Sleep Apnea Oxygen Detection, Analysis and Auxiliary Regulation Method, System and Device

Abstract

The present invention provides a method, system and device for detecting, analyzing, and assisting in adjusting sleep breathing and blood oxygen. By detecting and analyzing sleep breathing signals and sleep blood oxygen signals, respiratory events and blood oxygen events are identified, and sleep breathing and blood oxygen events and event intensities are extracted. Through the combined features of sleep breathing and blood oxygen, the event intensities are further corrected, realizing systematic detection, analysis, and quantitative evaluation of sleep breathing and blood oxygen events. Through the prediction of the current state and development of sleep breathing and blood oxygen events, a sleep breathing and blood oxygen assisted adjustment strategy is generated and sent to a sleep breathing adjustment device to achieve efficient dynamic assisted adjustment of the user's sleep breathing and blood oxygen. The present invention can achieve scientific detection and evaluation of sleep breathing and blood oxygen and efficient dynamic assisted adjustment.

Description

Technical Field

[0001] The present invention relates to the field of sleep apnea blood oxygen detection, analysis and auxiliary regulation, and particularly to a method, a system and a device for sleep apnea blood oxygen detection, analysis and auxiliary regulation. Background Art

[0002] Blood oxygen level is an important physiological indicator for the circulatory work of the respiratory and cardiovascular systems. The blood oxygen levels in different limb parts or organ regions are not only direct indicators of physiological metabolism levels, but also manifestations of the physiological capabilities of the respiratory and cardiovascular systems. Due to factors such as obesity, fatigue, aging, infection, and disorder of respiratory muscle tone control, etc., the normal maintenance of blood oxygen levels will be affected; especially the occurrence of severe and continuous blood oxygen level drops during sleep has a serious impact on people's sleep quality, physical health and life safety. During sleep, the sleep blood oxygen level (arterial oxygen saturation) of healthy people is maintained at 95% or above; while in different sleep phase states, people with different degrees of sleep apnea problems often have blood oxygen levels below 90%, 80% or even 65%, which has a serious impact on life health and sleep quality.

[0003] Currently, whether it is consumer electronics (such as smart bracelets and smart watches) or professional medical devices (such as pulse oximeters), most of them only complete the data collection of blood oxygen and simple numerical statistical analysis. The existing technical solution CN112971763A relates to a sleep apnea event detection device, including: a signal preprocessing module, a tidal volume calculation module, a threshold setting module, a respiratory event judgment module, and a calibration module; it preliminarily determines sleep apnea events based on the chest and abdominal respiratory signals of the subject, and then uses the blood oxygen saturation data and triaxial acceleration data corresponding to the chest and abdominal respiratory signals for calibration to determine sleep apnea events; the sleep apnea events are respiratory sleep apnea or hypopnea; among them, respiratory sleep apnea is obstructive sleep apnea, central sleep apnea, or mixed sleep apnea. This technical solution mainly relies on detecting and analyzing the respiratory signal waveform through threshold and counting methods to determine sleep apnea events. Generally speaking, the existing technical solutions for sleep apnea detection and evaluation mainly rely on sensors or devices such as oral-nasal pressure, oral-nasal thermosensitivity, and chest-abdominal belt pressure to directly detect and analyze, and complete respiratory event detection, event type classification, and simple statistical analysis of events through waveform analysis; the existing technical solutions lack systematic detection and analysis, quantitative evaluation, and identification and correction of sleep apnea events and blood oxygen events, especially lack of a clear definition or feature quantification of the intensity of sleep apnea blood oxygen events from the perspective of respiratory dynamics and sleep phase state. In addition, currently, sleep apnea regulation devices are usually independently separated from sleep apnea detection devices. Most sleep apnea regulation devices can connect to the network to feedback regulation parameters but still use an offline preset program to control the device feedback regulation, and cannot perform intelligent, precise, and personalized dynamic auxiliary adjustment according to the user's real-time state.

[0004] As can be seen from the above, how to systematically detect and analyze and quantitatively evaluate sleep apnea events and blood oxygen events, how to optimize and improve the efficiency and effectiveness of existing sleep apnea regulation devices, so as to ensure the normal sleep blood oxygen level of users, assist users in sleeping and improve sleep quality, are problems that need to be further solved in the current product technical solutions and actual application scenarios at home and abroad. Summary of the Invention

[0005] In view of the above defects and improvement requirements of the existing methods, the purpose of the present invention is to provide a method for detecting, analyzing, and assisting in adjusting sleep respiration and blood oxygen. By detecting and analyzing sleep respiration signals and sleep blood oxygen signals, identifying respiration events and blood oxygen events, and extracting sleep respiration and blood oxygen events and event intensities, and further correcting the event intensities through the combined features of sleep respiration and blood oxygen, the systematic detection, analysis, and quantitative evaluation of sleep respiration and blood oxygen events are realized; through the prediction of the current state and development of sleep respiration and blood oxygen events, a sleep respiration and blood oxygen assisted adjustment strategy is generated and sent to a sleep respiration adjustment device to achieve efficient dynamic assisted adjustment of the user's sleep respiration and blood oxygen. The present invention also provides a system for detecting, analyzing, and assisting in adjusting sleep respiration and blood oxygen for implementing the above method. The present invention also provides a device for detecting, analyzing, and assisting in adjusting sleep respiration and blood oxygen for implementing the above system.

[0006] According to the purpose of the present invention, the present invention proposes a method for detecting, analyzing, and assisting in adjusting sleep respiration and blood oxygen, including the following steps:

[0007] Collect and process the sleep respiration signal and sleep blood oxygen signal during the user's sleep process to obtain a combined sleep respiration and blood oxygen signal;

[0008] Perform event detection and analysis on the combined sleep respiration and blood oxygen signal, identify respiration events and blood oxygen events, and extract sleep respiration and blood oxygen events and event intensities;

[0009] Perform feature analysis on the combined sleep respiration and blood oxygen signal to obtain combined sleep respiration and blood oxygen features and compare the relative changes in features during the occurrence of the sleep respiration and blood oxygen events, and correct the event intensities;

[0010] Perform prediction analysis on the combined sleep respiration and blood oxygen signal and the combined sleep respiration and blood oxygen features, identify sleep respiration and blood oxygen prediction events and event occurrence trends, combine a sleep respiration knowledge base and a user sleep respiration database, generate a sleep respiration and blood oxygen assisted adjustment strategy, and send it to a sleep respiration adjustment device through a signal interface;

[0011] Generate and output a sleep respiration and blood oxygen evaluation and assisted adjustment report according to a preset cycle or strategy, and update the user sleep respiration database.

[0012] More preferably, the specific steps of collecting and processing the sleep respiration signal and sleep blood oxygen signal during the user's sleep process to obtain a combined sleep respiration and blood oxygen signal further include:

[0013] Collect and monitor the user's sleep respiration behavior and perform signal processing to obtain the sleep respiration signal;

[0014] Collect and monitor the user's sleep blood oxygen level and perform signal processing to obtain the sleep blood oxygen signal;

[0015] Aggregate the sleep respiratory signal and the sleep blood oxygen signal to obtain the combined sleep respiratory and blood oxygen signal.

[0016] Preferably, the acquisition and processing at least include acquisition and obtaining, analog-to-digital conversion, resampling, rereferencing, artifact removal, noise reduction, power frequency notch filtering, bandpass filtering, mean filtering, smoothing processing, and signal time window segmentation.

[0017] Preferably, the combined sleep respiratory and blood oxygen signal at least includes the sleep respiratory signal and the sleep blood oxygen signal; wherein, the sleep respiratory signal at least includes any one of an oral and nasal temperature monitoring signal, a nasal pressure monitoring signal, an oral and nasal CO2 monitoring signal, a thoracic and abdominal respiratory movement signal, an electrocardiogram-derived respiratory signal, and a pharyngeal electromyogram signal, and the sleep blood oxygen signal at least includes any one of a photoplethysmography blood oxygen signal and a blood oxygenation level-dependent signal.

[0018] Preferably, the specific steps of performing event detection and analysis on the combined sleep respiratory and blood oxygen signal, identifying respiratory events and blood oxygen events, and extracting sleep respiratory and blood oxygen events and event intensities further include:

[0019] Perform event detection and analysis on the sleep respiratory signal, identify respiratory events, and obtain sleep respiratory events;

[0020] Perform event detection and analysis on the sleep blood oxygen signal, identify blood oxygen events, and obtain sleep blood oxygen events;

[0021] Calculate the event intensity according to the sleep respiratory events and the sleep respiratory signal, and the sleep blood oxygen events and the sleep blood oxygen signal;

[0022] Aggregate the sleep respiratory events, the sleep blood oxygen events, and the event intensities in the order of event occurrence time to obtain the sleep respiratory and blood oxygen events.

[0023] Preferably, the event detection and analysis specifically involve performing waveform feature recognition and marker extraction on the sleep respiratory signal or the sleep blood oxygen signal according to a preset sleep respiratory-blood oxygen event knowledge base and / or a machine learning model to obtain the basic event information of the sleep respiratory event or the sleep blood oxygen event; the basic event information at least includes event type, start time, end time, duration, peak and valley values, and time at peak and valley values.

