Cardiovascular-based sleep respiratory evaluation and assisted adjustment method, system, and apparatus

Abstract

The application provides a cardiovascular-based sleep respiration evaluation and auxiliary adjustment method, system and device, sleep respiration cardiovascular dynamics signals are obtained through collection and processing of sleep cardiovascular signals and dynamics analysis, sleep respiration cardiovascular event information and event intensity are identified; further, signal feature analysis is performed on the sleep cardiovascular signals and the sleep respiration cardiovascular dynamics signals, sleep phases are identified, characteristic changes during sleep respiration cardiovascular event occurrence are compared and analyzed, and the event intensity of the sleep respiration cardiovascular event is corrected; finally, through occurrence trend prediction of the sleep respiration cardiovascular event, real-time generation and interface sending of a sleep respiration auxiliary adjustment strategy are completed, and the efficiency and effectiveness of a sleep respiration regulation device are optimized and improved. The application can realize scientific detection and evaluation and dynamic auxiliary adjustment of sleep respiration cardiovascular.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of sleep respiration detection evaluation and auxiliary adjustment, in particular to a cardiovascular-based sleep respiration evaluation and auxiliary adjustment method, system and device. BACKGROUND

[0002] The cardiovascular system and the respiratory system of the human body are a tightly coordinated and mutually influenced cardiopulmonary coupling working system, especially during sleep, the coupling coordination relationship becomes stronger. In addition, compared with the respiratory system, the cardiovascular system is a faster physiological system, and the heart rate is obviously higher than the respiratory frequency. This also makes the respiratory events such as smooth breathing, rapid breathing, and apnea can be directly observed and more finely described from the state changes of the cardiovascular system, whether in the task state or the resting state, the wake period or the sleep period.

[0003] At present, the existing sleep respiration detection evaluation technical scheme mainly relies on oral-nasal pressure, oral-nasal thermal and chest-abdominal belt pressure sensors or devices to directly detect and analyze, and completes the respiratory event detection, event type classification and event simple statistical analysis through waveform analysis. The existing technical scheme CN112971763A relates to a sleep respiration event detection device, which comprises a signal preprocessing module, a tidal volume calculation module, a threshold setting module, a respiratory event judgment module and a calibration module. According to the chest and abdominal respiration signals of the subject, the sleep respiration event is preliminarily determined, and then the blood oxygen saturation data and three-axis acceleration data corresponding to the chest and abdominal respiration signals are calibrated to determine the sleep respiration event. The sleep respiration event is respiratory sleep apnea or low ventilation. The respiratory sleep apnea is obstructive sleep apnea, central sleep apnea or mixed sleep apnea. The technical scheme mainly relies on threshold and counting method to detect and analyze the respiratory signal waveform to determine the sleep respiration event. Overall, the existing technical scheme lacks process observation, system quantification and evaluation calibration of sleep respiration events from the perspective of cardiovascular system to cardiopulmonary coupling system, especially lacks clear definition or feature quantification of sleep respiration event intensity from the perspective of respiratory dynamics and sleep phase state. In addition, the current sleep respiration regulation device is usually independent of the sleep respiration detection device, and most sleep respiration regulation devices can be connected to feedback regulation parameters but still use offline preset program control to complete device feedback regulation, which cannot perform intelligent, accurate and personalized dynamic auxiliary adjustment according to the real-time state of the user.

[0004] From the above, how to detect, observe and quantitatively evaluate sleep respiratory events through physiological changes of the cardiovascular system, further optimize and improve the efficiency of the existing sleep respiratory regulation device, and assist the user in sleep respiration and improve the sleep quality are problems to be further solved in the current product technical scheme and actual application scene. SUMMARY

