavelet noise, carrying out 50Hz power frequency notch filtering and carrying out 0.5-70 Hz band-pass filtering; removing artifacts from the electrocardiosignal, correcting, reducing the noise of the wavelet, and carrying out band-pass filtering at 0.1-40 Hz; removing artifacts from the electromyographic signals, correcting, reducing the noise of wavelets, carrying out power frequency notch filtering at 50Hz and 100Hz, and carrying out band-pass filtering at 20-150 Hz; removing artifacts from the triaxial acceleration signals, correcting, reducing wavelet noise, and carrying out 0.1-40 Hz band-pass filtering; the pretreatment of respiratory signals and blood oxygen signals is mainly to remove artifacts, correct signals and filter 2Hz low-pass signals. Secondly, the signals are subjected to sliding segmentation in a preset time step of 15 seconds and a preset time window of 30 seconds to respectively obtain physiological state data and behavior state data, namely, the dynamic adjustment is carried out on the user according to the sleep behavior activity level of the last 30 seconds every 15 seconds.

And thirdly, carrying out real-time feature analysis on the physiological state real-time data and the behavior state real-time data to generate physiological state real-time features and behavior state real-time features.

In this embodiment, the feature analysis at least includes numerical feature, physical feature analysis, time-frequency feature analysis, envelope feature, and nonlinear feature analysis; wherein the numerical features include at least mean, root mean square, maximum, minimum, variance, standard deviation, coefficient of variation, kurtosis, and skewness; the physical characteristics at least comprise time information, duration information, amplitude characteristics, intensity characteristics and frequency characteristics; the time-frequency characteristics at least comprise total power, characteristic frequency band power duty ratio, characteristic frequency band center frequency, heart rate and heart rate variability; the envelope features at least comprise an original signal, an envelope signal, a normalized envelope signal, an envelope mean, an envelope root mean square, an envelope maximum, an envelope minimum, an envelope variance, an envelope standard deviation, an envelope variation coefficient, an envelope kurtosis and an envelope skewness; the nonlinear features include at least entropy features, fractal features, and complexity features.

In this embodiment, the physiological state real-time characteristic at least includes a numerical characteristic, a time-frequency characteristic, an envelope characteristic, and a nonlinear characteristic of the physiological state signal; the behavior state real-time features include at least a numerical feature, a physical feature, and a time-frequency feature of the behavior state signal.

In an actual use scene, proper characteristics can be selected for real-time calculation and characteristic analysis according to specific situations of users, so that different requirements of individual demands, scene operation efficiency and response speed of people are guaranteed.

And fourthly, recognizing the real-time state of the sleep time phase according to the real-time features of the physiological state and the real-time features of the behavior state, and obtaining a sleep time phase curve.

In this embodiment, the method for extracting the sleep phase curve specifically includes:

1) The method comprises the steps of performing learning training and data modeling on physiological state real-time characteristics, behavior state real-time characteristics and corresponding sleep stage data of a scale sleep user sample through a deep learning algorithm to obtain a sleep time phase automatic stage model;

2) Inputting the physiological state real-time characteristics and the behavior state real-time characteristics of the current user into a sleep time phase automatic stage model to obtain corresponding sleep time phase stage values;

3) And acquiring the sleep time phase stage values of the physiological state real-time characteristic and the behavior state real-time characteristic according to the time sequence, and generating a sleep time phase curve.

In this embodiment, the sleep phase state and the sleep phase stage are identified in real time, which mainly provides a key basis for the subsequent formulation of dynamic adjustment strategies, because different sleep phase states have a relatively large difference in sleep behavior activity level, for example, the sleep behavior activity level during fast eye movement sleep is much higher than that during non-fast eye movement deep sleep stage.

P200: and carrying out behavioral energy quantitative analysis, baseline change analysis and extremum harmonic analysis on the physiological state real-time characteristics and the behavioral state real-time characteristics in real time to generate a myotonia level real-time index, a central motor capacity real-time index and a behavioral action level real-time index.

In this embodiment, the extremum harmonic analysis is a data analysis method that uses the maximum value, the minimum value, the maximum value of the absolute value and the minimum value of the absolute value as the observation base point, uses the average value, the median, the quantile, the variance, the variation coefficient, the kurtosis, the skewness, the absolute value average value, the absolute value median, the absolute value quantile, the absolute value variance, the absolute value variation coefficient, the absolute value average value, the absolute value kurtosis and the absolute value skewness of the numerical value array as the main analysis harmonic items to observe and analyze the extremum fluctuation state and the overall trend change of the numerical value array.