[0024] Preferably, the sleep respiratory and blood oxygen events are specifically composed of the sleep respiratory events and the sleep blood oxygen events, and at least include event type, start time, end time, duration, peak and valley values, time at peak and valley values, and event intensity.

[0025] Preferably, the event intensity is specifically determined by the duration, start time, end time, peak-valley value, and relative time at the peak-valley value of the sleep breathing event and the sleep blood oxygen event, as well as the amplitude change of the sleep breathing blood oxygen characteristic and the sleep phase.

[0026] Preferably, the preliminary generation method of the event intensity is specifically as follows:

[0027] 1) Obtain the start time and end time of the sleep breathing event and the sleep blood oxygen event respectively, calculate the ratio of the time difference between the start times of the two events and the time difference between the end times of the two events to obtain the respiratory blood oxygen event delay ratio;

[0028] 2) Obtain the start time, end time, duration, peak-valley value, and time at the peak-valley value of the sleep blood oxygen event, as well as a preset sleep blood oxygen signal threshold;

[0029] 3) Calculate the relative change amount between the peak-valley value of the sleep blood oxygen event and the preset sleep blood oxygen signal threshold to obtain the peak-valley relative value;

[0030] 4) Calculate the linear slope according to the start time, peak-valley value, and time at the peak-valley value of the sleep blood oxygen event to obtain the peak-valley front slope;

[0031] 5) Calculate the linear slope according to the end time, peak-valley value, and time at the peak-valley value of the sleep blood oxygen event to obtain the peak-valley rear slope;

[0032] 6) Calculate the event intensity through numerical fusion of the duration of the sleep breathing event, the duration of the sleep blood oxygen event, the respiratory blood oxygen event delay ratio, the peak-valley relative value, the peak-valley front slope, and the peak-valley rear slope.

[0033] Preferably, the specific steps for performing feature analysis on the sleep breathing blood oxygen combined signal to obtain the sleep breathing blood oxygen combined feature and comparing the relative change in features during the occurrence of the sleep breathing blood oxygen event to correct the event intensity further include:

[0034] Perform feature analysis on the sleep breathing blood oxygen combined signal to obtain the sleep breathing blood oxygen combined feature;

[0035] Identify the sleep phase according to the sleep breathing blood oxygen combined feature and generate a sleep phase curve;

[0036] Compare the relative change in features of the sleep breathing blood oxygen combined feature according to the occurrence period of the sleep breathing blood oxygen event to obtain the change amount of the sleep breathing blood oxygen combined event feature;

[0037] Correct the event intensity according to the sleep phase and the change amount of the sleep breathing blood oxygen combined event feature.

[0038] Preferably, the feature analysis at least includes numerical feature analysis, envelope feature analysis, time-frequency feature analysis, and non-linear feature analysis; wherein, the numerical features at least include mean value, root mean square, maximum value, minimum value, variance, standard deviation, coefficient of variation, kurtosis, and skewness, the time-frequency features at least include band power, band power ratio, and band center frequency, and the non-linear features at least include entropy features, fractal features, and complexity features.

[0039] Preferably, the combined sleep apnea and blood oxygen features at least include sleep apnea features and sleep blood oxygen features; the sleep apnea features at least include respiratory rate, numerical features, and time-frequency features of the sleep apnea signal; the sleep blood oxygen features at least include any one of pulse features, PPG blood oxygen level features, and BOLD blood oxygen level features, and the BOLD blood oxygen level features at least include oxygenated hemoglobin concentration features and deoxygenated hemoglobin concentration features.

[0040] Preferably, the sleep stages at least include wakefulness, light sleep, deep sleep, and rapid eye movement sleep; the method for generating the sleep stages and the sleep stage curve is specifically as follows:

[0041] 1) Through machine learning, the combined sleep apnea and blood oxygen features of a large-scale sleep user sample and their corresponding sleep stage data are learned, trained, and data modeled to obtain a sleep stage classification model;

[0042] 2) The combined sleep apnea and blood oxygen features of the current user are input into the sleep stage classification model to obtain the corresponding sleep stage classification;

[0043] 3) According to the time sequence, the numerical values of the sleep stages of all signal time windows are extracted to obtain the sleep stage curve.

[0044] Preferably, the method for correcting the event intensity is specifically as follows:

[0045] 1) Obtain the sleep stage, and extract the sleep stage correction coefficient according to the preset sleep stage - correction coefficient comparison table;

[0046] 2) Obtain the change amount of the combined sleep apnea and blood oxygen event features, determine and select the change amount of the target feature, and through numerical weighting calculation, obtain the relative change coefficient of the event feature;

[0047] 3) Use the numerical product of the relative change coefficient of the event feature and the sleep stage correction coefficient to correct the initially generated event intensity to obtain the corrected event intensity.

[0048] Preferably, the specific steps of predicting and analyzing the sleep breathing blood oxygen combined signal and the sleep breathing blood oxygen combined feature, identifying the sleep breathing blood oxygen prediction event and the event occurrence trend, combining the sleep breathing knowledge base and the user sleep breathing database, generating the sleep breathing blood oxygen auxiliary regulation strategy and sending it to the sleep breathing regulation device through the signal interface further include:

[0049] Predict and analyze the sleep breathing blood oxygen combined signal and the sleep breathing blood oxygen combined feature to generate a sleep breathing blood oxygen combined prediction signal and a sleep breathing blood oxygen combined prediction feature respectively;

[0050] Conduct event detection and analysis on the sleep breathing blood oxygen combined prediction signal to identify the sleep breathing blood oxygen event, and obtain the sleep breathing blood oxygen prediction event and the event prediction intensity;

[0051] According to the sleep breathing blood oxygen prediction event and the sleep breathing blood oxygen combined prediction feature, correct the event prediction intensity and identify the event occurrence trend;

[0052] According to the sleep breathing blood oxygen prediction event and the event occurrence trend, combine the sleep breathing knowledge base and the user sleep breathing database to generate the sleep breathing blood oxygen auxiliary regulation strategy;

[0053] Send the sleep breathing blood oxygen auxiliary regulation strategy to the sleep breathing regulation device through the signal interface to optimize the operation control of the sleep breathing regulation device.

[0054] Preferably, the method of the prediction analysis includes at least one of exponential smoothing method, Holt-Winters method, AR, MA, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA, VARMAX, and machine learning.

[0055] Preferably, the event occurrence trend includes at least the event occurrence type, the event occurrence probability, and the event occurrence intensity; the event occurrence trend is obtained by machine learning through learning and training, data modeling, and analysis and calculation on the sleep breathing blood oxygen combined signal and the sleep breathing blood oxygen combined feature related to the sleep breathing event and the sleep blood oxygen event in a large-scale sleep user sample data.

[0056] Preferably, the sleep apnea knowledge base mainly comes from the knowledge and experience of sleep apnea-related health management and clinical medicine, and at least includes sleep apnea-blood oxygen laws, characteristics of common sleep apnea-blood oxygen events, common sleep apnea regulation methods, and scene intervention parameter guidance; the user sleep apnea database at least includes the sleep apnea-blood oxygen events, the sleep apnea-blood oxygen combined signals, the sleep apnea-blood oxygen combined characteristics, the event occurrence trend, and the sleep apnea-blood oxygen auxiliary regulation strategy.

[0057] Preferably, the sleep apnea-blood oxygen auxiliary regulation strategy at least includes blood oxygen target value, breathing frequency target value, breathing depth target value, regulation method, regulation time point, duration, and device control parameters.

[0058] Preferably, the sleep apnea regulation device at least includes any one of a ventilator, an odor stimulation device, an electrical stimulation device, a tactile stimulation device, an environmental temperature and humidity regulation device, and an environmental CO2 concentration regulation device.

[0059] Preferably, the specific steps of generating and outputting a sleep apnea-blood oxygen evaluation and auxiliary regulation report according to a preset cycle or strategy and updating the user sleep apnea database further include:

[0060] Generating the sleep apnea-blood oxygen evaluation and auxiliary regulation report according to a preset report cycle;

[0061] Outputting the sleep apnea-blood oxygen evaluation and auxiliary regulation report according to the user's scene requirements;

[0062] Updating the user sleep apnea database according to a preset data update strategy.

[0063] Preferably, the sleep apnea-blood oxygen evaluation and auxiliary regulation report at least includes the sleep blood oxygen signal, the statistical analysis of the sleep apnea-blood oxygen events, the curve of key characteristic indexes in the sleep apnea-blood oxygen combined characteristics, the sleep phase curve, detection quantification and auxiliary regulation summary, and sleep apnea optimization suggestions.