[0005] In view of the above defects and improvement needs of the existing method, the purpose of the present application is to provide a cardiovascular-based sleep respiratory evaluation and auxiliary regulation method, sleep respiratory cardiovascular dynamics signals are obtained through collection and processing of sleep cardiovascular signals and dynamics analysis, sleep respiratory cardiovascular event information and event intensity are identified; further, feature analysis is performed on the sleep cardiovascular signals and the sleep respiratory cardiovascular dynamics signals, sleep phases are identified, feature changes during the occurrence of the sleep respiratory cardiovascular events are compared and analyzed, and the event intensity of the sleep respiratory cardiovascular events is corrected; finally, through prediction of the occurrence trend of the sleep respiratory cardiovascular events, real-time generation and interface sending of the sleep respiratory auxiliary regulation strategy are completed, thereby optimizing and improving the efficiency of the sleep respiratory regulation device, and realizing scientific detection and evaluation and dynamic auxiliary regulation of the user's sleep respiration, and efficiently assisting the user in sleep. The present application also provides a cardiovascular-based sleep respiratory evaluation and auxiliary regulation system for realizing the above method. The present application also provides a cardiovascular-based sleep respiratory evaluation and auxiliary regulation device for realizing the above system.

[0006] According to the purpose of the present application, the present application provides a cardiovascular-based sleep respiratory evaluation and auxiliary regulation method, comprising the following steps:

[0007] Collection and processing of sleep cardiovascular signals of a user and dynamics analysis are performed, sleep respiratory cardiovascular dynamics signals are extracted, and sleep respiratory cardiovascular events and event intensity are identified;

[0008] Feature analysis is performed on the sleep cardiovascular signals and the sleep respiratory cardiovascular dynamics signals, sleep respiratory cardiovascular features are obtained, and relative changes in features during the occurrence of the sleep respiratory cardiovascular events are compared, and the event intensity is corrected;

[0009] Trend prediction is performed on the sleep respiratory cardiovascular dynamics signals and the sleep respiratory cardiovascular features, and the sleep respiratory cardiovascular events and event occurrence trend are identified;

[0010] According to the sleep respiratory cardiovascular events and the event occurrence trend, a sleep respiratory auxiliary regulation strategy is generated in combination with a sleep respiratory knowledge base and a user sleep respiratory database, and a sleep respiratory regulation device is sent through a signal interface.

[0011] More preferably, the specific steps of collecting, processing and dynamically analyzing the sleep cardiovascular signals of the user, extracting sleep respiratory cardiovascular dynamics signals, and identifying sleep respiratory cardiovascular events and event intensity further comprise:

[0012] collecting and processing the cardiovascular signals of the user during sleep to obtain the sleep cardiovascular signals;

[0013] dynamically analyzing the sleep cardiovascular signals to obtain the sleep respiratory cardiovascular dynamics signals;

[0014] performing event detection analysis on the sleep respiratory cardiovascular dynamics signals to identify and extract the sleep respiratory cardiovascular events and the event intensity.

[0015] More preferably, the cardiovascular signals at least include any of electrocardiogram signals and pulse signals.

[0016] More preferably, the collection and processing at least include collection and acquisition, analog-to-digital conversion, resampling, re-referencing, artifact removal, noise reduction, notch filtering, band-pass filtering, mean filtering, smoothing processing, and signal time window segmentation.

[0017] More preferably, the dynamic analysis at least includes electrocardiogram-derived respiratory signal extraction, heart rate curve extraction, electrocardiogram PQRST wave pattern identification and extraction; and the sleep respiratory cardiovascular dynamics signals at least include any of electrocardiogram-derived respiratory signals, heart rate signals, electrocardiogram P wave signals, electrocardiogram QRS wave signals, and electrocardiogram T wave signals.

[0018] More preferably, the event detection analysis specifically comprises waveform feature identification and label extraction of the sleep respiratory cardiovascular dynamics signals according to a pre-set sleep respiratory event knowledge base and / or a machine learning model to obtain relevant information of the sleep respiratory cardiovascular events.

[0019] More preferably, the sleep respiratory cardiovascular events at least include event type, start time, end time, duration, peak-to-valley value, peak-to-valley value time, event intensity, and event level.

[0020] More preferably, the event intensity is specifically determined by the duration of the sleep respiratory cardiovascular events, the peak-to-valley value and the relative time at the peak-to-valley value of the dynamics signals, and the amplitude changes of sleep respiratory cardiovascular characteristics and sleep phase staging.