In this embodiment, a specific calculation method of extremum harmonic analysis is as follows:

for numerical value arrays

For example, the extremum harmonic value is calculated by

；

wherein ,

is the extremum harmonic value of the value array Y, < ->

The first step, the central nervous physiological characteristics in the physiological state real-time characteristics are subjected to central movement capacity analysis, baseline change analysis and extremum harmonic analysis in real time, and a central movement capacity real-time index is extracted.

In this embodiment, the calculation and generation method of the real-time index of the central movement capability includes:

1) Collecting central nerve physiological real-time signals and autonomic nerve physiological real-time signals in a resting state when a current user wakes up, and carrying out feature analysis and feature value mean value calculation to obtain a nerve resting motor ability baseline feature index set;

2) Extracting features corresponding to the central nervous physiological real-time signal and the autonomic nervous physiological real-time signal from the physiological state real-time features to generate the central physiological state real-time features;

3) Calculating a baseline variation value of a baseline characteristic index value in the real-time characteristic of the central physiological state and the baseline characteristic index value in the baseline characteristic index set of the nerve resting exercise capacity, namely analyzing the baseline variation to obtain a relative variation index set of the central exercise capacity characterization characteristic;

And secondly, carrying out real-time myotensor level analysis, baseline change analysis and extremum harmonic analysis on the physiological characteristics of the muscle system in the physiological state real-time characteristics, and extracting a myotensor level real-time index.

In this embodiment, the calculation and generation method of the myopic tension level real-time index includes:

1) Collecting physiological real-time signals of a muscle system in a resting state when a current user wakes up, and carrying out feature analysis and feature value average calculation to obtain a muscle nerve resting behavior baseline feature index set;

2) Extracting the corresponding characteristics of the physiological signals of the muscle system from the real-time characteristics of the physiological states to generate the real-time characteristics of the physiological states of the muscles;

3) Calculating a baseline variation value of a baseline characteristic index value in the real-time characteristic of the muscle physiological state and the baseline characteristic index value in the baseline characteristic index set of the muscle nerve rest behavior, namely analyzing the baseline variation to obtain a relative variation index set of the muscle tension characterization characteristic;

Thirdly, analyzing the behavior action level of the behavior state real-time characteristics in real time, and extracting a behavior action level real-time index.

In this embodiment, the calculation and generation method of the behavior action level real-time index includes:

1) Acquiring real-time characteristics of behavior states, analyzing and quantifying time distribution, duration, motion amplitude, motion frequency, motion intensity and motion regularity of behavior motions, and generating a behavior action level representation index set;

2) And carrying out weighted fusion calculation on all indexes in the behavior action level representation index set to obtain the current behavior action level real-time index.

In this embodiment, the specific calculation modes of the baseline variation analysis and the baseline variation value are as follows:

for real-valued variables

And its non-zero base line sequence +.>

For the baseline variation value of

；

wherein ,

respectively real value variable +.>

In this embodiment, the myotonic level real-time index reflects the activity or tension of the peripheral nerve and the muscular system, the central motor ability real-time index reflects the movement preparation and reserve conditions of the central nerve and the autonomic nerve, and the behavioural action level real-time index reflects the actual conditions of the quantity, intensity, frequency and the like of the motor action, so that the three conditions can well describe the behavioural level or the state of the behavioural ability of the user in different sleeping time phases.

P300: and carrying out real-time foundation line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index to obtain a sleep behavior activity level real-time index and generate a sleep behavior activity level real-time curve, and predicting and calculating to generate a sleep behavior activity level trend curve.

The method comprises the steps of firstly, carrying out real-time foundation line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index to obtain a sleep behavior activity level real-time index, and generating or updating a sleep behavior activity level real-time curve.

In this embodiment, the method for generating the sleep behavior activity level real-time index and the sleep behavior activity level real-time curve specifically includes:

1) Acquiring standard behavior active curve characteristic baseline index sets of healthy user groups with different sexes, different age groups and different scale numbers in a rest state of a waking period and a motion task state of the waking period;

2) Acquiring a current myotonic level real-time index, a central movement capability real-time index and a behavior movement level real-time index, calculating a rest baseline value and a baseline variation value of a task baseline value in a standard behavior activity curve characteristic baseline index set of healthy people of the same age group, namely, analyzing the baseline variation to obtain a sleep behavior activity level characteristic variation set;

3) Carrying out extremum harmonic analysis on all indexes in the characteristic change quantity set of the active level of the sleep behavior to obtain extremum harmonic values, namely the current real-time index of the active level of the sleep behavior;

4) And obtaining the sleep behavior activity level real-time index of the whole process according to the time sequence, and generating or updating to obtain a sleep behavior activity level real-time curve.