[0064] According to the purpose of the present invention, the present invention provides a sleep apnea-blood oxygen detection, analysis and auxiliary regulation system, including the following modules:

[0065] A respiratory blood oxygen detection module, configured to collect and process the sleep apnea signal and the sleep blood oxygen signal during the user's sleep process to obtain a sleep apnea-blood oxygen combined signal;

[0066] An event detection and analysis module, configured to perform event detection and analysis on the sleep apnea-blood oxygen combined signal, identify respiratory events and blood oxygen events, and extract sleep apnea-blood oxygen events and event intensities;

[0067] An event intensity adjustment module for performing feature analysis on the combined sleep apnea and blood oxygen signal to obtain combined sleep apnea and blood oxygen features and comparing the relative changes in features during the occurrence of the sleep apnea and blood oxygen event to correct the event intensity;

[0068] A respiratory blood oxygen regulation module for performing predictive analysis on the combined sleep apnea and blood oxygen signal and the combined sleep apnea and blood oxygen features, identifying sleep apnea and blood oxygen prediction events and event occurrence trends, combining a sleep apnea knowledge base and a user's sleep apnea database, generating a sleep apnea and blood oxygen assisted regulation strategy and sending it to a sleep apnea regulation device through a signal interface;

[0069] A user report management module for generating and outputting a sleep apnea and blood oxygen evaluation and assisted regulation report according to a preset cycle or strategy, and updating the user's sleep apnea database;

[0070] A data operation management module for visual management, unified storage, and operation management of all process data of the system.

[0071] Preferably, the respiratory blood oxygen detection module further includes the following functional units:

[0072] A sleep apnea detection unit for collecting and monitoring a user's sleep apnea behavior and performing signal processing to obtain the sleep apnea signal;

[0073] A sleep blood oxygen detection unit for collecting and monitoring a user's sleep blood oxygen level and performing signal processing to obtain the sleep blood oxygen signal;

[0074] A combined signal collection unit for collecting the sleep apnea signal and the sleep blood oxygen signal to obtain the combined sleep apnea and blood oxygen signal.

[0075] Preferably, the event detection and analysis module further includes the following functional units:

[0076] A respiratory event recognition unit for performing event detection and analysis on the sleep apnea signal to identify respiratory events and obtain sleep apnea events;

[0077] A blood oxygen event recognition unit for performing event detection and analysis on the sleep blood oxygen signal to identify blood oxygen events and obtain sleep blood oxygen events;

[0078] An event intensity calculation unit for calculating the event intensity according to the sleep apnea event and the sleep apnea signal, the sleep blood oxygen event and the sleep blood oxygen signal;

[0079] An event information collection unit for collecting the sleep apnea event, the sleep blood oxygen event, and the event intensity in chronological order of event occurrence to obtain the sleep apnea and blood oxygen event.

[0080] Preferably, the event intensity adjustment module further includes the following functional units:

[0081] The combined feature analysis unit is used to perform feature analysis on the sleep apnea blood oxygen combined signal to obtain the sleep apnea blood oxygen combined feature;

[0082] The sleep phase recognition unit is used to recognize the sleep phase according to the sleep apnea blood oxygen combined feature and generate a sleep phase curve;

[0083] The feature change analysis unit is used to compare the relative change of the features of the sleep apnea blood oxygen combined feature during the occurrence of the sleep apnea blood oxygen event to obtain the feature change amount of the sleep apnea blood oxygen combined event;

[0084] The event intensity correction unit is used to correct the event intensity according to the sleep phase and the feature change amount of the sleep apnea blood oxygen combined event.

[0085] Preferably, the breathing blood oxygen regulation module further includes the following functional units:

[0086] The trend prediction analysis unit is used to perform prediction analysis on the sleep apnea blood oxygen combined signal and the sleep apnea blood oxygen combined feature, and generate a sleep apnea blood oxygen combined prediction signal and a sleep apnea blood oxygen combined prediction feature respectively;

[0087] The event detection analysis unit is used to perform event detection analysis on the sleep apnea blood oxygen combined prediction signal to identify the sleep apnea blood oxygen event, and obtain the sleep apnea blood oxygen prediction event and the event prediction intensity;

[0088] The event occurrence prediction unit is used to correct the event prediction intensity according to the sleep apnea blood oxygen prediction event and the sleep apnea blood oxygen combined prediction feature, and identify the event occurrence trend;

[0089] The auxiliary strategy generation unit is used to generate the sleep apnea blood oxygen auxiliary regulation strategy according to the sleep apnea blood oxygen prediction event and the event occurrence trend, in combination with the sleep apnea knowledge base and the user's sleep apnea database;

[0090] The strategy sending control unit is used to send the sleep apnea blood oxygen auxiliary regulation strategy to the sleep apnea regulation device through a signal interface to optimize the operation control of the sleep apnea regulation device.

[0091] Preferably, the user report management module further includes the following functional units:

[0092] The user report generation unit is used to generate the sleep apnea blood oxygen evaluation and auxiliary regulation report according to a preset report period;

[0093] A user report output unit for outputting the sleep apnea blood oxygen evaluation and auxiliary adjustment report according to user scenario requirements;

[0094] A database update unit for updating the user sleep apnea database according to a preset data update strategy.

[0095] Preferably, the data operation management module further includes the following functional units:

[0096] A user information management unit for registering, inputting, editing, querying, outputting, and deleting user information;

[0097] A data visualization management unit for visual display management of all data in the system;

[0098] A data storage management unit for unified storage management of all data in the system;

[0099] A data operation management unit for backing up, migrating, and exporting all data in the system.

[0100] According to the object of the present invention, the present invention provides a sleep apnea blood oxygen detection, analysis and auxiliary adjustment device, including the following modules:

[0101] A respiratory blood oxygen detection module for collecting and processing sleep apnea signals and sleep blood oxygen signals during a user's sleep to obtain a combined sleep apnea blood oxygen signal;

[0102] An event detection and analysis module for performing event detection and analysis on the combined sleep apnea blood oxygen signal, identifying respiratory events and blood oxygen events, and extracting sleep apnea blood oxygen events and event intensities;

[0103] An event intensity adjustment module for performing feature analysis on the combined sleep apnea blood oxygen signal to obtain combined sleep apnea blood oxygen features and comparing the relative changes in features during the occurrence of the sleep apnea blood oxygen events to correct the event intensity;

[0104] A respiratory blood oxygen adjustment module for performing predictive analysis on the combined sleep apnea blood oxygen signal and the combined sleep apnea blood oxygen features, identifying sleep apnea blood oxygen prediction events and event occurrence trends, combining a sleep apnea knowledge base and a user sleep apnea database, generating a sleep apnea blood oxygen auxiliary adjustment strategy, and sending it to a sleep apnea adjustment device through a signal interface;

[0105] A user report management module for generating and outputting a sleep apnea blood oxygen evaluation and auxiliary adjustment report according to a preset period or strategy, and updating the user sleep apnea database;

[0106] A data visualization module, used for unified visualization and management of all process data and / or result data in the device;

[0107] 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.

[0108] The present invention provides a method, system and device for sleep breathing blood oxygen detection and analysis and auxiliary regulation, and innovatively proposes to identify sleep breathing blood oxygen events and accurately quantify and correct the event intensity through the collection and processing of sleep blood oxygen signals and sleep breathing signals, event detection and analysis, feature analysis, feature comparison and correction calculation. Based on the current state of the sleep breathing blood oxygen event, the development trend of the event (event type, event probability and event intensity) is predicted, and the real-time generation and interface sending of the sleep breathing auxiliary regulation strategy are completed, thereby optimizing and improving the efficiency and effectiveness of the sleep breathing control equipment, thereby realizing the integration of scientific detection and evaluation of the user's sleep breathing blood oxygen and dynamic auxiliary regulation, and efficiently assisting the user to sleep.

[0109] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0110] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0111] Figure 1 It is a schematic diagram of the process steps of a sleep breathing blood oxygen detection analysis and auxiliary adjustment method provided by an embodiment of the present invention;

[0112] Figure 2 This is a schematic diagram of the module composition of a sleep breathing and blood oxygen detection and analysis and auxiliary regulation system provided by an embodiment of the present invention;

[0113] Figure 3 It is a schematic diagram of the module structure of a sleep breathing blood oxygen detection analysis and auxiliary adjustment device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0114] To more clearly illustrate the purpose and technical solutions of the present invention, the present invention will be further introduced below in conjunction with the accompanying drawings in the embodiments of the present invention application. Obviously, the embodiments described below are only a part of the embodiments of the present invention, rather than all embodiments. Without creative labor, other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention should fall within the protection scope of the present invention. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be arbitrarily combined with each other.

[0115] The applicant has found that there are significant differences and completely different characteristic laws in human breathing behavior during sleep and wakefulness: First, as one of the direct results of breathing behavior, the oscillation level and change law of sleep blood oxygen level are almost all brought by sleep breathing events, and the sleep blood oxygen level curve has good characteristics for sleep breathing dynamics analysis; Second, the sleep breathing behavior during sleep is mainly determined by the brainstem respiratory center, the muscle function and nerve function of the respiratory organ link, etc., and the breathing regulation process during sleep is more dependent on the underlying biological and physiological functions; Finally, the blood oxygen saturation level of the limbs and organs is the direct result of sleep breathing behavior and the blood supply and oxygen supply of the physiological link, is the most direct and obvious manifestation of sleep breathing behavior, and is particularly sensitive to various sleep blood oxygen events.

[0116] Therefore, the present invention will be based on sleep breathing signals and sleep blood oxygen signals to scientifically and comprehensively detect, quantify and evaluate sleep breathing blood oxygen events, further complete the predictive analysis of sleep breathing blood oxygen events and the real-time generation of sleep breathing regulation strategies, so as to realize the integration of scientific detection and evaluation of user sleep breathing and dynamic auxiliary regulation, and efficiently assist the user to sleep.

