Methods, systems, and devices for combined detection, evaluation, and auxiliary regulation of sleep brain and respiration

Abstract

This invention provides a method, system, and device for the joint detection, evaluation, and auxiliary regulation of sleep-related brain and respiratory functions. It collects, detects, and dynamically analyzes sleep-related respiratory behavior signals, extracts sleep-related respiratory dynamic signals, and identifies sleep-related respiratory events through event detection and analysis. Furthermore, it combines quantitative characteristics of sleep-related respiratory and brain states with sleep phase segmentation to extract event intensity and event level, achieving scientific quantification and joint evaluation of sleep-related breathing. By predicting the development trends of respiratory dynamics, respiratory state, and brain state, it generates sleep-related brain state auxiliary regulation strategies in real time and sends them to sleep regulation intervention devices via a signal interface, thereby achieving dynamic auxiliary regulation of the user's sleep-related brain state. This invention enables scientific detection, evaluation, and efficient dynamic auxiliary regulation of sleep-related breathing.

Description

Technical Field

[0001] This invention relates to the field of sleep breathing detection, evaluation, and assisted regulation, and particularly to methods, systems, and devices for combined sleep brain and breathing detection, evaluation, and assisted regulation. Background Technology

[0002] Normally, respiratory behavior alters blood circulation, directly affecting the functioning of the brain and other central nervous systems, especially during sleep, a state of drastic decline or loss of self-awareness. Sleep breathing in healthy individuals exhibits cyclical, trend-like, and stable patterns; memories and dreams during sleep also influence sleep breathing states and patterns. Overall, sleep breathing and sleep brain states maintain physiological stability, continuity, and crucial synergy across different sleep phases and states.

[0003] Currently, existing technologies incorporate the coupling relationship between brain state and respiratory behavior into detection, analysis, and assisted regulation. While sleep apnea detection and assessment technologies primarily focus on respiratory movement behavior itself, addressing respiratory event detection and classification, and using simple statistical analyses such as the hypoventilation index and respiratory event frequency commonly used in clinical practice, they fail to provide a clear definition or quantification of the sleep breathing process, the intensity and level of sleep breathing events. In particular, they lack the crucial element of detecting or evaluating the coupling influence between sleep brain state and sleep breathing events. Furthermore, current sleep regulation devices or sleep apnea regulation devices are typically separate from sleep detection devices or sleep apnea detection devices. Most regulation devices can connect to the network to provide feedback on regulation parameters but still rely on offline preset programs for control, failing to provide intelligent, precise, and personalized dynamic assisted regulation based on the user's real-time status.

[0004] As can be seen from the above, how to scientifically and comprehensively quantify and evaluate the processes and events of sleep breathing, sleep breathing and sleep brain state, and further optimize and improve the efficiency and effectiveness of existing sleep breathing brain state regulation devices to assist users in sleep breathing and improve sleep quality are problems that need to be further solved in current domestic and foreign product technology solutions and practical application scenarios. Summary of the Invention

[0005] To address the shortcomings and improvement needs of existing methods, this invention aims to provide a method for the joint detection, evaluation, and auxiliary regulation of sleep-related brain and respiration. This method involves collecting, detecting, and analyzing sleep-related respiratory behavior signals, extracting sleep-related respiratory dynamic signals, and identifying sleep-related respiratory events through event detection and analysis. Furthermore, it combines quantitative characteristics of sleep-related respiratory and brain states with sleep phase segmentation to extract event intensity and level, achieving scientific detection, quantification, and joint evaluation of sleep-related breathing. By predicting the development trends of respiratory dynamics, respiratory state, and brain state, a sleep-related brain state auxiliary regulation strategy is generated in real time and sent to a sleep regulation intervention device via a signal interface to achieve dynamic auxiliary regulation of the user's sleep-related brain state. This invention also provides a system for the joint detection, evaluation, and auxiliary regulation of sleep-related brain and respiration to implement the above method. Finally, this invention provides a device for the joint detection, evaluation, and auxiliary regulation of sleep-related brain and respiration to implement the above system.

[0006] According to the purpose of this invention, a method for joint detection, evaluation, and auxiliary regulation of sleep brain and respiration is proposed, comprising the following steps:

[0007] The system collects and processes users' sleep brain state signals and performs feature analysis to identify sleep phases and obtain quantitative features of sleep brain state.

[0008] The system collects, processes, and performs dynamic analysis on the user's sleep breathing behavior signals to obtain the first sleep breathing dynamic signal. Through event detection and analysis, it extracts the basic information of sleep breathing events.

[0009] The first sleep respiratory dynamics signal is decomposed and / or analyzed in time and frequency to obtain the second sleep respiratory dynamics signal, and feature analysis is performed on the two dynamics signals to generate quantitative features of sleep breathing state.

[0010] Based on the quantitative characteristics of sleep brain state, the quantitative characteristics of sleep breathing state, and the sleep phase segmentation, the relevant characteristic changes of the sleep breathing events are compared and analyzed to identify the event intensity and event level that generated the sleep breathing events;

[0011] Trend prediction is performed on the first sleep respiratory dynamics signal, the quantitative characteristics of sleep respiratory state and the quantitative characteristics of sleep brain state. Combined with the sleep respiratory knowledge base and the user sleep respiratory database, a sleep respiratory brain state auxiliary regulation strategy is generated and sent to the sleep regulation intervention device through the signal interface.

[0012] According to a preset cycle or strategy, a joint detection, evaluation and auxiliary adjustment report of sleep breathing brain state is generated and output, and the user's sleep breathing database is updated.

[0013] More preferably, the specific steps of collecting and processing the user's sleep brain state signals and performing feature analysis to identify sleep phase stages and obtain quantitative features of sleep brain state further include:

[0014] The brain state signals during the user's sleep process are collected and processed to obtain the sleep brain state signals;

[0015] Feature analysis is performed on the sleep brain state signal to obtain the quantitative features of the sleep brain state;

[0016] Based on the sleep brain state signals, the sleep phase stages are identified, and a sleep phase curve is generated.

[0017] More preferably, the sleep brain state signal includes at least one of electroencephalogram (EEG) signals, magnetic resonance imaging (MRI) signals, and functional near-infrared imaging (FIR) signals.

[0018] More preferably, the acquisition and processing includes at least acquisition, analog-to-digital conversion, resampling, rereference, artifact removal, signal correction, noise reduction, power frequency notch filtering, bandpass filtering, mean filtering, smoothing, and signal time window segmentation; wherein, the signal time window segmentation specifically involves continuously segmenting the target signal into time windows with a preset length time window and a preset translation time step to obtain a multi-time window signal set.

[0019] More preferably, the feature analysis includes at least numerical feature analysis, envelope feature analysis, time-frequency feature analysis, and nonlinear feature analysis; wherein, the numerical features include at least the mean, root mean square, maximum, minimum, variance, standard deviation, coefficient of variation, kurtosis, and skewness; the time-frequency features include at least the band power, band power percentage, and band center frequency; and the nonlinear features include at least the entropy feature, fractal feature, and complexity feature.

[0020] More preferably, the sleep phase segmentation includes at least the wakefulness stage, light sleep stage, deep sleep stage, and REM sleep stage; the method for generating the sleep phase segmentation and the sleep phase curve is as follows:

[0021] 1) The sleep brain state signals and their corresponding sleep stage data of a large-scale sleep user sample are learned and trained and modeled by machine learning to obtain a sleep phase stage model;

[0022] 2) Input the current user's sleep brain state signal into the sleep phase segmentation model to obtain the corresponding sleep phase segmentation;

[0023] 3) Extract the values ​​of the sleep phase stages of all signal time windows according to the time sequence to obtain the sleep phase curve.

[0024] More preferably, the specific steps of acquiring, processing, and analyzing the user's sleep breathing behavior signals to obtain a first sleep breathing dynamic signal and extracting sleep breathing events through event detection and analysis further include:

[0025] The breathing behavior signals during the user's sleep process are collected and processed to obtain the sleep breathing behavior signals;

[0026] The sleep breathing behavior signal is subjected to dynamic analysis to obtain the first sleep breathing dynamic signal;

[0027] The first sleep respiratory dynamics signal is subjected to event detection analysis to identify the sleep breathing events and extract basic event information.

[0028] More preferably, the sleep breathing behavior signal includes at least sleep breathing movement signal and sleep position signal; the sleep breathing movement signal includes at least one of oral and nasal temperature monitoring signal, nasal pressure monitoring signal, oral and nasal CO2 monitoring signal, chest and abdominal respiratory movement signal, electrocardiogram-derived respiratory signal, and pharyngeal electromyography signal; and the sleep position signal includes at least sleeping posture direction signal and sleeping posture angle signal.

[0029] More preferably, the first sleep respiratory dynamics signal is specifically a dynamic curve describing the continuous strength changes of sleep breathing, including at least one of the following: oral and nasal temperature dynamics signal, nasal pressure dynamics signal, oral and nasal CO2 dynamics signal, chest and abdominal respiratory movement dynamics signal, electrocardiogram-derived respiratory dynamics signal, pharyngeal electromyography dynamics signal, as well as a sleep posture direction description signal and a sleep posture angle description signal.

[0030] More preferably, the dynamic analysis specifically involves analyzing the dynamic properties, signal quality, and morphology of each physiological signal in the sleep breathing behavior signal, selecting target signals, and performing signal fusion to obtain a sleep breathing dynamic description signal.

[0031] More preferably, the sleep breathing event includes at least the event type, start time, end time, duration, peak and trough values, time between peak and trough values, event intensity, and event level.

[0032] More preferably, the event detection and analysis specifically involves identifying and extracting waveform features and labels from the sleep breathing dynamics signal based on a preset sleep breathing event knowledge base and / or machine learning model to obtain basic event information of the sleep breathing event; the basic event information includes at least the event type, start time, end time, duration, peak and trough values, and time at the peak and trough values.

[0033] More preferably, the specific steps of performing signal decomposition and / or time-frequency analysis on the first sleep respiratory dynamics signal to obtain a second sleep respiratory dynamics signal, and performing feature analysis on the two dynamics signals to generate quantitative features of sleep breathing state, further include:

[0034] The first sleep respiratory dynamics signal is decomposed and / or analyzed in time and frequency to obtain the second sleep respiratory dynamics signal.