And secondly, carrying out real-time trend analysis and prediction calculation according to the sleep behavior activity level real-time curve to generate or update the sleep behavior activity level trend curve.

In this embodiment, the method for calculating and generating the sleep activity level trend curve is as follows:

1) Acquiring a current sleep behavior activity level real-time index and a current sleep behavior activity level real-time curve of a user;

3) And incorporating the sleep behavior activity level real-time prediction index according to the time sequence, and generating or updating a sleep behavior activity level trend curve.

In the actual adaptation scene, trend analysis and index prediction may adopt a time-series prediction method commonly used in AR, MR, ARMA, ARIMA, SARIMA, VAR and the like, and prediction calculation of the sleep activity level real-time prediction index can also be completed through a deep learning model.

P400: and generating a sleep behavior activity level dynamic regulation strategy in real time according to the sleep behavior level optimization knowledge base, the sleep time phase curve, the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve, and dynamically regulating the behavior activity level of the sleep process of the user in real time.

The method comprises the steps of optimizing a knowledge base, a sleep time phase curve, a sleep activity level real-time curve and a sleep activity level trend curve according to sleep activity levels, and generating a sleep activity level dynamic regulation strategy by combining the sleep activity level dynamic regulation purpose.

In this embodiment, the dynamic regulation strategy of the sleep behavior activity level at least includes a regulation mode, an execution part, a regulation method and a regulation intensity; the adjusting mode at least comprises sound, light, taste, electricity, magnetism, ultrasound and sleeping environment, the executing part comprises a head, a neck, a trunk part, left and right upper limbs, left and right lower limbs and various large sense organs, the adjusting method at least comprises a constant, an increasing curve, a decreasing curve, an index curve, a sine curve, a periodic square wave and a random curve, and the adjusting intensity is determined by the current real-time index of the sleeping activity level and the current real-time prediction index of the sleeping activity level.

In this embodiment, the sleep behavior level optimization knowledge base includes not only the professional knowledge, technical means, operation parameters, safety guidance, and other information of sleep behavior level adjustment, but also historical information of sleep behavior activity level dynamic adjustment of the user, namely, historical sleep time phase curves, sleep behavior activity level real-time curves, sleep behavior activity level trend curves, sleep behavior activity level dynamic adjustment strategies, sleep behavior activity level dynamic adjustment strategy effects, and the like.

In the actual adaptation scene, a proper adjusting mode, an execution part combination, an adjusting method and an adjusting intensity range can be selected according to different user scene requirements.

And secondly, dynamically adjusting the activity level of the sleeping process of the user in real time according to the activity level dynamic adjustment strategy of the sleeping.

In this embodiment, according to the sleep activity level dynamic adjustment policy, corresponding hardware devices are connected, and adjustment parameters are sent, so that the real-time dynamic adjustment of the activity level of the user in the sleep process is realized, and the personal safety and other unexpected factors in the adjustment process are monitored.

P500: repeating the steps to complete the circulation dynamic regulation of all the sleep behavior activity levels, evaluating the dynamic regulation effect, extracting the phase behavior activity correlation coefficient and the behavior level dynamic regulation effect coefficient, generating a sleep behavior activity level regulation report and establishing a personalized behavior regulation long-term database.

And the first step is to complete the circulation dynamic regulation of all the sleep behavior activity levels, and obtain a sleep time phase curve, a sleep behavior activity level real-time curve and a sleep behavior activity level trend curve of all the regulation processes.

In the whole sleeping process of the user, the central nervous state, the autonomic nervous state and the behavior movement state of the user are continuously collected and analyzed, the sleeping behavior activity level of the user is evaluated and quantified in real time, and the dynamic regulation strategy is further formulated or optimized according to the dynamic regulation purpose of the sleeping behavior activity level and the last regulation result effect, so that the continuous dynamic training and regulation of the sleeping behavior activity level of the user are realized.

And secondly, analyzing and calculating the relation characteristics of the sleep time phase curve and the sleep behavior activity level real-time curve, and extracting the time phase activity correlation coefficient.