[0117] Combined with Figure 1 As shown, a sleep breathing blood oxygen detection, analysis and auxiliary regulation method provided by an embodiment of the present invention includes the following steps:

[0118] P100: Collect and process the sleep breathing signal and sleep blood oxygen signal during the user's sleep process to obtain a sleep breathing blood oxygen combined signal.

[0119] In the embodiment, the collection and processing at least include collection and acquisition, analog-to-digital conversion, resampling, rereferencing, artifact removal, noise reduction, power frequency notch filtering, band-pass filtering, mean filtering, smoothing processing and signal time window segmentation.

[0120] In this embodiment, the sleep apnea and blood oxygen combined signal at least includes a sleep apnea signal and a sleep blood oxygen signal; wherein, the sleep apnea signal at least includes any one of an oral and nasal temperature monitoring signal, a nasal pressure monitoring signal, an oral and nasal CO2 monitoring signal, a thoracic and abdominal respiration movement signal, an electrocardiogram-derived respiration signal, and a pharyngeal electromyogram signal, and the sleep blood oxygen signal at least includes any one of a photoplethysmography blood oxygen signal and a blood oxygenation level-dependent signal.

[0121] Step 1: Collect and monitor the sleep apnea behavior of the user and perform signal processing to obtain a sleep apnea signal.

[0122] In this embodiment, the thoracic and abdominal respiration movement signal is used as the sleep apnea signal to describe the specific implementation process of the present invention. The sleep apnea movement of the user's chest and abdomen is collected by an impedance sensor type thoracic and abdominal belt, and the sampling rate is 128 Hz; the thoracic and abdominal belt is divided into a chest belt and an abdominal belt, and only the abdominal belt signal is collected in this embodiment. The signal processing of the thoracic and abdominal respiration movement signal mainly includes artifact removal, wavelet denoising, DC removal high-pass filtering, and signal time window segmentation.

[0123] Step 2: Collect and monitor the sleep blood oxygen level of the user and perform signal processing to obtain a sleep blood oxygen signal.

[0124] In this embodiment, the photoplethysmography blood oxygen signal is used as the sleep blood oxygen signal to describe the specific implementation method of the present invention. The fingertip blood oxygen signal of the user's left index finger is collected by a fingertip oximeter, and the sampling rate is 64 Hz. The signal processing of the sleep blood oxygen signal mainly includes resampling (128 Hz), artifact removal, wavelet denoising, DC removal high-pass filtering, and signal time window segmentation.

[0125] In the actual use scenario, various medical fingertip oximeters or mid- to high-end health monitoring function watches can obtain high-quality blood oxygen signal records, and they are all PPG photoplethysmography blood oxygen signals; while the BOLD blood oxygenation level-dependent signal, although the device use and detection process are slightly complicated, can obtain more accurate blood oxygen level state data descriptions; it is necessary to select the PPG photoplethysmography blood oxygen signal or the BOLD blood oxygenation level-dependent signal according to the actual scenario requirements.

[0126] Step 3: Aggregate the sleep apnea signal and the sleep blood oxygen signal to obtain a sleep apnea and blood oxygen combined signal.

[0127] In this embodiment, the sleep apnea and blood oxygen combined signal includes a thoracic and abdominal respiration movement signal and a fingertip blood oxygen signal.

[0128] P200: Perform event detection and analysis on the sleep apnea and blood oxygen combined signal, identify respiration events and blood oxygen events, and extract sleep apnea and blood oxygen events and event intensities.

[0129] In this embodiment, the event detection and analysis specifically refers to identifying waveform features and extracting markers from the sleep respiration signal or sleep blood oxygen signal according to a preset sleep respiration - blood oxygen event knowledge base and / or a machine learning model, so as to obtain the basic event information of the sleep respiration event or sleep blood oxygen event. The basic event information includes at least the event type, start time, end time, duration, peak - valley value, and time at the peak - valley value.

[0130] First step: Perform event detection and analysis on the sleep respiration signal to identify respiration events and obtain sleep respiration events.

[0131] In this embodiment, machine learning is used to perform learning training and data modeling on the scale sample data of the sleep respiration signal and sleep respiration events to construct a sleep respiration event recognition model; after the sleep respiration signal passes through the sleep respiration event recognition model and event information extraction, sleep respiration events are obtained.

[0132] Second step: Perform event detection and analysis on the sleep blood oxygen signal to identify blood oxygen events and obtain sleep blood oxygen events.

[0133] In this embodiment, machine learning is used to perform learning training and data modeling on the scale sample data of the sleep blood oxygen signal and sleep blood oxygen events to construct a sleep blood oxygen event recognition model; after the sleep blood oxygen signal passes through the sleep blood oxygen event recognition model and event information extraction, sleep respiration events are obtained.

[0134] Third step: Calculate the event intensity based on the sleep respiration events and sleep respiration signals, and sleep blood oxygen events and sleep blood oxygen signals.

[0135] In this embodiment, the event intensity is specifically determined by the duration, start time, end time, peak - valley value, and relative time at the peak - valley value of the sleep respiration event and sleep blood oxygen event, as well as the change in the amplitude of the sleep respiration - blood oxygen characteristics and the sleep phase. In the actual application scenario, attention should be paid to the paired relationship between the sleep respiration event and the sleep blood oxygen event, where the sleep respiration event occurs first and is immediately followed by the sleep blood oxygen event.

[0136] In this embodiment, the preliminary generation method of the event intensity is specifically as follows:

[0137] 1) Obtain the start time and end time of the sleep respiration event and sleep blood oxygen event respectively, and calculate the ratio of the time difference between the start times of the two events and the time difference between the end times of the two events to obtain the respiration - blood oxygen event delay ratio;

[0138] 2) Obtain the start time, end time, duration, peak - valley value, and time at the peak - valley value of the sleep blood oxygen event, as well as a preset sleep blood oxygen signal threshold;

[0139] 3) Calculate the peak-to-valley value of the sleep blood oxygen event and the relative change amount of the preset sleep blood oxygen signal threshold to obtain the peak-to-valley relative value;

[0140] 4) Calculate the linear slope based on the start time, peak-to-valley value, and the time at the peak-to-valley value of the sleep blood oxygen event to obtain the peak-to-valley leading edge slope;

[0141] 5) Calculate the linear slope based on the end time, peak-to-valley value, and the time at the peak-to-valley value of the sleep blood oxygen event to obtain the peak-to-valley trailing edge slope;

[0142] 6) Calculate through numerical fusion of the duration of the sleep apnea event, the duration of the sleep blood oxygen event, the respiratory blood oxygen event delay ratio, the peak-to-valley relative value, the peak-to-valley leading edge slope, and the peak-to-valley trailing edge slope,

[0143] to obtain the event intensity.

[0144] In this embodiment, the respiratory blood oxygen event delay ratio reflects the quantification of the basic physiological function state and its time delay characteristics. The peak-to-valley leading edge slope and the peak-to-valley trailing edge slope in the blood oxygen event represent the physiological obstruction rate and physiological recovery ability, which can be corrected with different weight coefficients in the observation and analysis to adapt to different analysis objectives. In addition, the duration and the peak-to-valley relative value (blood oxygen drop amount) reflect the most direct severity of the ischemic event. The longer the duration and the larger the peak-to-valley relative value (blood oxygen drop amount), the greater the damage to the human body and the brain. It is necessary to adjust the numerical fusion calculation method of the duration, the peak-to-valley relative value, the peak-to-valley leading edge slope, and the peak-to-valley trailing edge slope according to the actual analysis task to obtain a more accurately characterized event intensity. In the actual application scenario, select the numerical combination calculation method of the event intensity according to the actual situation of different user scenarios. Two calculation formulas are provided in this embodiment to illustrate the construction method.

[0145] The first calculation formula method of the event intensity is specifically as follows:

[0146]

[0147] The second calculation formula method of the event intensity is specifically as follows:

[0148]

[0149] Among them, eti is the event intensity, bTdurt and oTdurt are the duration of the sleep apnea event and the duration of the sleep blood oxygen event respectively, iTdelay is the respiratory blood oxygen event delay ratio, Pv is the peak-to-valley relative value, LSp is the peak-to-valley leading edge slope, RSp is the peak-to-valley trailing edge slope, || || is the absolute value operation, and min() is the minimum value operation.

[0150] In this embodiment, the preset sleep blood oxygen signal threshold is set to 95. In the actual use scenario, the blood oxygen level of a normal person during normal sleep remains above 95, so the preset sleep blood oxygen signal threshold is set to 95. In addition, different levels such as 100 - 95, 95 - 90, 90 - 85, 85 - 80, 80 - 75, 75 - 70, 70 - 65, 65 - 60, 60 - 50, 50 - below, etc. of the blood oxygen level also illustrate an important level consideration of the sleep blood oxygen event intensity. It is important to track and analyze the deviation of the user's blood oxygen level from the preset sleep blood oxygen signal threshold of 95.

[0151] Step 4: Aggregate the sleep apnea events, sleep blood oxygen events, and event intensities according to the event occurrence time sequence to obtain sleep apnea blood oxygen events.

[0152] In this embodiment, the sleep apnea blood oxygen event is specifically composed of sleep apnea events and sleep blood oxygen events, and at least includes event type, start time, end time, duration, peak valley value, time at the peak valley value, and event intensity.