[0035] Feature analysis is performed on the first sleep respiratory dynamics signal and the second sleep respiratory dynamics signal to obtain the quantitative features of the sleep breathing state.

[0036] More preferably, the second sleep respiratory dynamics signal includes at least a sleep respiratory dynamics cycle signal, a sleep respiratory dynamics trend signal, and a sleep respiratory dynamics oscillation signal.

[0037] More preferably, the signal decomposition or the time-frequency analysis specifically involves extracting the periodic component, trend component, residual or oscillatory component of the sleep respiratory dynamics signal from the first sleep respiratory dynamics signal to generate the second sleep respiratory dynamics signal, and further extracting the periodic feature, trend feature and stationary feature of the sleep respiratory dynamics signal through feature analysis.

[0038] More preferably, the signal decomposition method includes at least time series decomposition, empirical mode decomposition, variational mode decomposition, local mean decomposition, wavelet transform, wavelet packet transform, time-frequency transform, blind source separation, linear discriminant analysis, detrending analysis, principal component analysis, independent component analysis, waveform analysis, and numerical fitting, as well as their evolutionary variants; the time-frequency analysis method includes at least any one of time-frequency transform, time-domain filtering, and frequency-domain filtering.

[0039] More preferably, the quantitative characteristics of sleep breathing state include at least sleep respiratory dynamics characteristics, sleep respiratory dynamics periodicity characteristics, sleep respiratory dynamics trend characteristics, sleep respiratory dynamics stability characteristics, and sleep posture characteristics; the sleep respiratory dynamics characteristics include at least respiratory rate, numerical characteristics, time-frequency characteristics, and nonlinear characteristics; and the sleep posture characteristics include at least sleeping position direction characteristics and sleeping position angle characteristics.

[0040] More preferably, the specific steps of comparing and analyzing the relevant feature changes of the sleep breathing event based on the quantitative features of the sleep brain state, the quantitative features of the sleep breathing state, and the sleep phase segmentation, and identifying the event intensity and event level that generated the sleep breathing event, further include:

[0041] Based on the occurrence period and comparison interval of the sleep breathing events, the relative changes of the quantitative characteristics of sleep brain state and the quantitative characteristics of sleep breathing state are compared and analyzed to obtain a set of event description feature changes.

[0042] The intensity of the event is identified and generated based on the sleep breathing events, the set of changes in the event description features, and the sleep phases.

[0043] The event level is identified and generated based on the sleep breathing events, sleep posture characteristics, and sleep phase stages.

[0044] More preferably, the method for generating the event intensity is as follows:

[0045] 1) Obtain the set of changes in the event description features, select key target features and perform numerical weighted calculation of feature values ​​to obtain the event feature change correction coefficient;

[0046] 2) Obtain the start time, end time, duration, peak and trough values, and time at the peak and trough values ​​of the sleep breathing events, as well as the preset sleep breathing dynamics signal threshold;

[0047] 3) Calculate the linear slope based on the onset time, peak and trough values, and time at the peak and trough values ​​of the sleep breathing events to obtain the slope of the peak and trough leading edge;

[0048] 4) Calculate the linear slope based on the end time, peak and trough values, and time at the peak and trough values ​​of the sleep breathing events to obtain the slope at the trailing edge of the peak and trough.

[0049] 5) Based on the current sleep phase segmentation, extract the sleep phase correction coefficient according to the preset sleep phase correction coefficient reference table;

[0050] 6) The event intensity is obtained by fusing the values ​​of the event characteristic change correction coefficient, the peak-valley front slope, the peak-valley rear slope, and the sleep phase correction coefficient.

[0051] More preferably, the method for generating the event level is as follows:

[0052] 1) Obtain the start time, end time, duration, peak and trough values, and time at the peak and trough values ​​of the sleep breathing events, as well as a preset sleep breathing dynamics signal threshold;

[0053] 2) Calculate the relative changes between the peak and trough values ​​of the sleep breathing events and the preset sleep breathing dynamics signal threshold to obtain the peak-trough relative values;

[0054] 3) Calculate the product of the duration of the sleep breathing event and the relative value of the peak and trough to obtain the event level description feature value;

[0055] 4) Based on the sleep position characteristics, extract the sleeping posture angle and sleeping posture direction information or preset the sleep position correction calculation strategy, and extract the sleep position correction coefficient according to the preset sleep position correction coefficient reference table.

[0056] 5) Based on the current sleep phase segmentation, extract the sleep phase correction coefficient according to the preset sleep phase correction coefficient reference table;

[0057] 6) The event level is obtained by multiplying the event level description feature value, the sleep position correction coefficient, and the sleep phase correction coefficient according to the prediction level classification rules.

[0058] More preferably, the specific steps of performing trend prediction on the first sleep respiratory dynamics signal, the quantitative features of sleep respiratory state, and the quantitative features of sleep brain state, combining the sleep respiratory knowledge base and the user's sleep respiratory database, generating a sleep respiratory brain state-assisted regulation strategy, and sending it to the sleep regulation intervention device through a signal interface further include:

[0059] Trend prediction is performed on the first sleep respiratory dynamics signal to obtain the first sleep respiratory dynamics prediction signal;

[0060] Trend prediction is performed on the quantitative characteristics of sleep breathing state to obtain predicted characteristics of sleep breathing state.

[0061] Trend prediction is performed on the quantitative characteristics of sleep brain state to obtain sleep brain state prediction characteristics;

[0062] The first sleep respiratory dynamics prediction signal, the sleep respiratory state prediction feature and the sleep brain state prediction feature are aggregated to obtain a sleep respiratory and brain state prediction description.

[0063] The first sleep respiratory dynamics signal, the quantitative features of sleep respiratory state, and the quantitative features of sleep brain state are aggregated to obtain the current description of sleep respiratory and brain state.

[0064] Based on the predicted description of sleep breathing and brain state and the current description of sleep breathing and brain state, combined with the sleep breathing knowledge base and the user's sleep breathing database, the sleep breathing brain state auxiliary regulation strategy is generated.

[0065] The sleep-breathing-brain state auxiliary regulation strategy is sent to the sleep regulation intervention device through the signal interface, and the control execution of the sleep regulation intervention device is optimized to achieve dynamic auxiliary regulation of the user's sleep-breathing-brain state.

[0066] More preferably, the trend prediction includes at least one of the following: exponential smoothing, Holt-Winters method, AR, MA, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA, VARMAX, and machine learning.

[0067] More preferably, the sleep breathing knowledge base mainly comes from knowledge and experience related to health management and clinical medicine related to sleep breathing, including at least sleep knowledge, sleep breathing patterns, characteristics of common sleep breathing events, commonly used sleep breathing regulation methods, and guidance on scenario intervention parameters; the user sleep breathing database includes at least the sleep breathing events, the first sleep breathing dynamics signal, the second sleep breathing dynamics signal, the quantitative characteristics of the sleep breathing state, the quantitative characteristics of the sleep brain state, and the sleep breathing brain state auxiliary regulation strategy.

[0068] More preferably, the sleep breathing brain state-assisted regulation strategy includes at least a target value for respiratory rate, a target value for respiratory depth, a regulation mode, a regulation time point, a duration, and device control parameters.

[0069] More preferably, the sleep regulation intervention device includes at least one of the following: a ventilator, a smart mattress, an odor stimulation device, an electrical stimulation device, a magnetic stimulation device, a tactile stimulation device, an environmental temperature and humidity control device, and an environmental CO2 concentration control device.

[0070] More preferably, the specific steps of generating and outputting a sleep breathing brain state joint detection evaluation and auxiliary adjustment report according to a preset cycle or strategy, and updating the user's sleep breathing database, further include:

[0071] According to the preset reporting cycle, generate or output the sleep breathing brain state joint detection evaluation and auxiliary regulation report;

[0072] The user's sleep breathing database is updated according to a preset data update strategy.

[0073] More preferably, the sleep breathing brain state joint detection evaluation and auxiliary regulation report includes at least the statistical analysis of the sleep breathing events, the sleep breathing state evaluation summary, the sleep brain state evaluation summary, the auxiliary regulation evaluation summary, and suggestions for improving sleep behavior.

[0074] According to the purpose of this invention, a sleep brain and breathing joint detection, evaluation, and auxiliary regulation system is proposed, comprising the following modules:

[0075] The brain state detection module is used to collect and process the user's sleep brain state signals and, through feature analysis, identify sleep phase stages to obtain quantitative features of sleep brain state.

[0076] The breathing event detection module is used to collect, process, and perform dynamic analysis on the user's sleep breathing behavior signals to obtain the first sleep breathing dynamic signal and extract the basic information of the sleep breathing events through event detection and analysis.

[0077] The breathing state quantification module is used to perform signal decomposition and / or time-frequency analysis on the first sleep breathing dynamics signal to obtain the second sleep breathing dynamics signal, and to perform feature analysis on the two dynamics signals to generate sleep breathing state quantification features.

[0078] The intensity level assessment module is used to compare and analyze the relevant feature changes of the sleep breathing event based on the quantitative features of the sleep brain state, the quantitative features of the sleep breathing state, and the sleep phase stage, and to identify the event intensity and event level that generated the sleep breathing event.

[0079] The sleep-assisted regulation module is used to predict the trends of the first sleep respiratory dynamics signal, the quantitative characteristics of sleep respiratory state and the quantitative characteristics of sleep brain state, and, in combination with the sleep respiratory knowledge base and the user's sleep respiratory database, generate a sleep respiratory brain state-assisted regulation strategy and send it to the sleep regulation intervention device through the signal interface.

[0080] The report data management module is used to generate and output a sleep breathing brain state joint detection evaluation and auxiliary adjustment report according to a preset cycle or strategy, and update the user's sleep breathing database.

[0081] The data operation management module is used to perform visual management, unified storage, and operation management of all process data of the system.

[0082] More preferably, the brain state detection module further includes the following functional units:

[0083] The brain signal detection unit is used to collect and process brain state signals during the user's sleep process to obtain the sleep brain state signal.

[0084] A brain state feature analysis unit is used to perform feature analysis on the sleep brain state signal to obtain the quantitative features of the sleep brain state.

[0085] The sleep phase analysis unit is used to identify the sleep phase stages based on the sleep brain state signals and generate a sleep phase curve.