In this embodiment, the method for calculating the phase behavior activity correlation coefficient specifically includes:

1) Acquiring a sleep time phase curve and a sleep behavior activity level real-time curve;

2) Analyzing and calculating the relation characteristic of the sleep time phase curve and the sleep behavior activity level real-time curve to obtain a time phase behavior activity level relation characteristic index set;

3) And carrying out weighted fusion calculation on the phase behavior activity level relation characteristic index set to obtain a phase behavior activity correlation coefficient.

In this embodiment, the relationship features include at least an association feature and a distance feature; the correlation features at least comprise a coherence coefficient, a pearson correlation coefficient, a Jacquard similarity coefficient, a linear mutual information coefficient and a linear correlation coefficient, and the distance features at least comprise an Euclidean distance, a Manhattan distance, a Chebyshev distance, a Minkowski distance, a normalized Euclidean distance, a Mahalanobis distance, a Papanic distance, a Hamming distance and an included angle cosine.

In this embodiment, the pearson correlation coefficient and the euclidean distance are selected as the relational features. For two arrays of the same length

and />

Pirson correlation coefficient->

The calculation formula of (2) is as follows:

；

wherein ,

for array->

Average value of>

For array->

Average value of (2).

Euclidean distance

The calculation formula of (2) is as follows: />

。

Thirdly, analyzing and calculating relation characteristics of the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve, and extracting a behavior level dynamic adjustment effect coefficient.

In this embodiment, the calculation method of the dynamic adjustment effect coefficient of the behavior level specifically includes:

1) Acquiring a sleep behavior activity level real-time curve and a sleep behavior activity level trend curve;

2) Analyzing and calculating relation characteristics of a sleep behavior activity level real-time curve and a sleep behavior activity level trend curve to obtain a behavior level dynamic regulation effect characteristic index set;

3) And carrying out weighted fusion calculation on the characteristic index set of the dynamic behavior level adjusting effect to obtain the dynamic behavior level adjusting effect coefficient.

In this embodiment, the relationship features include at least an association feature and a distance feature.

And fourthly, analyzing, calculating and generating a sleep behavior activity level regulation report according to the sleep time phase curve, the sleep behavior activity level real-time curve, the sleep behavior activity level trend curve, the time phase activity correlation coefficient and the behavior level dynamic regulation effect coefficient.

In this embodiment, the sleep behavior activity level adjustment report at least includes a sleep phase curve, a sleep behavior activity level real-time curve, a sleep behavior activity level trend curve, a phase behavior activity correlation coefficient, a behavior level dynamic adjustment effect coefficient, a total sleep behavior activity level dynamic adjustment policy, a behavior activity level phase distribution statistic, a peak activity period summary, a low peak activity period summary, an abnormal activity period summary, and a sleep behavior activity level adjustment report summary.

In this embodiment, the phase distribution statistics of the behavioral activity level are specifically the average behavioral activity level, the maximum behavioral activity level, and the minimum behavioral activity level of different sleep phases; the peak activity time summary is specifically peak time distribution, time numerical sum and duty ratio of peak time distribution corresponding to a segment exceeding a preset peak threshold in the sleep behavior activity level real-time curve; the low peak activity period summary is specifically a low peak period distribution corresponding to a segment exceeding a preset low peak threshold value in a sleep behavior activity level real-time curve, and a time numerical sum and a duty ratio of the low peak period distribution; the abnormal activity period summary is specifically an abnormal period distribution corresponding to an abnormal segment which deviates from the curve baseline trend in the sleep activity level real-time curve, a time numerical sum of the abnormal period distribution and a duty ratio.

And fifthly, establishing or updating a personalized behavior adjustment long-term database according to the sleep behavior activity level adjustment report and the current state information of the user, and providing a data analysis inheritance model for the continuous sleep behavior activity level dynamic adjustment of the subsequent user.

After the whole cycle dynamic adjustment is finished, the current age, physical and psychological state, sleeping environment and other information of the user and the sleeping behavior activity level adjustment report are combined, a personalized behavior adjustment long-term database is built and continuously updated, so that the subsequent user individual sleeping behavior activity level dynamic adjustment strategy is continuously optimized and adjusted, a quantized-adjusted long-term influence model is built, complete individuation and intellectualization are realized, and a better dynamic adjustment effect is achieved.

The database at least comprises phase activity related coefficients and dynamic regulation effect coefficients of the behavior level of the individual, and the two coefficients are reserved in the database due to different behavior activity degrees and regulation influence factors of different individuals, so that dynamic regulation can be completed more quickly and pertinently.