[0153] P300: Perform feature analysis on the sleep apnea blood oxygen combined signal to obtain sleep apnea blood oxygen combined features and compare the relative changes in features during the occurrence of the sleep apnea blood oxygen event, and correct the event intensity.

[0154] In this embodiment, the feature analysis at least includes numerical feature analysis, envelope feature analysis, time - frequency feature analysis, and non - linear feature analysis; among them, the numerical features at least include mean value, root mean square, maximum value, minimum value, variance, standard deviation, coefficient of variation, kurtosis, and skewness, the time - frequency features at least include band power, band power ratio, and band center frequency, and the non - linear features at least include entropy features, fractal features, and complexity features.

[0155] Step 1: Perform feature analysis on the sleep apnea blood oxygen combined signal to obtain sleep apnea blood oxygen combined features.

[0156] In this embodiment, the sleep apnea blood oxygen combined features at least include sleep apnea features and sleep blood oxygen features; the sleep apnea features at least include respiratory rate, numerical features, and time - frequency features of the sleep apnea signal; the sleep blood oxygen features at least include any one of pulse features, PPG blood oxygen level features, and BOLD blood oxygen level features, and the BOLD blood oxygen level features at least include oxygenated hemoglobin concentration features and deoxygenated hemoglobin concentration features.

[0157] In this embodiment, the respiratory rate and the minimum blood oxygen value are used as the sleep apnea blood oxygen combined features.

[0158] Step 2: Identify the sleep phase according to the sleep apnea blood oxygen combined features and generate a sleep phase curve.

[0159] In this embodiment, the sleep phases at least include the wake period, light sleep period, deep sleep period, and rapid eye movement (REM) sleep period; the method for generating the sleep phases and the sleep phase curve is specifically as follows:

[0160] 1) Through machine learning, learn and train the combined sleep respiration and blood oxygen characteristics of a large-scale sleep user sample and their corresponding sleep stage data, and perform data modeling to obtain a sleep phase staging model;

[0161] 2) Input the combined sleep respiration and blood oxygen characteristics of the current user into the sleep phase staging model to obtain the corresponding sleep phase staging;

[0162] 3) Extract the numerical values of the sleep phases of all signal time windows in chronological order to obtain a sleep phase curve.

[0163] Step 3: According to the occurrence period of the sleep respiration and blood oxygen event, compare the relative changes in the characteristics of the combined sleep respiration and blood oxygen characteristics to obtain the change amount of the combined sleep respiration and blood oxygen event characteristics.

[0164] In this embodiment, according to the start time and end time in the sleep respiration and blood oxygen event, determine the occurrence period of the current sleep respiration and blood oxygen event, and select the end time of the previous event to the start time of the current event as the comparison interval. Compare and analyze the relative changes in the characteristics of the respiratory rate and the minimum blood oxygen value during the occurrence period of the sleep respiration and blood oxygen event and the comparison interval to obtain the change amount of the combined sleep respiration and blood oxygen event characteristics.

[0165] In this embodiment, the calculation formula for the relative change in characteristics is:

[0166]

[0167] where iFea is the relative change in characteristics, FE i and FB i are the characteristic values during the occurrence period and the comparison interval respectively, and || || is the absolute value calculation.

[0168] Step 4: Correct the event intensity according to the sleep phases and the change amount of the combined sleep respiration and blood oxygen event characteristics.

[0169] In this embodiment, the method for correcting the event intensity is specifically as follows:

[0170] 1) Obtain the sleep phases, and extract the sleep phase correction coefficients according to the preset sleep phase - correction coefficient comparison table;

[0171] 2) Obtain the change amount of the combined sleep respiration and blood oxygen event characteristics, determine and select the change amount of the characteristics of the target feature, and calculate through numerical weighting to obtain the relative change coefficient of the event characteristics;

[0172] 3) Use the numerical product of the relative change coefficient of event features and the sleep phase correction coefficient to correct the initially generated event intensity, and obtain the corrected event intensity.

[0173] In this embodiment, a numerical weighted calculation is performed on the change amounts of the combined event features of sleep respiration and blood oxygen for respiratory rate and minimum blood oxygen. The weight of the former is 0.3, and the weight of the latter is 0.7, to obtain the relative change coefficient of event features.

[0174] In this embodiment, the control relationships in the preset sleep phase - correction coefficient comparison table are as follows: wakefulness period - 0.80, light sleep period - 0.95, deep sleep period - 0.90, rapid eye movement sleep period - 1.0.

[0175] P400: Perform predictive analysis on the combined signal of sleep respiration and blood oxygen and the combined features of sleep respiration and blood oxygen, identify sleep respiration and blood oxygen prediction events and event occurrence trends, combine the sleep respiration knowledge base and the user's sleep respiration database, generate a sleep respiration and blood oxygen auxiliary regulation strategy, and send it to the sleep respiration regulation device through the signal interface.

[0176] The first step: Perform predictive analysis on the combined signal of sleep respiration and blood oxygen and the combined features of sleep respiration and blood oxygen, and respectively generate a combined prediction signal of sleep respiration and blood oxygen and a combined prediction feature of sleep respiration and blood oxygen.

[0177] In this embodiment, the AR method is selected to perform predictive analysis on the combined signal of sleep respiration and blood oxygen and the combined features of sleep respiration and blood oxygen. In actual application scenarios, the methods of predictive analysis at least include exponential smoothing method, Holt - Winters method, AR, MA, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA, VARMAX, machine learning, and can be flexibly selected and used according to the scenario.

[0178] The second step: Perform event detection analysis on the combined prediction signal of sleep respiration and blood oxygen, identify sleep respiration and blood oxygen events, and obtain sleep respiration and blood oxygen prediction events and event prediction intensities.

[0179] In this embodiment, based on the combined prediction signal of sleep respiration and blood oxygen, using the aforementioned sleep respiration event recognition model, sleep blood oxygen event recognition model, and event intensity initial generation method, identify respiratory events and blood oxygen events, extract sleep respiration and blood oxygen events and event intensities, and obtain sleep respiration and blood oxygen prediction events and event prediction intensities.

[0180] The third step: According to the sleep respiration and blood oxygen prediction events and the combined prediction features of sleep respiration and blood oxygen, correct the event prediction intensity and identify the event occurrence trend.

[0181] In this embodiment, according to the sleep apnea blood oxygen prediction events and the combined prediction features of sleep apnea blood oxygen, the event prediction intensity is corrected by using the aforementioned event intensity correction method.

[0182] In this embodiment, the event occurrence trend is obtained after machine learning performs learning training, data modeling, and analysis calculations on the large-scale sleep user sample data such as the combined signals of sleep apnea events and sleep blood oxygen events, and the combined features of sleep apnea blood oxygen. The event occurrence trend at least includes the event occurrence type, event occurrence probability, and event occurrence intensity.

[0183] Step 4: According to the sleep apnea blood oxygen prediction events and the event occurrence trend, combined with the sleep apnea knowledge base and the user's sleep apnea database, generate a sleep apnea blood oxygen auxiliary regulation strategy.

[0184] In this embodiment, the sleep apnea knowledge base mainly comes from the knowledge and experience of health management and clinical medicine related to sleep apnea, and at least includes the sleep apnea - blood oxygen rules, the characteristics of common sleep apnea - blood oxygen events, the common sleep apnea regulation methods, and the guidance of scenario intervention parameters; the user's sleep apnea database at least includes sleep apnea blood oxygen events, combined signals of sleep apnea blood oxygen, combined features of sleep apnea blood oxygen, event occurrence trend, and sleep apnea blood oxygen auxiliary regulation strategy.

[0185] In this embodiment, the sleep apnea blood oxygen auxiliary regulation strategy at least includes the blood oxygen target value, breathing frequency target value, breathing depth target value, regulation method, regulation time point, duration, and device control parameters.

[0186] In this embodiment, with the help of the preset sleep apnea knowledge base and the pre - constructed expert system, the sleep apnea blood oxygen events, event occurrence trend, blood oxygen target value, and breathing frequency target value are input into the expert system, and the expert system can output parameters such as the regulation method, regulation time point, duration, and device control parameters. It is worth mentioning that the expert system can be a traditional knowledge retrieval and application system or a machine learning model system.

[0187] Step 5: Send the sleep apnea blood oxygen auxiliary regulation strategy to the sleep apnea regulation device through the signal interface to optimize the operation control of the sleep apnea regulation device.

[0188] In this embodiment, the sleep apnea regulation device at least includes any one of a ventilator, an odor stimulation device, an electrical stimulation device, a tactile stimulation device, an environmental temperature and humidity regulation device, and an environmental CO2 concentration regulation device.

[0189] In this embodiment, the environmental temperature and humidity control device (intelligent air conditioner) and the odor stimulation device (intelligent aromatherapy system) are preferentially selected as the sleep breathing regulation devices. In the actual application scenario, the sleep breathing regulation device only needs to be able to connect to the network, receive and analyze parameters, and perform remote control, etc., to meet the basic requirements of the user for sleep breathing blood oxygen auxiliary regulation. In the actual application scenario, different sleep breathing blood oxygen auxiliary regulation strategies and sleep breathing regulation devices need to be selected according to the specific situation of the user and the facility conditions. For example, if the user has relatively severe sleep breathing events or blood oxygen events, an intelligent ventilator should be used to achieve breathing assistance.