[0086] More preferably, the respiratory event detection module further includes the following functional units:

[0087] A respiratory signal detection unit is used to collect and process respiratory behavior signals during the user's sleep process to obtain the sleep respiratory behavior signal.

[0088] A dynamic analysis unit is used to perform dynamic analysis on the sleep breathing behavior signal to obtain the first sleep breathing dynamic signal;

[0089] The event detection and analysis unit is used to perform event detection and analysis on the first sleep respiratory dynamics signal, identify the sleep breathing event, and extract the basic information of the event.

[0090] More preferably, the respiratory state quantification module further includes the following functional units:

[0091] The signal decomposition and extraction unit is used to perform signal decomposition and / or time-frequency analysis on the first sleep respiratory dynamics signal to obtain the second sleep respiratory dynamics signal.

[0092] The breathing feature analysis unit is used to perform feature analysis on the first sleep breathing dynamics signal and the second sleep breathing dynamics signal to obtain the quantitative features of the sleep breathing state.

[0093] More preferably, the strength level assessment module further includes the following functional units:

[0094] The feature change calculation unit is used to compare and analyze the relative changes of the quantitative features of sleep brain state and the quantitative features of sleep breathing state based on the event occurrence period and event comparison interval of the sleep breathing event, and obtain a set of event description feature changes.

[0095] An event intensity quantification unit is used to identify and generate the event intensity based on the sleep breathing event, the event description feature change set, and the sleep phase stage.

[0096] The event level quantification unit is used to identify and generate the event level based on the sleep breathing event, the sleep posture characteristics, and the sleep phase stage.

[0097] More preferably, the sleep assistance adjustment module further includes the following functional units:

[0098] A respiratory dynamics prediction unit is used to perform trend prediction on the first sleep respiratory dynamics signal to obtain a first sleep respiratory dynamics prediction signal.

[0099] A breathing state prediction unit is used to predict the trend of the quantitative characteristics of sleep breathing state to obtain sleep breathing state prediction characteristics.

[0100] A brain state prediction unit is used to predict the trend of the quantitative characteristics of sleep brain state to obtain sleep brain state prediction characteristics.

[0101] The prediction description aggregation unit is used to aggregate the first sleep respiratory dynamics prediction signal, the sleep respiratory state prediction feature and the sleep brain state prediction feature to obtain a sleep respiratory and brain state prediction description.

[0102] The baseline description aggregation unit is used to aggregate the first sleep respiratory dynamics signal, the sleep respiratory state quantification feature and the sleep brain state quantification feature to obtain the current description of sleep breathing and brain state.

[0103] The auxiliary strategy generation unit is used to generate the sleep breathing and brain state auxiliary regulation strategy based on the predicted description of sleep breathing and brain state and the current description of sleep breathing and brain state, combined with the sleep breathing knowledge base and the user's sleep breathing database.

[0104] The strategy-assisted adjustment unit is used to send the sleep-breathing-brain state auxiliary adjustment strategy to the sleep regulation intervention device through a signal interface, optimize the control execution of the sleep regulation intervention device, and realize the dynamic auxiliary adjustment of the user's sleep-breathing-brain state.

[0105] More preferably, the report data management module also includes the following functional units:

[0106] The report generation and output unit is used to generate or output the sleep breathing brain state joint detection evaluation and auxiliary adjustment report according to a preset reporting cycle;

[0107] The database update unit is used to update the user's sleep breathing database according to a preset data update strategy.

[0108] More preferably, the data operation management module further includes the following functional units:

[0109] The user information management unit is used for registering, inputting, editing, querying, outputting, and deleting basic user information;

[0110] The data visualization management unit is used for the visualization and management of all data in the system.

[0111] The data operation management unit is used for storing, backing up, migrating, and exporting all data in the system.

[0112] According to the purpose of this invention, a sleep brain and breathing joint detection, evaluation, and auxiliary regulation device is proposed, comprising the following modules:

[0113] The brain state detection module is used to collect and process the user's sleep brain state signals and, through feature analysis, identify sleep phases and obtain quantitative features of sleep brain state.

[0114] The breathing event detection module is used to collect, process, and perform dynamic analysis on the user's sleep breathing behavior signals to obtain the first sleep breathing dynamic signal and extract the basic information of the sleep breathing event through event detection and analysis.

[0115] The breathing state quantification module is used to perform signal decomposition and / or time-frequency analysis on the first sleep breathing dynamics signal to obtain the second sleep breathing dynamics signal, and to perform feature analysis on the two dynamics signals to generate sleep breathing state quantification features.

[0116] The intensity level assessment module is used to compare and analyze the relevant feature changes of the sleep breathing event based on the quantitative features of the sleep brain state, the quantitative features of the sleep breathing state, and the sleep phase stage, and to identify the event intensity and event level that generated the sleep breathing event.

[0117] The sleep-assisted regulation module is used to predict the trends of the first sleep respiratory dynamics signal, the quantitative characteristics of sleep respiratory state and the quantitative characteristics of sleep brain state, and, in combination with the sleep respiratory knowledge base and the user's sleep respiratory database, generate a sleep respiratory brain state-assisted regulation strategy and send it to the sleep regulation intervention device through a signal interface.

[0118] The report data management module is used to generate and output a sleep breathing brain state joint detection evaluation and auxiliary adjustment report according to a preset cycle or strategy, and update the user's sleep breathing database.

[0119] A data visualization module is used for unified visualization and management of all process data and / or result data in the device.

[0120] The data management center module is used for unified storage and data operation management of all process data and / or result data in the device.

[0121] This invention provides a method, system, and device for the joint detection, evaluation, and auxiliary regulation of sleep-brain and respiration. By scientifically and comprehensively quantifying and evaluating sleep-breathing behavior processes and events, sleep breathing, and sleep-brain states, and through trend prediction of sleep states and dynamic generation and transmission of sleep-breathing-brain state auxiliary regulation strategies, it further optimizes and improves the efficiency and effectiveness of existing sleep-breathing-brain state regulation devices. This achieves an integrated architecture for innovative assessment and auxiliary regulation of sleep-breathing-brain states, assisting users in sleep breathing and improving sleep quality. In practical applications, sleep-breathing-related detection and regulation systems or devices can integrate all or part of the technical points or functions provided by this invention to better meet the needs of different user service scenarios.

[0122] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0123] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of the present invention and do not constitute a limitation on the technical solutions of the present invention.

[0124] Figure 1 This is a schematic diagram of the process steps of a method for joint detection, evaluation and auxiliary regulation of sleep brain and respiration provided in an embodiment of the present invention;

[0125] Figure 2 This is a schematic diagram of the module composition of a sleep brain and breathing joint detection, evaluation and auxiliary regulation system provided in an embodiment of the present invention;

[0126] Figure 3 This is a schematic diagram of the module structure of a sleep brain and breathing joint detection, evaluation and auxiliary regulation device provided in an embodiment of the present invention. Detailed Implementation

[0127] To more clearly illustrate the objectives and technical solutions of this invention, the invention will be further described below with reference to the accompanying drawings of the embodiments of this application. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort should fall within the protection scope of this invention. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other.

[0128] The applicant found that, firstly, the sleep breathing behavior of normal healthy individuals possesses a relatively stable physiological process with excellent periodicity and stability. However, this periodicity and stability are also very fragile and easily affected by sleep breathing events. Sleep breathing disturbances can transform a relatively stable process into a non-stationary one. Secondly, respiratory movements of the mouth, nose, chest, and abdomen are the most direct and accurate representation of respiration. Changes in signals such as pressure, tidal volume, CO2 concentration, temperature, and skin conductance all exhibit excellent sleep respiratory dynamics characteristics, accurately depicting sleep breathing patterns, intensity, and abrupt events. Furthermore, sleep breathing patterns and regularities, as well as the probability and intensity of respiratory events, vary under different sleeping positions and sleep phases. Thirdly, changes or disturbances in the central nervous system are largely influenced by different sleep phases and sleep breathing behavior disturbances, particularly leading to shallow sleep, disordered and fragmented sleep phases. Sleep memories and dreams also affect people's sleep breathing state and patterns. Therefore, a comprehensive consideration of these factors is necessary to achieve a more scientific and comprehensive evaluation and dynamic auxiliary regulation of sleep breathing behavior.

[0129] Therefore, the technical solution of this invention creatively proposes a path for the joint detection, quantification, and auxiliary regulation of sleep breathing and sleep brain state, so as to realize the scientific detection and quantification of the system, assist users in sleep breathing, and improve sleep quality.

[0130] Combination Figure 1 As shown in the figure, the method for joint detection, evaluation, and auxiliary regulation of sleep brain and respiration provided by the present invention includes the following steps:

[0131] P100: Collects and processes the user's sleep brain state signals and performs feature analysis to identify sleep phases and obtain quantitative features of sleep brain state.

[0132] In this embodiment, the sleep brain state signal includes at least one of the following: electroencephalogram (EEG) signal, magnetic resonance imaging (MRI) signal, functional magnetic resonance imaging (fMRI) signal, and functional near-infrared imaging (FIR) signal.

[0133] In this embodiment, the acquisition and processing includes at least acquisition, analog-to-digital conversion, resampling, rereference, artifact removal, signal correction, noise reduction, power frequency notch filtering, bandpass filtering, mean filtering, smoothing, and signal time window segmentation; wherein, signal time window segmentation specifically involves continuously segmenting the target signal into time windows with a preset length time window and a preset translation time step to obtain a multi-time window signal set.

[0134] In this embodiment, the feature analysis includes at least numerical feature analysis, envelope feature analysis, time-frequency feature analysis, and nonlinear feature analysis; wherein, the numerical features include at least the mean, root mean square, maximum, minimum, variance, standard deviation, coefficient of variation, kurtosis, and skewness; the time-frequency features include at least the band power, the band power percentage, and the band center frequency; and the nonlinear features include at least the entropy features, fractal features, and complexity features.

[0135] The first step is to collect and process the brain state signals during the user's sleep process to obtain sleep brain state signals.