As shown in fig. 2, a system for dynamic adjustment of sleep activity level is provided according to an embodiment of the present invention, and is configured to perform the above-described method steps. The system comprises the following modules:

The time phase state analysis module S100 is used for carrying out real-time acquisition monitoring, signal processing and time frame characteristic analysis on physiological state signals and behavior state signals of a sleeping process of a user, generating physiological state real-time characteristics and behavior state real-time characteristics, identifying sleeping time phase states in real time and generating a sleeping time phase curve;

the characterization index extraction module S200 is configured to perform behavioral energy quantization analysis, baseline variation analysis and extremum harmonic analysis on the physiological state real-time feature and the behavioral state real-time feature in real time, so as to generate a myotonia level real-time index, a central exercise capacity real-time index and a behavioral action level real-time index;

the behavior level quantification module S300 is used for carrying out real-time foundation line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index to obtain a sleep behavior activity level real-time index and generate a sleep behavior activity level real-time curve, and predicting and calculating to generate a sleep behavior activity level trend curve;

the dynamic strategy adjustment module S400 is used for optimizing a knowledge base, a sleep time phase curve, a sleep activity level real-time curve and a sleep activity level trend curve according to the sleep activity level, generating a sleep activity level dynamic adjustment strategy in real time, and dynamically adjusting the activity level of the user in real time in the sleep process;

The behavior adjustment report module S500 is used for completing the cyclic dynamic adjustment of all sleep behavior activity levels, evaluating the dynamic adjustment effect, extracting the phase behavior activity correlation coefficient and the behavior level dynamic adjustment effect coefficient, generating a sleep behavior activity level adjustment report and establishing a personalized behavior adjustment long-term database;

the data management center module S600 is used for visual display and data operation management of all process data in the system.

In this embodiment, the phase state analysis module S100 further includes the following functional units:

the signal real-time monitoring unit is used for collecting and monitoring the physiological state and the behavior state of the sleeping process of the user in real time and generating a physiological state real-time signal and a behavior state real-time signal;

the feature real-time analysis unit is used for carrying out real-time feature analysis on the physiological state real-time data and the behavior state real-time data to generate physiological state real-time features and behavior state real-time features;

the time phase state identification unit is used for identifying the sleep time phase real-time state according to the physiological state real-time characteristic and the behavior state real-time characteristic, and obtaining a sleep time phase curve.

In this embodiment, the characterization index extraction module S200 further includes the following functional units:

the central movement capacity analysis unit is used for carrying out central movement capacity analysis, baseline change analysis and extremum harmonic analysis on the central nerve physiological characteristics in real time in the physiological state real-time characteristics, and extracting a central movement capacity real-time index;

the myotonic level analysis unit is used for carrying out real-time myotonic level analysis, baseline change analysis and extremum harmonic analysis on the physiological characteristics of the muscle system in the physiological state real-time characteristics, and extracting a myotonic level real-time index;

the behavior operation analysis unit is used for analyzing the behavior state real-time characteristics in real time and extracting the behavior action level real-time index.

In this embodiment, the behavior level quantization module S300 further includes the following functional units:

the behavior level quantification unit is used for carrying out real-time foundation line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index to obtain a sleep behavior activity level real-time index and generating or updating a sleep behavior activity level real-time curve;

and the exponential trend prediction unit is used for carrying out real-time trend analysis and prediction calculation according to the sleep behavior activity level real-time curve to generate or update the sleep behavior activity level trend curve.

In this embodiment, the dynamic policy adjustment module S400 further includes the following functional units:

the dynamic strategy generation unit is used for optimizing a knowledge base, a sleep time phase curve, a sleep activity level real-time curve and a sleep activity level trend curve according to the sleep activity level, and generating a sleep activity level dynamic regulation strategy by combining the sleep activity level dynamic regulation purpose;

In this embodiment, the behavior adjustment reporting module S500 further includes the following functional units:

the circulation dynamic adjusting unit is used for completing circulation dynamic adjustment of all sleep behavior active levels to obtain a sleep time phase curve, a sleep behavior active level real-time curve and a sleep behavior active level trend curve in all adjustment processes;

The regulation report generation unit is used for generating a sleep behavior activity level regulation report according to the sleep time phase curve, the sleep behavior activity level real-time curve, the sleep behavior activity level trend curve, the time phase activity correlation coefficient and the behavior level dynamic regulation effect coefficient through analysis and calculation;

and the behavior adjustment inheritance unit is used for establishing or updating a personalized behavior adjustment long-term database according to the sleep behavior activity level adjustment report and the current state information of the user, and providing a data analysis inheritance model for the continuous sleep behavior activity level dynamic adjustment of the subsequent user.