[0190] P500: Generate and output a sleep breathing blood oxygen evaluation and auxiliary regulation report according to a preset cycle or strategy, and update the user sleep breathing database.

[0191] Step 1: Generate a sleep breathing blood oxygen evaluation and auxiliary regulation report according to a preset report cycle.

[0192] In this embodiment, the sleep breathing blood oxygen evaluation and auxiliary regulation report at least includes sleep blood oxygen signal, statistical analysis of sleep breathing blood oxygen events, key feature index curves in sleep breathing blood oxygen combined features, sleep phase curves, detection quantification and auxiliary regulation summary, and sleep breathing optimization suggestions.

[0193] Step 2: Output a sleep breathing blood oxygen evaluation and auxiliary regulation report according to the user scenario requirements.

[0194] In the actual application scenario, different generation cycles and different output methods of the sleep breathing blood oxygen evaluation and auxiliary regulation report are required to meet the user needs in multiple scenarios.

[0195] Step 3: Update the user sleep breathing database according to a preset data update strategy.

[0196] In this embodiment, the preset data update strategy is as follows: when no event occurs, the user sleep breathing database is updated every 1 minute; when an event occurs, the sleep breathing blood oxygen auxiliary regulation strategy is sent once, that is, the user sleep breathing database is updated immediately.

[0197] Combined Figure 2 As shown, a sleep breathing blood oxygen detection, analysis and auxiliary regulation system provided by an embodiment of the present invention includes the following modules:

[0198] A respiratory blood oxygen detection module S100, configured to collect and process sleep breathing signals and sleep blood oxygen signals during the user's sleep process to obtain a sleep breathing blood oxygen combined signal;

[0199] An event detection and analysis module S200, configured to perform event detection and analysis on the sleep breathing blood oxygen combined signal, identify respiratory events and blood oxygen events, and extract sleep breathing blood oxygen events and event intensities;

[0200] An event intensity adjustment module S300 is configured to perform feature analysis on the combined sleep apnea and blood oxygen signal, obtain the combined sleep apnea and blood oxygen features, and compare the relative changes in features during the occurrence of the sleep apnea and blood oxygen events to correct the event intensity.

[0201] A respiratory blood oxygen regulation module S400 is configured to perform predictive analysis on the combined sleep apnea and blood oxygen signal and the combined sleep apnea and blood oxygen features, identify the sleep apnea and blood oxygen prediction events and the event occurrence trends, generate a sleep apnea and blood oxygen assisted regulation strategy in combination with the sleep apnea knowledge base and the user's sleep apnea database, and send the sleep apnea regulation device through the signal interface.

[0202] A user report management module S500 is configured to generate and output a sleep apnea and blood oxygen evaluation and assisted regulation report according to a preset cycle or strategy, and update the user's sleep apnea database.

[0203] A data operation management module S600 is configured to perform visual management, unified storage, and operation management on all process data of the system.

[0204] In this embodiment, the respiratory blood oxygen detection module S100 further includes the following functional units:

[0205] A sleep apnea detection unit is configured to collect and monitor the user's sleep apnea behavior and perform signal processing to obtain a sleep apnea signal.

[0206] A sleep blood oxygen detection unit is configured to collect and monitor the user's sleep blood oxygen level and perform signal processing to obtain a sleep blood oxygen signal.

[0207] A combined signal collection unit is configured to collect the sleep apnea signal and the sleep blood oxygen signal to obtain a combined sleep apnea and blood oxygen signal.

[0208] In this embodiment, the event detection and analysis module S200 further includes the following functional units:

[0209] A respiratory event recognition unit is configured to perform event detection and analysis on the sleep apnea signal, identify respiratory events, and obtain sleep apnea events.

[0210] A blood oxygen event recognition unit is configured to perform event detection and analysis on the sleep blood oxygen signal, identify blood oxygen events, and obtain sleep blood oxygen events.

[0211] An event intensity calculation unit is configured to calculate the event intensity according to the sleep apnea events and the sleep apnea signal, the sleep blood oxygen events and the sleep blood oxygen signal.

[0212] An event information collection unit is configured to collect the sleep apnea events, the sleep blood oxygen events, and the event intensity according to the event occurrence time sequence to obtain sleep apnea and blood oxygen events.

[0213] In this embodiment, the event intensity adjustment module S300 further includes the following functional units:

[0214] The joint feature analysis unit is used to perform feature analysis on the sleep apnea and blood oxygen combined signal to obtain the sleep apnea and blood oxygen combined features;

[0215] The sleep phase recognition unit is used to recognize the sleep phase according to the sleep apnea and blood oxygen combined features and generate a sleep phase curve;

[0216] The feature change analysis unit is used to compare the relative change of the features of the sleep apnea and blood oxygen combined features during the occurrence of the sleep apnea and blood oxygen event to obtain the feature change amount of the sleep apnea and blood oxygen combined event;

[0217] The event intensity correction unit is used to correct the event intensity according to the sleep phase and the feature change amount of the sleep apnea and blood oxygen combined event.

[0218] In this embodiment, the respiratory blood oxygen regulation module S400 further includes the following functional units:

[0219] The trend prediction analysis unit is used to perform prediction analysis on the sleep apnea and blood oxygen combined signal and the sleep apnea and blood oxygen combined features, and generate a sleep apnea and blood oxygen combined prediction signal and a sleep apnea and blood oxygen combined prediction feature respectively;

[0220] The event detection analysis unit is used to perform event detection analysis on the sleep apnea and blood oxygen combined prediction signal to identify the sleep apnea and blood oxygen event, and obtain the sleep apnea and blood oxygen prediction event and the event prediction intensity;

[0221] The event occurrence prediction unit is used to correct the event prediction intensity and identify the event occurrence trend according to the sleep apnea and blood oxygen prediction event and the sleep apnea and blood oxygen combined prediction features;

[0222] The auxiliary strategy generation unit is used to generate a sleep apnea and blood oxygen auxiliary regulation strategy according to the sleep apnea and blood oxygen prediction event and the event occurrence trend, in combination with the sleep apnea knowledge base and the user's sleep apnea database;

[0223] The strategy sending control unit is used to send the sleep apnea and blood oxygen auxiliary regulation strategy to the sleep apnea regulation device through the signal interface to optimize the operation control of the sleep apnea regulation device.

[0224] In this embodiment, the user report management module S500 further includes the following functional units:

[0225] The user report generation unit is used to generate a sleep apnea and blood oxygen evaluation and auxiliary regulation report according to a preset report period;

[0226] A user report output unit for outputting a sleep apnea blood oxygen evaluation and auxiliary adjustment report according to user scenario requirements;

[0227] A database update unit for updating the user sleep apnea database according to a preset data update strategy.

[0228] In this embodiment, the data operation management module S600 further includes the following functional units:

[0229] A user information management unit for registering, inputting, editing, querying, outputting, and deleting user information;

[0230] A data visualization management unit for visual display management of all data in the system;

[0231] A data storage management unit for unified storage management of all data in the system;

[0232] A data operation management unit for backing up, migrating, and exporting all data in the system.

[0233] The system is configured to correspondingly execute Figure 1 each step in the method, which will not be elaborated here.

[0234] Combined with Figure 3 As shown, a sleep apnea blood oxygen detection, analysis, and auxiliary adjustment device provided by an embodiment of the present invention includes the following modules:

[0235] A respiratory blood oxygen detection module M100 for collecting and processing sleep apnea signals and sleep blood oxygen signals during a user's sleep to obtain a combined sleep apnea blood oxygen signal;

[0236] An event detection and analysis module M200 for performing event detection and analysis on the combined sleep apnea blood oxygen signal, identifying respiratory events and blood oxygen events, and extracting sleep apnea blood oxygen events and event intensities;

[0237] An event intensity adjustment module M300 for performing feature analysis on the combined sleep apnea blood oxygen signal to obtain combined sleep apnea blood oxygen features and comparing the relative changes in features during the occurrence of sleep apnea blood oxygen events to correct the event intensity;

[0238] A respiratory blood oxygen regulation module M400 for performing predictive analysis on the combined sleep apnea blood oxygen signal and the combined sleep apnea blood oxygen features, identifying sleep apnea blood oxygen prediction events and event occurrence trends, combining the sleep apnea knowledge base and the user sleep apnea database, generating a sleep apnea blood oxygen auxiliary regulation strategy, and sending it to a sleep apnea regulation device through a signal interface;

[0239] The user report management module M500 is used to generate and output a sleep apnea blood oxygen evaluation and auxiliary regulation report according to a preset period or strategy, and update the user's sleep apnea database;

[0240] The data visualization module M600 is used for unified visual display management of all process data and / or result data in the device;

[0241] The data management center module M700 is used for unified storage and data operation management of all process data and / or result data in the device.

[0242] The device is configured to correspondingly execute Figure 1 each step in the method, which will not be elaborated here.

[0243] The present invention also provides various programmable processors (FPGA, ASIC or other integrated circuits), and the processors are used to run a program. When the program runs, it executes the steps in the above embodiments.