[0136] In this embodiment, electroencephalogram (EEG) signals are selected as brain state signals to describe the specific implementation method of the present invention. Sleep EEG signals of the user are acquired using a portable EEG machine, with acquisition electrodes F3 and F4, and reference electrodes M1 and M2, at a sampling rate of 1024Hz. The unified processing of the EEG physiological signals includes artifact removal, signal correction, wavelet noise reduction, and notch filtering at 50Hz power frequency and its harmonics using a Hamming window and a zero-phase FIR digital filter, followed by bandpass filtering from 1-240Hz to obtain the sleep brain state signal.

[0137] The second step is to perform feature analysis on the sleep brain state signals to obtain quantitative features of the sleep brain state.

[0138] In this embodiment, single-channel F3-M2 EEG signal features are extracted, including SVD entropy, Katz fractal dimension, and the frequency band power ratio and frequency band center frequency of the 1-40Hz band, as sleep breathing brain state features.

[0139] The third step is to identify sleep phases based on sleep brain state signals and generate sleep phase curves.

[0140] In this embodiment, sleep phase segmentation includes at least the wakefulness stage, light sleep stage, deep sleep stage, and REM sleep stage; the method for generating sleep phase segmentation and sleep phase curve is as follows:

[0141] 1) By using machine learning to learn and train the sleep brain state signals and their corresponding sleep stage data of a large-scale sleep user sample, a sleep phase stage model is obtained.

[0142] 2) Input the current user's sleep brain state signal into the sleep phase segmentation model to obtain the corresponding sleep phase segmentation;

[0143] 3) Extract the values ​​of the sleep phase stages of all signal time windows according to the time sequence to obtain the sleep phase curve.

[0144] P200: Collects, processes, and performs dynamic analysis on the user's sleep breathing behavior signals to obtain the first sleep breathing dynamic signal. After event detection and analysis, it extracts the basic event information of sleep breathing events.

[0145] In this embodiment, the sleep breathing behavior signal includes at least the sleep breathing movement signal and the sleep position signal; the sleep breathing movement signal includes at least any one of the following: oral and nasal temperature monitoring signal, nasal pressure monitoring signal, oral and nasal CO2 monitoring signal, chest and abdominal respiratory movement signal, electrocardiogram-derived respiratory signal, and pharyngeal electromyography signal; the sleep position signal includes at least the sleeping posture direction signal and the sleeping posture angle signal.

[0146] In this embodiment, the first sleep respiratory dynamics signal is specifically a dynamic curve describing the continuous strength changes of sleep breathing, including at least one of the following: oral and nasal temperature dynamics signal, nasal pressure dynamics signal, oral and nasal CO2 dynamics signal, chest and abdominal respiratory movement dynamics signal, electrocardiogram-derived respiratory dynamics signal, pharyngeal electromyography dynamics signal, as well as a sleeping posture direction description signal and a sleeping posture angle description signal.

[0147] The first step is to collect and process the breathing behavior signals during the user's sleep process to obtain sleep breathing behavior signals.

[0148] In this embodiment, nasal pressure monitoring signals and postural acceleration signals are used as sleep breathing behavior signals to describe the specific implementation process of the present invention.

[0149] In this embodiment, a nasal pressure sensor is used to collect the user's nasal breathing airflow during sleep, with a sampling rate of 128Hz, to obtain a sleep nasal pressure monitoring signal. In practical applications, oral and nasal thermistors, nasal pressure sensors, and exhaled CO2 sensors all detect sleep breathing movements through the oral and nasal airflow channels; pressure sensors, impedance sensors, etc., can detect sleep breathing movements in the chest, abdomen, and esophagus, obtaining respiratory effort channel signal data. The appropriate method and means to collect and record sleep breathing behavior signals need to be selected based on the specific circumstances of the user and the task. Subsequently, the signal analysis and processing of the nasal pressure monitoring signal mainly involves artifact removal, signal correction, 6Hz low-pass filtering, and signal time window segmentation (2s window, 1s step). It is worth mentioning that the sleep nasal breathing airflow signal can very directly, sensitively, and accurately express the user's sleep breathing behavior; simple signal processing can yield a very good sleep respiratory dynamics signal.

[0150] In this embodiment, the acceleration signal at the center of the forehead (Fpz) and above the right chest is collected by a wireless three-axis gyroscope sensor patch device at a sampling rate of 128Hz to generate a sleep acceleration signal. Further, the sleep acceleration signal is subjected to artifact removal, signal correction, 4Hz low-pass filtering, and signal time window segmentation (2s window, 1s step). The sleeping posture direction signal and sleeping posture angle signal are extracted by coordinate conversion, thus obtaining the sleep position signal.

[0151] The second step is to perform dynamic analysis on the sleep breathing behavior signals to obtain the first sleep breathing dynamic signal.

[0152] In this embodiment, the dynamic analysis specifically involves analyzing the dynamic properties, signal quality, and morphology of each physiological signal in the sleep breathing behavior signal, selecting target signals, and fusing the signals to obtain the sleep breathing dynamic description signal.

[0153] In this embodiment, the first sleep respiratory dynamics signal includes nasal pressure dynamics signal, sleeping posture direction description signal, and sleeping posture angle description signal.

[0154] The third step is to perform event detection and analysis on the first sleep respiratory dynamics signal to identify sleep breathing events and extract basic event information.

[0155] In this embodiment, a sleep breathing event includes at least the event type, start time, end time, duration, peak and trough values, time at the peak and trough values, event intensity, and event level. Basic event information includes at least the event type, start time, end time, duration, peak and trough values, and time at the peak and trough values.

[0156] In this embodiment, the event detection and analysis specifically involves identifying and extracting waveform features from the sleep breathing dynamics signal based on a preset sleep breathing event knowledge base and / or machine learning model to obtain basic event information of the sleep breathing event.

[0157] In this embodiment, firstly, machine learning is used to train and model large-scale sample data of sleep respiratory dynamics signals and sleep breathing events from various sources (oral and nasal pressure, thermal sensitivity, chest and abdominal girdle pressure, etc.) to pre-construct a sleep breathing event recognition model. Then, basic information about sleep breathing events is extracted through model application and data annotation of the sleep breathing event recognition model.

[0158] In practical applications, machine learning can deliver excellent detection and analysis results after learning from large-scale data, and it can continuously relearn and update the model.

[0159] P300: Perform signal decomposition and / or time-frequency analysis on the first sleep respiratory dynamics signal to obtain the second sleep respiratory dynamics signal, and perform feature analysis on the above two dynamics signals to generate quantitative features of sleep breathing state.

[0160] In this embodiment, the second sleep respiratory dynamics signal includes at least a sleep respiratory dynamics cycle signal, a sleep respiratory dynamics trend signal, and a sleep respiratory dynamics oscillation signal.

[0161] In this embodiment, signal decomposition or time-frequency analysis specifically involves extracting the periodic components, trend components, residuals, or oscillation components of the sleep respiratory dynamics signal from the first sleep respiratory dynamics signal to generate a second sleep respiratory dynamics signal, and further extracting the periodic features, trend features, and stationary features of the sleep respiratory dynamics signal through feature analysis.

[0162] In this embodiment, the signal decomposition methods include at least time series decomposition, empirical mode decomposition, variational mode decomposition, local mean decomposition, wavelet transform, wavelet packet transform, time-frequency transform, blind source separation, linear discriminant analysis, detrending analysis, principal component analysis, independent component analysis, waveform analysis, and numerical fitting, as well as their evolutionary variants; the time-frequency analysis methods include at least any one of time-frequency transform, time-domain filtering, and frequency-domain filtering.

[0163] The first step is to perform signal decomposition and / or time-frequency analysis on the first sleep respiratory dynamics signal to obtain the second sleep respiratory dynamics signal.

[0164] In this embodiment, the time series decomposition method - STL decomposition method - is selected to decompose the sleep respiratory dynamics signal, extract the seasonal component, trend component and residual component of the signal, that is, to obtain the sleep respiratory dynamics periodic signal, the sleep respiratory dynamics trend signal and the sleep respiratory dynamics residual signal, and generate the second sleep respiratory dynamics signal.

[0165] It is worth noting that in practical applications, decomposition methods such as empirical mode decomposition and variational mode decomposition, combined with time-frequency analysis methods such as time-domain filtering and frequency-domain filtering, can meet the needs of extracting periodic signals, trend signals, and oscillating signals in different scenarios. However, the filtering parameter settings need to be considered in conjunction with the signal sampling rate. Using the sleep respiratory frequency range of normal healthy people corresponding to the current user's basic information, the upper and lower frequency limits in time-frequency analysis can be easily obtained through conversion. Then, the decomposed signal components are identified and combined to obtain the second sleep respiratory dynamics signal.

[0166] The second step is to perform feature analysis on the first and second sleep respiratory dynamics signals to obtain quantitative features of sleep breathing state.

[0167] In this embodiment, the quantitative characteristics of sleep breathing state include at least sleep respiratory dynamic characteristics, sleep respiratory dynamic periodic characteristics, sleep respiratory dynamic trend characteristics, sleep respiratory dynamic stationary characteristics, and sleep posture characteristics; the sleep respiratory dynamic characteristics include at least respiratory rate, numerical characteristics, time-frequency characteristics, and nonlinear characteristics; the sleep posture characteristics include at least sleeping position direction characteristics and sleeping position angle characteristics.

[0168] In this embodiment, respiratory rate, sample entropy, trend intensity, period intensity, stability intensity, sleeping posture direction characteristics, and sleeping posture angle characteristics are used as quantitative features of sleep breathing state.

[0169] In this embodiment, the calculation formulas for trend strength, periodic strength, and stability strength are as follows:

[0170]

[0171]

[0172]

[0173] Among them, F T For trend strength, F S Periodic intensity, F R Stable strength, T t S t R t These represent the sleep respiratory dynamics trend signal, the sleep respiratory dynamics cycle signal, and the sleep respiratory dynamics residual signal, respectively. Var() is the variance operator, and max() is the maximum value operator.

[0174] P400: Based on the quantitative characteristics of sleep brain state, the quantitative characteristics of sleep breathing state, and the sleep phase segmentation, compare and analyze the relevant characteristic changes of the sleep breathing events to identify the event intensity and event level that generated the sleep breathing events.

[0175] The first step is to compare and analyze the relative changes in the quantitative characteristics of sleep brain state and the quantitative characteristics of sleep breathing state based on the occurrence period and comparison interval of sleep breathing events, and obtain a set of changes in event description features.