In this embodiment, the data management center module S600 further includes the following specific functional units:

As shown in fig. 3, an apparatus for dynamically adjusting sleep activity level according to an embodiment of the present invention includes the following modules:

The time phase state analysis module M100 is used for carrying out real-time acquisition monitoring, signal processing and time frame characteristic analysis on physiological state signals and behavior state signals of a sleeping process of a user, generating physiological state real-time characteristics and behavior state real-time characteristics, identifying sleeping time phase states in real time and generating a sleeping time phase curve;

the characterization index extraction module M200 is used for performing real-time behavioral energy quantitative analysis, baseline change analysis and extremum harmonic analysis on the physiological state real-time characteristics and the behavioral state real-time characteristics to generate a myotonia level real-time index, a central movement capacity real-time index and a behavioral movement level real-time index;

the behavior level quantification module M300 is used for carrying out real-time foundation line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central movement capability real-time index and the behavior movement level real-time index to obtain a sleep behavior activity level real-time index and generate a sleep behavior activity level real-time curve, and predicting and calculating to generate a sleep behavior activity level trend curve;

the dynamic strategy adjustment module M400 is used for optimizing a knowledge base, a sleep time phase curve, a sleep activity level real-time curve and a sleep activity level trend curve according to the sleep activity level, generating a sleep activity level dynamic adjustment strategy in real time, and dynamically adjusting the activity level of the user in real time in the sleep process;

The behavior adjustment report module M500 is used for completing the circulation dynamic adjustment of all sleep behavior activity levels, evaluating the dynamic adjustment effect, extracting the phase behavior activity correlation coefficient and the behavior level dynamic adjustment effect coefficient, generating a sleep behavior activity level adjustment report and establishing a personalized behavior adjustment long-term database;

the data visualization pipe module M600 is used for performing visualization display management on all data in the device;

the data operation management module M700 is used for storing, backing up, migrating and exporting all data in the device.

The apparatus is configured to correspondingly perform the steps of the method of fig. 1, and will not be described in detail herein.

The present invention also provides various types of programmable processors (FPGA, ASIC or other integrated circuit) for running a program, wherein the program when run performs the steps of the embodiments described above.

The invention also provides corresponding computer equipment, comprising a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the memory realizes the steps in the embodiment when the program is executed.

Although the embodiments of the present invention are described above, the embodiments are only used for facilitating understanding of the present invention, and are not intended to limit the present invention. Any person skilled in the art to which the present invention pertains may make any modifications, changes, equivalents, etc. in form and detail of the implementation without departing from the spirit and principles of the present invention disclosed herein, which are within the scope of the present invention. Accordingly, the scope of the invention should be determined from the following claims.

Claims

1. A method for dynamic regulation of sleep activity levels, comprising the steps of:

Repeating the steps to complete the circulation dynamic adjustment of all the sleep behavior activity levels, evaluating the dynamic adjustment effect, extracting the phase behavior activity correlation coefficient and the behavior level dynamic adjustment effect coefficient, generating a sleep behavior activity level adjustment report and establishing a personalized behavior adjustment long-term database;

the specific steps of performing real-time behavioral energy quantitative analysis, baseline variation analysis and extremum harmonic analysis on the physiological state real-time feature and the behavioral state real-time feature to generate a myotonia level real-time index, a central exercise capacity real-time index and a behavioral action level real-time index further comprise:

performing real-time behavior action level analysis on the behavior state real-time characteristics, and extracting the behavior action level real-time index;

The extremum harmonic analysis is a data analysis method which takes at least one of the maximum value, the minimum value, the maximum value of the absolute value and the minimum value of the absolute value of the numerical value array as an observation base point basis, and takes at least one of the mean value, the median, the quantile, the variance, the variation coefficient, the kurtosis, the skewness, the absolute value mean value, the absolute value median, the absolute value quantile, the absolute value variance, the absolute value variation coefficient, the absolute value kurtosis and the absolute value skewness of the numerical value array as a main analysis harmonic item to observe and analyze the extremum fluctuation state and the general trend change of the numerical value array;

the method for generating the sleep behavior activity level real-time index and the sleep behavior activity level real-time curve specifically comprises the following steps:

2. The method of claim 1, wherein the specific steps of collecting and monitoring physiological status signals and behavioral status signals of the sleep process of the user in real time, processing the signals and analyzing the characteristics to generate physiological status real-time characteristics and behavioral status real-time characteristics, identifying sleep phase status in real time and generating a sleep phase curve further comprise:

the physiological state and the behavior state of the sleeping process of the user are collected and monitored in real time, and a physiological state real-time signal and a behavior state real-time signal are generated;