[0244] The present invention also provides a corresponding computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the memory executes the program, it implements the steps in the above embodiments.

[0245] Although the disclosed embodiments of the present invention are as above, the content described is only an embodiment adopted for facilitating the understanding of the present invention and is not intended to limit the present invention. Any person skilled in the art within the scope of the present invention can make any modifications, changes, equivalent replacements, etc. in the form and details of the implementation without departing from the spirit and principles disclosed by the present invention. These all fall within the protection scope of the present invention. Therefore, the patent protection scope of the present invention shall still be subject to the scope defined by the appended claims.

Claims

1. A method for sleep breathing blood oxygen detection, analysis and auxiliary regulation, characterized in that The following steps are involved: Collect and process the user's sleep breathing signal and sleep blood oxygen signal during sleep to obtain a sleep breathing and blood oxygen combined signal; Performing event detection and analysis on the sleep respiratory blood oxygen joint signal, identifying respiratory events and blood oxygen events, and extracting sleep respiratory blood oxygen events and event intensities, including: performing event detection and analysis on the sleep respiratory signal, identifying respiratory events, and obtaining sleep respiratory events; performing event detection and analysis on the sleep blood oxygen signal, identifying blood oxygen events, and obtaining sleep blood oxygen events; calculating event intensities based on sleep respiratory events and sleep respiratory signals, sleep blood oxygen events and sleep blood oxygen signals; and grouping sleep respiratory events, sleep blood oxygen events, and event intensities according to the event occurrence sequence to obtain sleep respiratory blood oxygen events; Perform feature analysis on the sleep-respiration-blood-oxygen combined signal to obtain the sleep-respiration-blood-oxygen combined feature and compare the relative changes of the features during the occurrence of the sleep-respiration-blood-oxygen event to correct the event intensity; Predict and analyze the sleep breathing blood oxygen joint signal and sleep breathing blood oxygen joint features, identify the sleep breathing blood oxygen prediction events and event trends, combine the sleep breathing knowledge base and the user sleep breathing database, generate the sleep breathing blood oxygen auxiliary adjustment strategy and send it to the sleep breathing adjustment device through the signal interface; Generate and output sleep breathing blood oxygen evaluation and auxiliary adjustment reports according to the preset report cycle and preset data update strategy, and update the user's sleep breathing database; The initial generation methods of event intensity include: 1) Obtain the start time and end time of the sleep breathing event and the sleep blood oxygen event respectively, calculate the ratio of the time difference between the start time of the sleep breathing event and the sleep blood oxygen event to the time difference between the end time of the sleep breathing event and the sleep blood oxygen event, and obtain the respiratory blood oxygen event delay ratio; 2) Obtain the start time, end time, duration, peak and valley values, and peak and valley time of the sleep blood oxygen event, as well as the preset sleep blood oxygen signal threshold; 3) Calculate the relative change of the peak and valley values ​​of the sleep blood oxygen event and the preset sleep blood oxygen signal threshold to obtain the peak and valley relative value; 4) Calculate the linear slope according to the start time, peak-valley value and time at the peak-valley value of the sleep blood oxygen event to obtain the peak-valley front slope; 5) Calculate the linear slope according to the end time, peak-to-valley value and time at the peak-to-valley value of the sleep blood oxygen event to obtain the peak-to-valley trailing edge slope; 6) The event intensity is obtained by numerically fusion calculation based on the duration of sleep breathing events, the duration of sleep blood oxygen events, the delay ratio of breathing blood oxygen events, the peak-to-valley relative value, the peak-to-valley leading edge slope, and the peak-to-valley trailing edge slope.

2. The method according to claim 1, characterized in that, The specific steps of collecting and processing the sleep breathing signal and the sleep blood oxygen signal of the user during sleep to obtain the sleep breathing blood oxygen combined signal also include: Collecting and monitoring the user's sleep breathing behavior and performing signal processing to obtain the sleep breathing signal; Collecting and monitoring the user's sleep blood oxygen level and performing signal processing to obtain the sleep blood oxygen signal; The sleep breathing signal and the sleep blood oxygen signal are aggregated to obtain the sleep breathing blood oxygen combined signal.

3. The method according to claim 2, characterized in that , the acquisition and processing at least include acquisition, analog-to-digital conversion, resampling, rereferencing, power frequency notch filtering, bandpass filtering, mean filtering, smoothing processing, and signal time window segmentation.

4. The method according to claim 1 or 2, characterized in that , the combined sleep breathing and blood oxygen signal at least includes the sleep breathing signal and the sleep blood oxygen signal; wherein, the sleep breathing signal at least includes any one of an oral and nasal temperature monitoring signal, a nasal pressure monitoring signal, an oral and nasal carbon dioxide monitoring signal, a thoracic and abdominal respiratory movement signal, an electrocardiogram-derived respiratory signal, and a pharyngeal electromyogram signal, and the sleep blood oxygen signal at least includes any one of a photoplethysmography blood oxygen signal and a blood oxygenation level-dependent signal.

5. The method according to claim 1, wherein The event detection and analysis specifically refers to performing waveform feature recognition and marker extraction on the sleep breathing signal or the sleep blood oxygen signal according to a preset sleep breathing-blood oxygen event knowledge base and / or a machine learning model to obtain the basic event information of the sleep breathing event or the sleep blood oxygen event.

6. The method according to claim 5, characterized in that, The basic event information at least includes event type, start time, end time, duration, peak and valley values, and time at peak and valley values.

7. The method according to claim 1, characterized in that, The sleep breathing-blood oxygen event specifically consists of the sleep breathing event and the sleep blood oxygen event, and at least includes event type, start time, end time, duration, peak and valley values, time at peak and valley values, and event intensity.

8. The method according to claim 1, characterized in that, The specific steps of performing feature analysis on the combined sleep breathing and blood oxygen signal to obtain combined sleep breathing and blood oxygen features and comparing the relative changes in features during the occurrence of the sleep breathing-blood oxygen event to correct the event intensity further include: Performing feature analysis on the combined sleep breathing and blood oxygen signal to obtain the combined sleep breathing and blood oxygen features; Identifying sleep phases according to the combined sleep breathing and blood oxygen features and generating a sleep phase curve; Comparing the relative changes in features of the combined sleep breathing and blood oxygen features according to the occurrence period of the sleep breathing-blood oxygen event to obtain the change amount of combined sleep breathing and blood oxygen event features; Correcting the event intensity according to the sleep phase and the change amount of combined sleep breathing and blood oxygen event features.

9. The method according to claim 8, wherein The feature analysis includes at least one of numerical feature analysis, envelope feature analysis, time-frequency feature analysis, and non-linear feature analysis; wherein, the numerical features at least include at least one of average value, root mean square, maximum value, minimum value, variance, standard deviation, coefficient of variation, kurtosis, and skewness, the time-frequency features include at least one of band power, band power ratio, and band center frequency, and the non-linear features include at least one of entropy features, fractal features, and complexity features.

10. The method according to claim 9, wherein The combined sleep breathing and blood oxygen features at least include sleep breathing features and sleep blood oxygen features; the sleep breathing features at least include respiratory rate, numerical features, and time-frequency features of the sleep breathing signal; the sleep blood oxygen features at least include BOLD blood oxygen level features, and the BOLD blood oxygen level features at least include oxygenated hemoglobin concentration features and deoxygenated hemoglobin concentration features.

11. The method according to claim 8, wherein, The sleep phases at least include a wakefulness period, a light sleep period, a deep sleep period, and a rapid eye movement sleep period; the generation method of the sleep phases and the sleep phase curve includes: 1) Training and learning the combined sleep breathing and blood oxygen characteristics of the sleep user samples and their corresponding sleep stage data through machine learning to obtain a sleep phase staging model; 2) Inputting the combined sleep breathing and blood oxygen characteristics of the current user into the sleep phase staging model to obtain the corresponding sleep phase; 3) Extracting the values of the sleep phase for all signal time windows according to the time sequence to obtain the sleep phase curve.

12. The method according to claim 11, characterized in that, The correction method for the event intensity includes: 1) Obtaining the sleep phase, and extracting the sleep phase correction coefficient according to the preset sleep phase - correction coefficient comparison table; 2) Obtaining the change amount of the combined sleep breathing and blood oxygen event characteristics, determining and selecting the change amount of the target feature, and calculating through numerical weighting to obtain the relative change coefficient of the event feature; 3) Using the numerical product of the relative change coefficient of the event feature and the sleep phase correction coefficient to correct the initially generated event intensity to obtain the corrected event intensity.

13. The method according to claim 1, characterized in that, The specific steps of performing predictive analysis on the combined sleep breathing and blood oxygen signal and the combined sleep breathing and blood oxygen characteristics, identifying sleep breathing and blood oxygen prediction events and event occurrence trends, generating a sleep breathing and blood oxygen auxiliary regulation strategy in combination with a sleep breathing knowledge base and a user sleep breathing database, and sending it to a sleep breathing regulation device through a signal interface further include: Performing predictive analysis on the combined sleep breathing and blood oxygen signal and the combined sleep breathing and blood oxygen characteristics to respectively generate a combined sleep breathing and blood oxygen prediction signal and a combined sleep breathing and blood oxygen prediction feature; Performing event detection and analysis on the combined sleep breathing and blood oxygen prediction signal to identify sleep breathing and blood oxygen prediction events, and obtaining the sleep breathing and blood oxygen prediction events and event prediction intensity; Correcting the event prediction intensity and identifying the event occurrence trend according to the sleep breathing and blood oxygen prediction events and the combined sleep breathing and blood oxygen prediction features; Generating the sleep breathing and blood oxygen auxiliary regulation strategy according to the sleep breathing and blood oxygen prediction events and the event occurrence trend, in combination with a sleep breathing knowledge base and a user sleep breathing database; Sending the sleep breathing and blood oxygen auxiliary regulation strategy to the sleep breathing regulation device through a signal interface to optimize the operation control of the sleep breathing regulation device.