[0176] In this embodiment, the event comparison interval is selected as the time interval between the end of the previous event and the start of the current event. If the time interval spans multiple time windows, the average of different target feature values ​​within these time windows is directly taken as the feature value of the comparison interval for that feature. The event description feature change set includes the respiratory rate, sample entropy, trend intensity, period intensity, and stationary intensity in the quantitative features of sleep breathing state, as well as the relative changes in SVD entropy, Katz fractal dimension, frequency band power ratio, and frequency band center frequency in the quantitative features of sleep brain state.

[0177] The second step is to identify the intensity of generated events based on sleep breathing events, the set of changes in event description features, and sleep phases.

[0178] In this embodiment, the method for generating event intensity is as follows:

[0179] 1) Obtain the set of event description feature changes, select key target features and perform numerical weighted calculation of feature values ​​to obtain the event feature change correction coefficient;

[0180] 2) Obtain the start time, end time, duration, peak and trough values, and time at the peak and trough values ​​of sleep breathing events, as well as preset sleep breathing dynamics signal thresholds;

[0181] 3) Calculate the linear slope based on the onset time, peak and trough values, and time at the peak and trough values ​​of sleep breathing events to obtain the slope of the peak and trough leading edge;

[0182] 4) Calculate the linear slope based on the end time of sleep breathing events, peak and trough values, and the time at the peak and trough values ​​to obtain the slope at the trailing edge of the peak and trough;

[0183] 5) Based on the current sleep phase stage, extract the sleep phase correction coefficient according to the preset sleep phase correction coefficient reference table;

[0184] 6) The event intensity is obtained by fusion calculation of the event characteristic change correction coefficient, peak and valley front slope, peak and valley back slope, and sleep phase correction coefficient.

[0185] In this embodiment, the event feature change correction coefficient is obtained by weighting the (average) values ​​of the eigenvalues ​​in the event description feature change set, including respiratory rate, sample entropy, trend intensity, periodic intensity, stationary intensity, SVD entropy, Katz fractal dimension, frequency band power ratio, and frequency band center frequency.

[0186] In this embodiment, the reference relationships in the preset sleep phase correction coefficient reference table are: wakefulness -0.80, light sleep -0.95, deep sleep -0.90, and REM sleep -1.0.

[0187] In this embodiment, the event intensity is equal to the product of the event feature change correction coefficient, the peak-valley leading edge slope, the peak-valley trailing edge slope, and the sleep phase correction coefficient. In practical applications, different user scenarios and different key feature indicators can be constructed using different numerical fusion calculation methods.

[0188] The third step is to identify and generate event levels based on sleep breathing events, sleep posture characteristics, and sleep phases.

[0189] In this embodiment, the method for generating event levels is as follows:

[0190] 1) Obtain the start time, end time, duration, peak and trough values, and time at the peak and trough values ​​of sleep breathing events, as well as preset sleep breathing dynamics signal thresholds;

[0191] 2) Calculate the relative changes between the peak and trough values ​​of sleep breathing events and the preset sleep breathing dynamics signal threshold to obtain the peak-trough relative values;

[0192] 3) Calculate the product of the duration of each sleep breathing event and the relative value of the peaks and troughs to obtain the event level description feature value;

[0193] 4) Based on the characteristics of sleep posture, extract the information of sleeping posture angle and sleeping posture direction or preset the sleep posture correction calculation strategy, and extract the sleep posture correction coefficient according to the preset sleep posture correction coefficient reference table.

[0194] 5) Based on the current sleep phase stage, extract the sleep phase correction coefficient according to the preset sleep phase correction coefficient reference table;

[0195] 6) The event level is obtained by multiplying the event level description feature value, sleep position correction coefficient, and sleep phase correction coefficient according to the prediction level classification rules.

[0196] In this embodiment, the relative value of peak and valley is the percentage of the increment, and the specific calculation formula is as follows:

[0197]

[0198] Among them, iF, V f V b These are the peak-to-trough relative values, the peak-to-trough values ​​of sleep breathing events, and the preset sleep breathing dynamics signal thresholds, respectively.

[0199] In this embodiment, a preset sleep position correction calculation strategy is used: the coefficient for a supine 0° angle is 0.5 and the coefficient for a prone 180° angle is 1.0 for linear angle calculation; the coefficient for a left lateral decubitus position is a constant of 0.95, the coefficient for a right lateral decubitus position is a constant of 0.90, the coefficient for a supine position is a constant of 0.85, and the coefficient for a prone position is a constant of 1.0.

[0200] In this embodiment, the event level descriptive feature value (the area between the duration and the relative peak and trough) can directly describe the severity level of sleep apnea events. The product of the event level descriptive feature value, the sleep position correction coefficient, and the sleep phase correction coefficient equals the event level index value. The prediction level classification rule starts with 0.5 as the first level and increases progressively.

[0201] P500: Performs trend prediction on the first sleep respiratory dynamics signal, the quantitative characteristics of sleep respiratory state, and the quantitative characteristics of sleep brain state. Combines the sleep respiratory knowledge base and the user's sleep respiratory database to generate a sleep respiratory brain state auxiliary regulation strategy and sends it to the sleep regulation intervention device through the signal interface.

[0202] In this embodiment, trend prediction includes at least one of the following: exponential smoothing, Holt-Winters method, AR, MA, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA, VARMAX, and machine learning.

[0203] The first step is to perform trend prediction on the first sleep respiratory dynamics signal to obtain the first sleep respiratory dynamics prediction signal.

[0204] In this embodiment, the ARMA method is selected to predict the trend of the first sleep respiratory dynamics signal, in order to predict the changing trend of respiratory dynamics (nasal pressure dynamics signal, sleeping position direction description signal and sleeping position angle description signal).

[0205] The second step is to perform trend prediction on the quantitative characteristics of sleep breathing state to obtain the predicted characteristics of sleep breathing state.

[0206] In this embodiment, the ARMA method is selected to predict the trend of quantitative features of sleep breathing state, in order to predict the changing trend of breathing state features (breathing frequency, sample entropy, trend intensity, period intensity, stability intensity, sleeping posture direction features, and sleeping posture angle features).

[0207] The third step is to perform trend prediction on the quantitative characteristics of sleep brain state to obtain the predicted characteristics of sleep brain state.

[0208] In this embodiment, the ARMA method is selected to predict the trend of brain state quantitative characteristics during sleep, in order to predict the changing trend of brain state characteristics (SVD entropy, Katz fractal dimension, frequency band power ratio and frequency band center frequency).

[0209] The fourth step is to aggregate the first sleep respiratory dynamics prediction signal, sleep breathing state prediction features, and sleep brain state prediction features to obtain a sleep breathing and brain state prediction description.

[0210] Step 5: Collect the first sleep respiratory dynamics signal, the quantitative characteristics of sleep breathing state, and the quantitative characteristics of sleep brain state to obtain the current description of sleep breathing and brain state.

[0211] Step 6: Based on the predicted description of sleep breathing and brain state and the current description of sleep breathing and brain state, combined with the sleep breathing knowledge base and the user's sleep breathing database, generate a sleep breathing brain state auxiliary regulation strategy.

[0212] In this embodiment, the sleep breathing knowledge base mainly comes from knowledge and experience related to health management and clinical medicine related to sleep breathing, including at least sleep knowledge, sleep breathing patterns, characteristics of common sleep breathing events, commonly used sleep breathing regulation methods, and guidance on scenario intervention parameters; the user sleep breathing database includes at least sleep breathing events, first sleep breathing dynamics signals, second sleep breathing dynamics signals, quantitative characteristics of sleep breathing state, quantitative characteristics of sleep brain state, and sleep breathing brain state auxiliary regulation strategies.

[0213] In this embodiment, the sleep breathing brain state assisted regulation strategy includes at least the target value of respiratory rate, the target value of respiratory depth, the regulation method, the regulation time point, the duration, and the device control parameters.

[0214] In this embodiment, the target respiratory rate is used as the observation index, and environmental temperature and humidity regulation is used as the adjustment method / means to generate a sleep breathing brain state auxiliary regulation strategy. First, based on the sleep breathing knowledge base, machine learning is used to model and learn from large-scale healthy population sample data with different breathing states, brain states, and different regulation strategy parameters, including first sleep breathing dynamics signals, quantitative characteristics of sleep breathing states, and quantitative characteristics of sleep brain states, to obtain a population breathing brain state regulation path planning model. Second, the predicted description of sleep breathing and brain states and the current description of sleep breathing and brain states are input into the population breathing brain state regulation path planning model to obtain the basic method of regulation path planning (target step size, adjustment time point, duration, and device control parameters). Finally, combined with the user's sleep breathing database, the user's historical auxiliary regulation effect is combined to generate the final sleep breathing brain state auxiliary regulation strategy.

[0215] In practical applications, it's best to choose relatively gentle and gradual sleep-breathing brain state regulation strategies that promote a positive sleep experience for the user, achieving the desired assistance while minimizing disruption to the user's normal sleep process. However, for more intense sleep breathing events, faster and more efficient sleep-breathing brain state regulation strategies are required.

[0216] Step 7: Send the sleep breathing brain state auxiliary regulation strategy to the sleep regulation intervention device through the signal interface, optimize the control execution of the sleep regulation intervention device, so as to realize the dynamic auxiliary regulation of the user's sleep breathing brain state.

[0217] In this embodiment, in conjunction with the foregoing, an environmental temperature and humidity control device is selected as a sleep regulation intervention device.

[0218] In practical applications, sleep regulation intervention devices include at least one of the following: ventilator, smart mattress, odor stimulation device, electrical stimulation device, magnetic stimulation device, tactile stimulation device, environmental temperature and humidity control device, and environmental CO2 concentration control device. The selection should be based on the specific equipment and facilities available to the user, as long as the sleep regulation intervention device possesses basic intelligent functions such as command reception and parsing, and remote wireless control.

[0219] In addition, non-contact, detachable, or patch-type miniature sleep regulation intervention devices or methods are always the preferred choice when ensuring regulatory effectiveness.

[0220] P600: Generates and outputs a joint detection, evaluation and auxiliary adjustment report of sleep breathing and brain state according to a preset cycle or strategy, and updates the user's sleep breathing database.

[0221] The first step is to generate or output a report on the combined detection, evaluation, and auxiliary regulation of sleep breathing and brain state according to the preset reporting cycle.