3. A method according to claim 1 or 2, characterized in that: the physiological status signals include at least one of central nervous physiological signals, autonomic nervous physiological signals, and musculature physiological signals; wherein the central nervous physiological signal comprises at least one of an electroencephalogram signal, a magnetoencephalography signal and an oxygen level dependent signal, the autonomic nervous physiological signal comprises at least one of an electrocardio signal, a pulse signal, a respiration signal, an oxygen level signal, a body temperature signal and a skin electrical signal, and the muscle system physiological signal comprises at least one of an oxygen level dependent signal, an electromyographic signal, a skin electrical signal and an acceleration signal.

4. A method as claimed in claim 3, wherein: the behavioral state signals include at least one of sleep posture position signals and limb movement signals.

5. The method of claim 1, wherein: the signal processing at least comprises A/D digital-to-analog conversion, resampling, re-referencing, noise reduction, artifact removal, power frequency notch, low-pass filtering, high-pass filtering, band-stop filtering, band-pass filtering, correction processing and time frame division; the correction processing specifically comprises signal correction and prediction smoothing processing of signal data fragments containing artifacts or distortion in signals, and the time frame division refers to moving interception processing of target signals according to a preset time window and a preset time step.

6. The method of claim 2, wherein: the feature analysis comprises at least one of numerical feature, physical feature analysis, time-frequency feature analysis, envelope feature and nonlinear feature analysis; wherein the numerical features include at least one of mean, root mean square, maximum, minimum, variance, standard deviation, coefficient of variation, kurtosis, and skewness; the physical characteristics comprise at least one of time information, duration information, amplitude characteristics, intensity characteristics and frequency characteristics; the time-frequency characteristic comprises at least one of total power, characteristic frequency band power duty ratio, characteristic frequency band central frequency, heart rate and heart rate variability; the envelope features comprise at least one of an original signal, an envelope signal, a normalized envelope signal, an envelope mean, an envelope root mean square, an envelope maximum, an envelope minimum, an envelope variance, an envelope standard deviation, an envelope variation coefficient, an envelope kurtosis and an envelope skewness; the nonlinear features include at least one of entropy features, fractal features, and complexity features.

7. The method of claim 6, wherein: the physiological state real-time characteristic comprises at least one of the numerical characteristic, the time-frequency characteristic, the envelope characteristic, and the nonlinear characteristic of the physiological state signal.

8. The method of claim 6 or 7, wherein: the behavioral state real-time characteristics include at least the numerical characteristics, the physical characteristics, and the time-frequency characteristics of the behavioral state signals.

9. A method according to claim 1 or 2, characterized in that: the extraction method of the sleep phase curve specifically comprises the following steps:

10. The method of claim 1, wherein one specific calculation of the extremum blending analysis is:

for the numerical array Y

For example, the extremum harmonic value is calculated by

；

wherein ,

is the extremum harmonic value of the value array Y, < ->

11. The method of claim 1, wherein the calculation and generation method of the real-time index of central movement capability is as follows:

1) Collecting central nerve physiological real-time signals and autonomic nerve physiological real-time signals in a resting state when a current user wakes up, and carrying out feature analysis and feature value mean value calculation to obtain a nerve resting motor capacity baseline feature index set;

12. The method of claim 1, wherein the calculation and generation method of the real-time index of the myopic level is as follows:

1) Collecting physiological signals of the muscle system in a resting state when the current user wakes up, and carrying out feature analysis and feature value mean value calculation to obtain a muscle resting behavior baseline feature index set;

3) Calculating a baseline variation value of a characteristic value in the real-time characteristic of the muscle physiological state and a baseline characteristic index value in the baseline characteristic index set of the muscle rest behavior, namely baseline variation analysis, and obtaining a relative variation index set of the muscle tension characterization characteristic;

13. The method of claim 1, wherein the calculation and generation method of the behavioral activity level real-time index is as follows:

14. The method of claim 11 or 12, wherein: the specific calculation modes of the baseline variation analysis and the baseline variation value are as follows:

for real-valued variables

And its non-zero base line sequence +.>

For the baseline variation value of

；

wherein ,

respectively real value variable +.>

15. The method according to claim 1 or 2, wherein the specific steps of performing real-time ground line change analysis and extremum harmonic analysis according to the myotonic level real-time index, the central motor ability real-time index and the behavioral activity level real-time index to obtain a sleep activity level real-time index and generate a sleep activity level real-time curve, and predicting and calculating to generate a sleep activity level trend curve further include:

16. The method of claim 15, wherein the method of computing the sleep activity level trend curve is as follows:

17. The method according to claim 1 or 2, wherein the specific steps of optimizing a knowledge base according to sleep behavior level, the sleep phase curve, the sleep behavior activity level real-time curve and the sleep behavior activity level trend curve, generating a sleep behavior activity level dynamic adjustment strategy in real time, and dynamically adjusting the behavior activity level of a sleep process of a user in real time further comprise:

18. The method of claim 17, wherein the sleep activity level dynamic adjustment strategy comprises at least one of an adjustment mode, an execution site, an adjustment method, and an adjustment intensity; the adjusting mode comprises at least one of sound, light, smell, electricity, magnetism, ultrasound and sleeping environment, the executing part comprises a head, a neck, a trunk, left and right upper limbs, left and right lower limbs and various large sensory organs, the adjusting method comprises at least one of a constant, an increasing curve, a decreasing curve, an exponential curve, a sinusoidal curve, a periodic square wave and a random curve, and the adjusting intensity is determined by the current real-time index of the sleeping activity level and the current real-time prediction index of the sleeping activity level.

19. A method according to claim 1 or 2, characterized in that: the steps are repeated, the circulation dynamic adjustment of all the sleep behavior activity levels is completed, the dynamic adjustment effect is evaluated, the phase behavior activity correlation coefficient and the behavior level dynamic adjustment effect coefficient are extracted, and the specific steps of generating the sleep behavior activity level adjustment report and establishing the personalized behavior adjustment long-term database further comprise:

20. The method of claim 19, wherein: the calculation method of the phase behavior activity correlation coefficient specifically comprises the following steps:

21. The method as recited in claim 20, wherein: the calculation method of the dynamic behavior level adjusting effect coefficient specifically comprises the following steps:

22. The method of claim 20 or 21, wherein: the relationship features include at least one of an association feature and a distance feature; wherein the correlation characteristic comprises at least one of a coherence coefficient, a pearson correlation coefficient, a jaccard similarity coefficient, a linear mutual information coefficient, and a linear correlation coefficient, and the distance characteristic comprises at least one of a euclidean distance, a manhattan distance, a chebyshev distance, a minkowski distance, a normalized euclidean distance, a mahalanobis distance, a barbita distance, a hamming distance, and an angle cosine.

23. The method of claim 21, wherein: the sleep activity level adjustment report comprises at least one of a sleep time phase curve, a sleep activity level real-time curve, a sleep activity level trend curve, a time phase activity correlation coefficient, a behavior level dynamic adjustment effect coefficient, all sleep activity level dynamic adjustment strategies, a behavior activity level time phase distribution statistic, a peak activity period summary, a low peak activity period summary, an abnormal activity period summary and a sleep activity level adjustment report summary.

24. The method of claim 23, wherein: the behavior activity level time phase distribution statistics specifically comprise average behavior activity level, maximum behavior activity level and minimum behavior activity level of different sleep time phases; the peak activity time summary is specifically peak time distribution corresponding to a segment exceeding a preset peak threshold value in the sleep behavior activity level real-time curve, and time numerical sum and duty ratio of the peak time distribution; the low peak activity period summary is specifically low peak period distribution corresponding to a segment exceeding a preset low peak threshold value in the sleep behavior activity level real-time curve, and time numerical sum and duty ratio of the low peak period distribution; the abnormal activity period summary is specifically an abnormal period distribution corresponding to an abnormal segment which is separated from a curve baseline trend in the sleep behavior activity level real-time curve, a time value sum and a duty ratio of the abnormal period distribution.

25. A system for dynamic adjustment of sleep activity levels, comprising the following modules:

The data management center module is used for visual display and data operation management of all process data in the system;

26. The system of claim 25, wherein the phase state analysis module further comprises the following functional units:

27. The system of claim 25 or 26, wherein the characterization index extraction module further comprises the following functional units:

28. The system of claim 27, wherein the behavioral level quantization module further comprises the following functional units:

29. The system of claim 25 or 26, wherein the dynamic policy adjustment module further comprises the following functional units:

30. The system of claim 29, wherein the behavior adjustment reporting module further comprises the following functional units:

31. The system of claim 25, wherein the data management center module further comprises the following specific functional units:

32. An apparatus for dynamic adjustment of sleep activity levels, comprising:

the data operation management module is used for storing, backing up, migrating and exporting all data in the device;