14. The method according to claim 13, wherein The method of predictive analysis includes at least any one of exponential smoothing method, AR, MA, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA, VARMAX.

15. The method according to claim 13, wherein The event occurrence trend includes at least event occurrence type, event occurrence probability, and event occurrence intensity.

16. The method according to claim 1 or 13, characterized in that, The sleep breathing knowledge base comes from the knowledge and experience of sleep breathing-related health management and clinical medicine, and at least includes sleep breathing - blood oxygen rules, common sleep breathing - blood oxygen event characteristics, common sleep breathing regulation methods, and scene intervention parameter guidance; the user sleep breathing database at least includes the sleep breathing and blood oxygen events, the combined sleep breathing and blood oxygen signal, the combined sleep breathing and blood oxygen characteristics, the event occurrence trend, and the sleep breathing and blood oxygen auxiliary regulation strategy.

17. The method according to claim 13, wherein The sleep apnea blood oxygen assisted regulation strategy includes at least one of a blood oxygen target value, a breathing frequency target value, a breathing depth target value, a regulation method, a regulation time point, and a duration.

18. The method according to claim 15, wherein The sleep apnea regulation device includes at least any one of a ventilator, an odor stimulation device, an electrical stimulation device, a tactile stimulation device, an environmental temperature and humidity regulation device, and an environmental carbon dioxide concentration regulation device.

19. The method according to claim 1, wherein The specific steps of generating and outputting a sleep apnea blood oxygen evaluation and assisted regulation report and updating the user sleep apnea database according to a preset reporting period and a preset data update strategy further include: Generating the sleep apnea blood oxygen evaluation and assisted regulation report according to a preset reporting period; Outputting the sleep apnea blood oxygen evaluation and assisted regulation report according to the user scenario requirements; Updating the user sleep apnea database according to a preset data update strategy.

20. The method according to claim 15, wherein The sleep apnea blood oxygen evaluation and assisted regulation report includes the sleep blood oxygen signal, the statistical analysis of the sleep apnea blood oxygen events, the sleep phase curve, and the sleep apnea optimization suggestions.

21. A sleep apnea blood oxygen detection, analysis and auxiliary regulation system, characterized in that, including: A respiratory blood oxygen detection module, configured to collect and process the sleep apnea signal and the sleep blood oxygen signal during the user's sleep process to obtain a combined sleep apnea blood oxygen signal; An event detection and analysis module, configured to perform event detection and analysis on the combined sleep apnea blood oxygen signal, identify respiratory events and blood oxygen events, and extract sleep apnea blood oxygen events and event intensities; the event detection and analysis module includes the following functional units: A respiratory event identification unit, configured to perform event detection and analysis on the sleep apnea signal to identify respiratory events and obtain sleep apnea events; A blood oxygen event identification unit, configured to perform event detection and analysis on the sleep blood oxygen signal to identify blood oxygen events and obtain sleep blood oxygen events; An event intensity calculation unit, configured to calculate the event intensity according to the sleep apnea events and the sleep apnea signal, and the sleep blood oxygen events and the sleep blood oxygen signal; An event information collection unit, configured to collect the sleep apnea events, the sleep blood oxygen events, and the event intensity in chronological order of event occurrence to obtain sleep apnea blood oxygen events; An event intensity adjustment module, configured to perform feature analysis on the combined sleep apnea blood oxygen signal to obtain combined sleep apnea blood oxygen features and compare the relative changes in features during the occurrence of sleep apnea blood oxygen events, and correct the event intensity; A respiratory blood oxygen regulation module, configured to perform prediction analysis on the combined sleep apnea blood oxygen signal and the combined sleep apnea blood oxygen features, identify sleep apnea blood oxygen prediction events and event occurrence trends, combine a sleep apnea knowledge base and a user sleep apnea database, generate a sleep apnea blood oxygen assisted regulation strategy, and send it to a sleep apnea regulation device through a signal interface; A user report management module, configured to generate and output a sleep apnea blood oxygen evaluation and assisted regulation report and update the user sleep apnea database according to a preset reporting period and a preset data update strategy; A data operation management module, configured to perform visual management, unified storage, and operation management on all process data of the system; The preliminary generation method of the event intensity includes: 1) Obtain the start time and end time of the sleep apnea event and the sleep blood oxygen event respectively, calculate the ratio of the time difference between the start times of the sleep apnea event and the sleep blood oxygen event to the time difference between the end times of the sleep apnea event and the sleep blood oxygen event, and obtain the respiratory blood oxygen event delay ratio; 2) Obtain the start time, end time, duration, peak valley value and the time at the peak valley value of the sleep blood oxygen event, as well as a preset sleep blood oxygen signal threshold; 3) Calculate the relative change amount between the peak valley value of the sleep blood oxygen event and the preset sleep blood oxygen signal threshold to obtain the peak valley relative value; 4) Calculate the linear slope based on the start time, peak valley value and the time at the peak valley value of the sleep blood oxygen event to obtain the peak valley front slope; 5) Calculate the linear slope based on the end time, peak valley value and the time at the peak valley value of the sleep blood oxygen event to obtain the peak valley rear slope; 6) Calculate the event intensity through numerical fusion of the duration of the sleep apnea event, the duration of the sleep blood oxygen event, the respiratory blood oxygen event delay ratio, the peak valley relative value, the peak valley front slope and the peak valley rear slope.

22. The system according to claim 21, wherein The respiratory blood oxygen detection module further includes the following functional units: A sleep apnea detection unit, configured to collect and monitor the sleep apnea behavior of the user and perform signal processing to obtain the sleep apnea signal; A sleep blood oxygen detection unit, configured to collect and monitor the sleep blood oxygen level of the user and perform signal processing to obtain the sleep blood oxygen signal; A combined signal collection unit, configured to collect the sleep apnea signal and the sleep blood oxygen signal to obtain the combined sleep apnea and blood oxygen signal.

23. The system according to any one of claims 21-22, characterized in that, The event intensity adjustment module includes the following functional units: A combined feature analysis unit, configured to perform feature analysis on the combined sleep apnea and blood oxygen signal to obtain the combined sleep apnea and blood oxygen feature; A sleep phase recognition unit, configured to recognize the sleep phase according to the combined sleep apnea and blood oxygen feature and generate a sleep phase curve; A feature change analysis unit, configured to compare the relative change of the combined sleep apnea and blood oxygen feature during the occurrence of the sleep apnea and blood oxygen event to obtain the change amount of the combined sleep apnea and blood oxygen event feature; An event intensity correction unit, configured to correct the event intensity according to the sleep phase and the change amount of the combined sleep apnea and blood oxygen event feature.

24. The system according to claim 23, wherein The respiratory blood oxygen regulation module includes the following functional units: A trend prediction analysis unit, configured to perform prediction analysis on the combined sleep apnea and blood oxygen signal and the combined sleep apnea and blood oxygen feature, and respectively generate a combined sleep apnea and blood oxygen prediction signal and a combined sleep apnea and blood oxygen prediction feature; An event detection analysis unit, configured to perform event detection analysis on the combined sleep apnea and blood oxygen prediction signal, identify the sleep apnea and blood oxygen prediction event, and obtain the sleep apnea and blood oxygen prediction event and the event prediction intensity; An event occurrence prediction unit, configured to correct the event prediction intensity according to the sleep apnea and blood oxygen prediction event and the combined sleep apnea and blood oxygen prediction feature and identify the event occurrence trend; An auxiliary strategy generation unit, configured to generate the sleep apnea blood oxygen auxiliary regulation strategy according to the sleep apnea blood oxygen prediction event and the event occurrence trend, in combination with the sleep apnea knowledge base and the user sleep apnea database; A strategy sending control unit, configured to send the sleep apnea blood oxygen auxiliary regulation strategy to the sleep apnea regulation device through a signal interface to optimize the operation control of the sleep apnea regulation device.

25. The system according to claim 21, wherein The user report management module includes the following functional units: A user report generation unit, configured to generate the sleep apnea blood oxygen evaluation and auxiliary regulation report according to a preset report period; A user report output unit, configured to output the sleep apnea blood oxygen evaluation and auxiliary regulation report according to the user scenario requirements; A database update unit, configured to update the user sleep apnea database according to a preset data update strategy.

26. The system according to claim 21, wherein The data operation management module further includes the following functional units: A user information management unit, configured for registration input, editing, querying, outputting, and deleting of user information; A data visualization management unit, configured for visual display management of all process data in the system; A data storage management unit, configured for unified storage management of all process data in the system; A data operation management unit, configured for backup of all process data in the system.

Images