[0222] In this embodiment, the combined detection, evaluation, and auxiliary regulation report of sleep breathing and brain state includes at least the statistical analysis of sleep breathing events, the summary of sleep breathing state evaluation, the summary of sleep brain state evaluation, the summary of auxiliary regulation evaluation, and suggestions for improving sleep behavior.

[0223] The second step is to update the user's sleep breathing database according to the preset data update strategy.

[0224] In this embodiment, each time a data operation such as data acquisition, processing, or calculation is completed, the data is directly updated and written to the user's sleep breathing database.

[0225] Combination Figure 2 As shown in the figure, the sleep brain and breathing joint detection, evaluation and auxiliary regulation system provided by the present invention includes the following modules:

[0226] The brain state detection module S100 is used to collect and process the user's sleep brain state signals and, through feature analysis, identify sleep phase stages to obtain quantitative features of sleep brain state.

[0227] The breathing event detection module S200 is used to collect, process, and perform dynamic analysis on the user's sleep breathing behavior signals to obtain the first sleep breathing dynamic signal and extract the basic information of the sleep breathing event through event detection and analysis.

[0228] The breathing state quantification module S300 is used to perform signal decomposition and / or time-frequency analysis on the first sleep breathing dynamics signal to obtain the second sleep breathing dynamics signal, and to perform feature analysis on the two dynamics signals to generate sleep breathing state quantification features.

[0229] The intensity level assessment module S400 is used to compare and analyze the relevant feature changes of sleep breathing events based on the quantitative characteristics of sleep brain state, the quantitative characteristics of sleep breathing state, and sleep phase stages, and to identify the event intensity and event level that generated the sleep breathing event.

[0230] The sleep-assisted regulation module S500 is used to predict the trends of the first sleep respiratory dynamics signal, the quantitative characteristics of sleep respiratory state and the quantitative characteristics of sleep brain state, and, combined with the sleep respiratory knowledge base and the user's sleep respiratory database, generate a sleep respiratory brain state-assisted regulation strategy and send it to the sleep regulation intervention device through the signal interface.

[0231] The report data management module S600 is used to generate and output a joint detection, evaluation and auxiliary adjustment report of sleep breathing brain state according to a preset cycle or strategy, and update the user's sleep breathing database.

[0232] The S700 data operation management module is used for visual management, unified storage, and operation management of all process data of the system.

[0233] In this embodiment, the brain state detection module S100 further includes the following functional units:

[0234] The brain signal detection unit is used to collect and process brain state signals during the user's sleep process to obtain sleep brain state signals.

[0235] The brain state feature analysis unit is used to perform feature analysis on sleep brain state signals to obtain quantitative features of sleep brain state.

[0236] The sleep phase analysis unit is used to identify sleep phase stages and generate sleep phase curves based on sleep brain state signals.

[0237] In this embodiment, the respiratory event detection module S200 further includes the following functional units:

[0238] The respiratory signal detection unit is used to collect and process the respiratory behavior signals of the user during sleep to obtain sleep respiratory behavior signals.

[0239] The dynamic analysis unit is used to perform dynamic analysis on sleep breathing behavior signals to obtain the first sleep breathing dynamic signal;

[0240] The event detection and analysis unit is used to perform event detection and analysis on the first sleep respiratory dynamics signal, identify sleep breathing events, and extract basic event information.

[0241] In this embodiment, the respiratory status quantification module S300 further includes the following functional units:

[0242] The signal decomposition and extraction unit is used to perform signal decomposition and / or time-frequency analysis on the first sleep respiratory dynamics signal to obtain the second sleep respiratory dynamics signal.

[0243] The respiratory feature analysis unit is used to perform feature analysis on the first sleep respiratory dynamics signal and the second sleep respiratory dynamics signal to obtain quantitative features of sleep breathing state.

[0244] In this embodiment, the strength level assessment module S400 further includes the following functional units:

[0245] The feature change calculation unit is used to compare and analyze the relative changes of the quantitative features of sleep brain state and the quantitative features of sleep breathing state based on the event occurrence period and event comparison interval of sleep breathing events, and obtain the event description feature change set.

[0246] The event intensity quantification unit is used to identify and generate event intensity based on sleep breathing events, the set of changes in event description features, and sleep phases.

[0247] The event grading unit is used to identify and generate event grades based on sleep breathing events, sleep posture characteristics, and sleep phases.

[0248] In this embodiment, the sleep assistance adjustment module S500 further includes the following functional units:

[0249] A respiratory dynamics prediction unit is used to predict the trend of the first sleep respiratory dynamics signal to obtain the first sleep respiratory dynamics prediction signal.

[0250] The breathing state prediction unit is used to predict the trend of quantitative characteristics of sleep breathing state and obtain sleep breathing state prediction characteristics.

[0251] The brain state prediction unit is used to predict the trend of quantitative characteristics of sleep brain state and obtain sleep brain state prediction characteristics.

[0252] The prediction description aggregation unit is used to aggregate the first sleep respiratory dynamics prediction signal, sleep breathing state prediction features and sleep brain state prediction features to obtain a sleep breathing and brain state prediction description.

[0253] The baseline description aggregation unit is used to aggregate the first sleep respiratory dynamics signal, the quantitative characteristics of sleep breathing state and the quantitative characteristics of sleep brain state to obtain the current description of sleep breathing and brain state.

[0254] The auxiliary strategy generation unit is used to generate auxiliary regulation strategies for sleep breathing and brain state based on the predicted description of sleep breathing and brain state and the current description of sleep breathing and brain state, combined with the sleep breathing knowledge base and the user's sleep breathing database.

[0255] The strategy-assisted adjustment unit is used to send the sleep breathing brain state auxiliary adjustment strategy to the sleep regulation intervention device through the signal interface, optimize the control execution of the sleep regulation intervention device, and realize the dynamic auxiliary adjustment of the user's sleep breathing brain state.

[0256] In this embodiment, the report data management module S600 further includes the following functional units:

[0257] The report generation and output unit is used to generate or output a report on the joint detection, evaluation and auxiliary regulation of sleep breathing and brain state according to a preset reporting cycle.

[0258] The database update unit is used to update the user's sleep breathing database according to a preset data update strategy.

[0259] In this embodiment, the data operation management module S700 further includes the following functional units:

[0260] The user information management unit is used for registering, inputting, editing, querying, outputting, and deleting basic user information;

[0261] The data visualization management unit is used for the visualization and management of all data in the system.

[0262] The data operations management unit is used for the storage, backup, migration, and export of all data in the system.

[0263] The system is configured for corresponding execution. Figure 1 The steps in the method will not be elaborated here.

[0264] Combination Figure 3 As shown in the figure, the sleep brain and breathing joint detection, evaluation and auxiliary regulation device provided by the present invention includes the following modules:

[0265] The brain state detection module M100 is used to collect and process the user's sleep brain state signals and, through feature analysis, identify sleep phase stages to obtain quantitative features of sleep brain state.

[0266] The M200 respiratory event detection module is used to collect, process, and perform dynamic analysis on the user's sleep breathing behavior signals to obtain the first sleep breathing dynamic signal and extract the basic information of the sleep breathing events through event detection and analysis.

[0267] The breathing state quantification module M300 is used to perform signal decomposition and / or time-frequency analysis on the first sleep breathing dynamics signal to obtain the second sleep breathing dynamics signal, and to perform feature analysis on the above two dynamics signals to generate sleep breathing state quantification features.

[0268] The intensity level assessment module M400 is used to compare and analyze the relevant feature changes of sleep breathing events based on the quantitative characteristics of sleep brain state, the quantitative characteristics of sleep breathing state, and sleep phase stages, and to identify the event intensity and event level that generated the sleep breathing event.

[0269] The sleep-assisted regulation module M500 is used to predict trends in the first sleep respiratory dynamics signal, the quantitative characteristics of sleep respiratory state and the quantitative characteristics of sleep brain state. Combined with the sleep respiratory knowledge base and the user's sleep respiratory database, it generates a sleep respiratory brain state-assisted regulation strategy and sends it to the sleep regulation intervention device through the signal interface.

[0270] The M600 report data management module is used to generate and output a joint detection, evaluation and auxiliary adjustment report of sleep breathing and brain state according to a preset cycle or strategy, and update the user's sleep breathing database.

[0271] The M700 data visualization module is used for unified visualization and management of all process data and / or result data in the device.

[0272] The M800 data management center module is used for unified storage and data operation management of all process data and / or result data in the device.

[0273] The device is configured to perform the corresponding action. Figure 1 The steps in the method will not be elaborated here.

[0274] The present invention also provides various programmable processors (FPGA, ASIC or other integrated circuits) for running programs, wherein the programs execute the steps in the above embodiments.

[0275] The present invention also provides a corresponding computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the memory executes the program to implement the steps in the above embodiments.

[0276] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art can make any modifications, changes, or equivalent substitutions in the form and details of the implementation without departing from the spirit and principles disclosed in this invention; all such modifications and changes fall within the scope of protection of this invention. Therefore, the scope of patent protection for this invention shall be determined by the scope defined in the appended claims.

Claims

1. A sleep brain and respiration combined detection, evaluation and auxiliary adjustment method, characterized in that, The method comprises the following steps: Collecting and processing sleep brain state signals of a user and performing feature analysis to identify sleep phase staging and obtain sleep brain state quantitative features; Collecting and processing sleep breathing behavior signals of the user and performing dynamic analysis to obtain first sleep breathing dynamic signals and perform event detection analysis to extract event basic information of sleep breathing events; Performing signal decomposition and / or time-frequency analysis on the first sleep breathing dynamic signals to obtain second sleep breathing dynamic signals, and performing feature analysis on the two dynamic signals to generate sleep breathing state quantitative features; Comparatively analyzing relevant feature changes of the sleep breathing events according to the sleep brain state quantitative features, the sleep breathing state quantitative features, and the sleep phase staging, and identifying event intensity and event level of the sleep breathing events; Performing trend prediction on the first sleep breathing dynamic signals, the sleep breathing state quantitative features, and the sleep brain state quantitative features, combining a sleep breathing knowledge base and a user sleep breathing database, generating sleep breathing brain state auxiliary adjustment strategies, and sending sleep adjustment intervention equipment through a signal interface; Generating and outputting sleep breathing brain state joint detection evaluation and auxiliary adjustment reports according to a preset period or strategy, and updating the user sleep breathing database; The specific steps of comparatively analyzing relevant feature changes of the sleep breathing events according to the sleep brain state quantitative features, the sleep breathing state quantitative features, and the sleep phase staging, and identifying event intensity and event level of the sleep breathing events further comprise: Comparatively analyzing relative changes of the sleep brain state quantitative features and the sleep breathing state quantitative features according to an event occurrence period and an event comparative interval of the sleep breathing events to obtain an event description feature change set; Identifying the event intensity according to the sleep breathing event, the event description feature change set, and the sleep phase staging; Identifying the event level according to the sleep breathing event, a sleep body position feature, and the sleep phase staging.

2. The method of claim 1, wherein, The specific steps of collecting and processing sleep brain state signals of a user and performing feature analysis to identify sleep phase staging and obtain sleep brain state quantitative features further comprise: Collecting and obtaining brain state signals during a sleep process of the user and performing signal processing to obtain the sleep brain state signals; Performing feature analysis on the sleep brain state signals to obtain the sleep brain state quantitative features; Identifying the sleep phase staging according to the sleep brain state signals and generating a sleep phase curve.

3. The method of claim 2, wherein The sleep brain state signals at least include any one of an electroencephalogram signal, an electroencephalogram signal, a functional magnetic resonance imaging signal, and a functional near-infrared imaging signal.

4. The method of claim 2, wherein The collection and processing at least include collection and acquisition, analog-to-digital conversion, resampling, re-referencing, artifact rejection, signal correction, noise reduction, power frequency notch, band-pass filtering, mean filtering, smoothing processing, and signal time window segmentation; wherein the signal time window segmentation specifically refers to continuously time window segmenting a target signal with a preset length time window and a preset translation time step to obtain a multi-time window signal set.

5. The method of claim 2, wherein The feature analysis at least includes any one of numerical feature analysis, envelope feature analysis, time-frequency feature analysis, and nonlinear feature analysis; wherein the numerical feature at least includes any one of mean value, root mean square, maximum value, minimum value, variance, standard deviation, coefficient of variation, kurtosis, and skewness; the time-frequency feature at least includes any one of frequency band power, frequency band power ratio, and frequency band center frequency; and the nonlinear feature at least includes any one of entropy feature, fractal feature, and complexity feature.

6. The method according to any one of claims 2 to 5, wherein, The sleep phase staging at least includes wakefulness period, light sleep period, deep sleep period, and rapid eye movement sleep period; and the generation method of the sleep phase curve and the sleep phase staging is specifically as follows:

1. Learning and training of the sleep brain state signal and corresponding sleep staging data of a large-scale sleep user sample through machine learning and data modeling to obtain a sleep phase staging model; 2. inputting the sleep brain state signal of a current user into the sleep phase staging model to obtain corresponding sleep phase staging; 3. extracting numerical values of the sleep phase staging of all signal time windows in time sequence to obtain the sleep phase curve.

7. The method of claim 1, wherein, The sleep respiratory behavior signal of a user is collected and processed and subjected to dynamics analysis to obtain a first sleep respiratory dynamics signal, and the specific steps of extracting a sleep respiratory event further include: collecting and acquiring a respiratory behavior signal of a user during sleep and performing signal processing to obtain the sleep respiratory behavior signal; performing dynamics analysis on the sleep respiratory behavior signal to obtain the first sleep respiratory dynamics signal; performing event detection analysis on the first sleep respiratory dynamics signal to identify and extract basic information of the sleep respiratory event.

8. The method of claim 7, wherein, The sleep respiratory behavior signal at least includes sleep respiratory movement signal and sleep body position signal; the sleep respiratory movement signal at least includes any one of oral-nasal temperature monitoring signal, nasal pressure monitoring signal, oral-nasal CO2 monitoring signal, thoraco-abdominal respiratory movement signal, electrocardiogram-derived respiratory signal, and laryngeal muscle electromyography signal; and the sleep body position signal at least includes sleep posture direction signal and sleep posture angle signal.

9. The method of claim 8, wherein, The first sleep respiratory dynamics signal is specifically a dynamics curve describing continuous strength change of sleep respiration, and at least includes any one of oral-nasal temperature dynamics signal, nasal pressure dynamics signal, oral-nasal CO2 dynamics signal, thoraco-abdominal respiratory movement dynamics signal, electrocardiogram-derived respiratory dynamics signal, and laryngeal muscle electromyography dynamics signal, as well as sleep posture direction description signal and sleep posture angle description signal.

10. The method of claim 7, wherein, The dynamics analysis is specifically analysis of dynamics attributes, signal quality, and morphology of each physiological signal in the sleep respiratory behavior signal, selection of a target signal therefrom, and signal fusion to obtain sleep respiratory dynamics description signal.

11. The method of claim 7, wherein, The sleep respiratory event at least includes event type, start time, end time, duration, peak-valley value, time at peak-valley value, event intensity, and event level.

12. The method according to any one of claims 7 to 11, wherein, The event detection analysis is specifically waveform feature recognition and label extraction of the sleep respiratory dynamics signal according to a preset sleep respiratory event knowledge base and / or machine learning model to obtain basic information of the sleep respiratory event.

13. The method of claim 12, wherein, The basic information of the event at least includes an event type, a start time, an end time, a duration, a peak value, and a time at the peak value.

14. The method of claim 1, wherein, The specific steps of signal decomposition and / or time-frequency analysis on the first sleep respiratory dynamics signal to obtain a second sleep respiratory dynamics signal, and feature analysis on the two dynamics signals to generate sleep respiratory state quantification features further include: signal decomposition and / or time-frequency analysis on the first sleep respiratory dynamics signal to obtain the second sleep respiratory dynamics signal; feature analysis on the first sleep respiratory dynamics signal and the second sleep respiratory dynamics signal to obtain the sleep respiratory state quantification features.

15. The method of claim 14, wherein, The second sleep respiratory dynamics signal at least includes a sleep respiratory dynamics periodic signal, a sleep respiratory dynamics trend signal, and a sleep respiratory dynamics oscillation signal.

16. The method of claim 14 or 15, wherein, The signal decomposition or the time-frequency analysis specifically extracts periodic components, trend components, residuals, or oscillation components of the sleep respiratory dynamics signal from the first sleep respiratory dynamics signal to generate the second sleep respiratory dynamics signal, and further extracts periodicity features, trend features, and stationarity features of the sleep respiratory dynamics signal through feature analysis.

17. The method of claim 16, wherein, The signal decomposition method at least includes time series decomposition, empirical mode decomposition, variational mode decomposition, local mean decomposition, wavelet transform, wavelet packet transform, time-frequency transform, blind source separation, linear discriminant analysis, detrend analysis, principal component analysis, independent component analysis, waveform analysis, and numerical fitting, and their evolutionary variants; and the time-frequency analysis method at least includes any one of time-frequency transform, time domain filtering, and frequency domain filtering.

18. The method of claim 14, wherein, The sleep respiratory state quantification features at least include any one of sleep respiratory dynamics features, sleep respiratory dynamics periodicity features, sleep respiratory dynamics trend features, sleep respiratory dynamics stationarity features, and sleep body position features; the sleep respiratory dynamics features at least include any one of respiratory frequency, numerical features, time-frequency features, and nonlinear features; and the sleep body position features at least include sleep posture direction features and sleep posture angle features.

19. The method of claim 1, wherein, The event intensity generation method includes: 1) obtaining the event description feature change set, selecting a key target feature, and performing numerical weighting calculation on the feature value to obtain an event feature change correction coefficient; 2) obtaining the start time, end time, duration, peak value, and time at the peak value of the sleep respiratory event, and a preset sleep respiratory dynamics signal threshold; 3) calculating a linear slope according to the start time, peak value, and time at the peak value of the sleep respiratory event to obtain a peak value front edge slope; 4) calculating a linear slope according to the end time, peak value, and time at the peak value of the sleep respiratory event to obtain a peak value rear edge slope; 5) extracting a sleep phase correction coefficient according to the current sleep phase staging and a preset sleep phase correction coefficient table; 6) performing numerical fusion calculation on the event feature change correction coefficient, the peak value front edge slope, the peak value rear edge slope, and the sleep phase correction coefficient to obtain the event intensity.

20. The method of claim 1 or 19, wherein, The event intensity generation method includes: 1) obtain the start time, end time, duration, peak-trough value and peak-trough value time of the sleep respiratory event, and a preset sleep respiratory dynamics signal threshold; 2) calculate the relative change amount of the peak-trough value of the sleep respiratory event and the preset sleep respiratory dynamics signal threshold to obtain a peak-trough relative value; 3) calculate the product of the duration of the sleep respiratory event and the peak-trough relative value to obtain an event level description characteristic value; 4) according to the sleep position feature, extract the sleeping posture angle and sleeping posture direction information or a preset sleep position correction calculation strategy, and extract a sleep position correction coefficient according to a preset sleep position correction coefficient table; 5) according to the current sleep phase stage, extract a sleep phase correction coefficient according to a preset sleep phase correction coefficient table; 6) according to the numerical product of the event level description characteristic value, the sleep position correction coefficient and the sleep phase correction coefficient, and according to a prediction level division rule, obtain the event level.

21. The method of claim 1, wherein, The specific steps of the trend prediction of the first sleep respiratory dynamics signal, the sleep respiratory state quantitative feature and the sleep brain state quantitative feature, combining the sleep respiratory knowledge base and the user sleep respiratory database, generating the sleep respiratory brain state auxiliary adjustment strategy and sending the specific sleep adjustment intervention device through the signal interface still include: performing trend prediction on the first sleep respiratory dynamics signal to obtain a first sleep respiratory dynamics prediction signal; performing trend prediction on the sleep respiratory state quantitative feature to obtain a sleep respiratory state prediction feature; performing trend prediction on the sleep brain state quantitative feature to obtain a sleep brain state prediction feature; aggregating the first sleep respiratory dynamics prediction signal, the sleep respiratory state prediction feature and the sleep brain state prediction feature to obtain a sleep respiratory and brain state prediction description; aggregating the first sleep respiratory dynamics signal, the sleep respiratory state quantitative feature and the sleep brain state quantitative feature to obtain a sleep respiratory and brain state current description; generating the sleep respiratory brain state auxiliary adjustment strategy according to the sleep respiratory and brain state prediction description and the sleep respiratory and brain state current description, combining the sleep respiratory knowledge base and the user sleep respiratory database; sending the sleep respiratory brain state auxiliary adjustment strategy to the sleep adjustment intervention device through the signal interface, optimizing the control execution of the sleep adjustment intervention device to realize the dynamic auxiliary adjustment of the user sleep respiratory brain state.

22. The method of claim 21, wherein, The trend prediction at least includes any one of exponential smoothing method, Holt-Winters method, AR, MA, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA, VARMAX, machine learning.

23. The method of claim 21, wherein, The sleep respiration knowledge base is mainly from the knowledge and experience of sleep respiration related health management and clinical medicine, and at least includes sleep knowledge, sleep respiration rules, characteristics of common sleep respiration events, commonly used sleep respiration regulation methods and scene intervention parameter guidance; the user sleep respiration database at least includes the sleep respiration event, the first sleep respiration dynamics signal, the second sleep respiration dynamics signal, the sleep respiration state quantitative feature, the sleep brain state quantitative feature and the sleep respiration brain state auxiliary regulation strategy.

24. The method of any one of claims 21-23, wherein, The sleep respiration brain state auxiliary regulation strategy at least includes any one of respiratory frequency target value, respiratory depth target value, regulation mode, regulation time point, duration and device control parameter.

25. The method of any one of claims 21-23, wherein, The sleep regulation intervention device at least includes any one of ventilator, intelligent mattress, odor stimulation device, electric stimulation device, magnetic stimulation device, tactile stimulation device, environmental temperature and humidity regulation device and environmental CO2 concentration regulation device.

26. The method of claim 1, wherein, The specific steps of generating the output sleep respiration brain state joint detection evaluation and auxiliary regulation report according to the preset period or strategy, and updating the user sleep respiration database further include: generating or outputting the sleep respiration brain state joint detection evaluation and auxiliary regulation report according to the preset report period; updating the user sleep respiration database according to the preset data update strategy.

27. The method of claim 26, wherein, The sleep respiration brain state joint detection evaluation and auxiliary regulation report at least includes statistical analysis of the sleep respiration event, sleep respiration state evaluation summary, sleep brain state evaluation summary, auxiliary regulation evaluation summary, and sleep behavior improvement suggestion.

28. A sleep brain and respiration combined detection evaluation and auxiliary adjustment system, characterized in that, The following modules are included: The brain state detection module is used for collecting and processing sleep brain state signals of a user, and identifying sleep phase stages through feature analysis, to obtain sleep brain state quantitative features. The respiration event detection module is used for collecting, processing and dynamics analyzing sleep respiration behavior signals of a user, to obtain a first sleep respiration dynamics signal and perform event detection analysis, and to extract basic information of sleep respiration events. The respiration state quantification module is used for signal decomposition and / or time-frequency analysis on the first sleep respiration dynamics signal, to obtain a second sleep respiration dynamics signal, and to perform feature analysis on the two dynamics signals to generate sleep respiration state quantitative features. The intensity grade evaluation module is used for comparing and analyzing related feature changes of the sleep respiration event according to the sleep brain state quantitative features, the sleep respiration state quantitative features and the sleep phase stages, to identify and generate event intensity and event grade of the sleep respiration event. The sleep auxiliary regulation module is used for trend prediction on the first sleep respiration dynamics signal, the sleep respiration state quantitative features and the sleep brain state quantitative features, to generate a sleep respiration brain state auxiliary regulation strategy in combination with a sleep respiration knowledge base and a user sleep respiration database, and to send a sleep regulation intervention device through a signal interface. The report data management module is used for generating and outputting a sleep respiration brain state joint detection evaluation and auxiliary regulation report according to a preset period or strategy, and updating the user sleep respiration database. A data operation management module is configured to visually manage, uniformly store and operate all process data of the system. The intensity level evaluation module further comprises the following functional units: A feature change calculation unit is configured to compare and analyze the relative changes of the sleep brain state quantitative features and the sleep respiratory state quantitative features according to the event occurrence period and the event comparison interval of the sleep respiratory event, to obtain an event description feature change set; An event intensity quantification unit is configured to identify and generate the event intensity according to the sleep respiratory event, the event description feature change set and the sleep phase staging; An event level quantification unit is configured to identify and generate the event level according to the sleep respiratory event, the sleep body position feature and the sleep phase staging.

29. The system of claim 28, wherein, The brain state detection module further comprises the following functional units: A brain signal detection unit is configured to collect and acquire brain state signals in a sleep process of a user and perform signal processing to obtain the sleep brain state signals; A brain state feature analysis unit is configured to perform feature analysis on the sleep brain state signals to obtain the sleep brain state quantitative features; A sleep phase analysis unit is configured to identify the sleep phase staging according to the sleep brain state signals and generate a sleep phase curve.

30. The system of claim 28 or 29, wherein, The respiratory event detection module further comprises the following functional units: A respiratory signal detection unit is configured to collect and acquire respiratory behavior signals in a sleep process of a user and perform signal processing to obtain the sleep respiratory behavior signals; A dynamics analysis unit is configured to perform dynamics analysis on the sleep respiratory behavior signals to obtain the first sleep respiratory dynamics signals; An event detection analysis unit is configured to perform event detection analysis on the first sleep respiratory dynamics signals to identify and obtain the sleep respiratory event and extract event basic information.

31. The system of claim 30, wherein, The respiratory state quantification module further comprises the following functional units: A signal decomposition and extraction unit is configured to perform signal decomposition and / or time-frequency analysis on the first sleep respiratory dynamics signals to obtain the second sleep respiratory dynamics signals; A respiratory feature analysis unit is configured to perform feature analysis on the first sleep respiratory dynamics signals and the second sleep respiratory dynamics signals to obtain the sleep respiratory state quantitative features.

32. The system of any one of claims 28, 31, wherein, The sleep auxiliary adjustment module further comprises the following functional units: A respiratory dynamics prediction unit is configured to perform trend prediction on the first sleep respiratory dynamics signals to obtain first sleep respiratory dynamics prediction signals; A respiratory state prediction unit is configured to perform trend prediction on the sleep respiratory state quantitative features to obtain sleep respiratory state prediction features; A brain state prediction unit is configured to perform trend prediction on the sleep brain state quantitative features to obtain sleep brain state prediction features; A prediction description collection unit is configured to collect the first sleep respiratory dynamics prediction signals, the sleep respiratory state prediction features and the sleep brain state prediction features to obtain sleep respiratory and brain state prediction descriptions; A baseline description collection unit is configured to collect the first sleep respiratory dynamics signal, the sleep respiratory state quantification feature, and the sleep brain state quantification feature to obtain a sleep respiratory and brain state current description; An auxiliary strategy generation unit is configured to generate the sleep respiratory and brain state auxiliary adjustment strategy according to the sleep respiratory and brain state prediction description and the sleep respiratory and brain state current description, in combination with a sleep respiratory knowledge base and a user sleep respiratory database; A strategy auxiliary adjustment unit is configured to send the sleep respiratory and brain state auxiliary adjustment strategy to a sleep adjustment intervention device through a signal interface, and optimize control execution of the sleep adjustment intervention device to achieve dynamic auxiliary adjustment of the user sleep respiratory and brain state.

33. The system of any one of claims 28, 31, wherein, The report data management module further includes the following functional units: A report generation output unit is configured to generate or output the sleep respiratory and brain state joint detection evaluation and auxiliary adjustment report according to a preset report period; A database update unit is configured to update the user sleep respiratory database according to a preset data update strategy.

34. The system of claim 33, wherein, The data operation management module further includes the following functional units: A user information management unit is configured to register, input, edit, query, output, and delete user basic information; A data visualization management unit is configured to visually display all data in the system; A data operation management unit is configured to store, back up, migrate, and export all data in the system.

35. A sleep brain and respiration combined detection, evaluation and auxiliary adjustment device, characterized in that, The following modules are included: A brain state detection module is configured to collect and process sleep brain state signals of a user, perform feature analysis, identify sleep phase stages, and obtain sleep brain state quantification features; A respiratory event detection module is configured to collect, process, and perform dynamics analysis on sleep respiratory behavior signals of a user, obtain a first sleep respiratory dynamics signal, perform event detection analysis, and extract basic information of sleep respiratory events; A respiratory state quantification module is configured to perform signal decomposition and / or time-frequency analysis on the first sleep respiratory dynamics signal to obtain a second sleep respiratory dynamics signal, and perform feature analysis on the two dynamics signals to generate sleep respiratory state quantification features; An intensity level evaluation module is configured to compare and analyze changes in related features of the sleep respiratory events according to the sleep brain state quantification features, the sleep respiratory state quantification features, and the sleep phase stages, and identify and generate event intensity and event level of the sleep respiratory events; A sleep auxiliary adjustment module is configured to perform trend prediction on the first sleep respiratory dynamics signal, the sleep respiratory state quantification features, and the sleep brain state quantification features, in combination with a sleep respiratory knowledge base and a user sleep respiratory database, to generate a sleep respiratory and brain state auxiliary adjustment strategy and send the strategy to a sleep adjustment intervention device through a signal interface; A report data management module is configured to generate and output a sleep respiratory and brain state joint detection evaluation and auxiliary adjustment report according to a preset period or strategy, and update the user sleep respiratory database; A data visualization module is configured to uniformly visually display all process data and / or result data in the device. A data management center module is configured to uniformly store and manage data operation of all process data and / or result data in the device. The specific steps of identifying and generating the event intensity and the event level of the sleep respiratory event by comparing and analyzing the related feature changes of the sleep respiratory event based on the sleep brain state quantification feature, the sleep respiratory state quantification feature and the sleep phase staging further include: According to the sleep respiratory event occurrence period and the event comparison interval, the feature relative changes of the sleep brain state quantification feature and the sleep respiratory state quantification feature are compared and analyzed to obtain an event description feature change set; According to the sleep respiratory event, the event description feature change set and the sleep phase staging, the event intensity is identified and generated; According to the sleep respiratory event, the sleep body position feature and the sleep phase staging, the event level is identified and generated.

Images