[0021] More preferably, the generation and calculation method of the event intensity specifically comprises:

[0022] 1) obtaining the start time, end time, duration, peak-to-valley value, peak-to-valley value time of the sleep respiratory cardiovascular events, and signal feature threshold value;

[0023] 2) calculating the relative change amount of the peak-trough value and the signal feature threshold value of the sleep respiratory cardiovascular event, to obtain a peak-trough relative value;

[0024] 3) calculating a linear slope according to the start time, peak-trough value and time at the peak-trough value of the sleep respiratory cardiovascular event, to obtain a peak-trough front slope;

[0025] 4) calculating a linear slope according to the end time, peak-trough value and time at the peak-trough value of the sleep respiratory cardiovascular event, to obtain a peak-trough rear slope;

[0026] 5) calculating the event intensity according to the combination of the duration, peak-trough relative value, peak-trough front slope and peak-trough rear slope of the sleep respiratory cardiovascular event.

[0027] More preferably, the specific steps of correcting the event intensity of the sleep respiratory cardiovascular event according to the sleep cardiovascular signal and the sleep respiratory cardiovascular dynamics signal, obtaining sleep respiratory cardiovascular features and comparing the relative changes of the features during the occurrence of the sleep respiratory cardiovascular event, further comprise:

[0028] performing feature analysis on the sleep cardiovascular signal and the sleep respiratory cardiovascular dynamics signal to obtain the sleep respiratory cardiovascular features;

[0029] identifying sleep phase stages according to the sleep respiratory cardiovascular features, and generating a sleep phase curve;

[0030] comparing and analyzing the relative changes of the features during the occurrence of the sleep respiratory cardiovascular event according to the sleep respiratory cardiovascular event and the sleep respiratory cardiovascular features, to obtain sleep respiratory cardiovascular event feature change amounts;

[0031] correcting the event intensity of the sleep respiratory cardiovascular event according to the sleep respiratory cardiovascular event feature change amounts and the sleep phase stages.

[0032] More preferably, the feature analysis at least includes numerical feature analysis, envelope feature analysis, time-frequency feature analysis and nonlinear feature analysis; wherein the nonlinear feature at least includes entropy feature, fractal feature and complexity feature.

[0033] More preferably, the sleep respiratory cardiovascular features at least include numerical features, envelope features, time-frequency features and nonlinear features; the numerical features at least include dynamics signal waveform features and heart rate variability features, the dynamics signal waveform features at least include wave type, amplitude, period and time course; 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 frequency band power, frequency band power ratio and frequency band center frequency.

[0034] More preferably, the sleep phase staging at least includes wakefulness period, light sleep period, deep sleep period and rapid eye movement sleep period; the method for generating the sleep phase staging and the sleep phase curve is specifically:

[0035] 1) Learning training and data modeling of the sleep respiratory cardiovascular features of a large-scale sleep user sample and its corresponding sleep staging data through machine learning, to obtain a sleep phase staging model;

[0036] 2) Inputting the sleep respiratory cardiovascular features of the current user into the sleep phase staging model to obtain the corresponding sleep phase staging;

[0037] 3) Extracting the numerical value of the sleep phase staging of all signal time windows in sequence to obtain the sleep phase curve.

[0038] More preferably, the correction method of the event intensity is specifically:

[0039] 1) Obtain the sleep respiratory cardiovascular event feature change amount, determine and select the feature change amount of the target feature, and obtain the event feature relative change coefficient through numerical weighting calculation;

[0040] 2) Obtain the sleep phase staging, extract the sleep phase correction coefficient according to the preset sleep phase-correction coefficient table;

[0041] 3) Identify the signal source of the sleep respiratory cardiovascular dynamics signal, and extract the signal source correction coefficient according to the preset signal source-correction coefficient table;

[0042] 4) Use the product of the event feature relative change coefficient, the sleep phase correction coefficient and the dynamics signal source correction coefficient, and the numerical value of the event intensity, as the event intensity of the corrected sleep respiratory cardiovascular event.

[0043] More preferably, the specific steps of the trend prediction of the sleep respiratory cardiovascular dynamics signal and the sleep respiratory cardiovascular features, and the identification of the sleep respiratory cardiovascular event and the event occurrence trend further include:

[0044] Performing trend prediction on the sleep respiratory cardiovascular dynamics signal to obtain a cardiovascular dynamics prediction signal;

[0045] Performing trend prediction on the sleep respiratory cardiovascular features to obtain a respiratory cardiovascular feature prediction signal;

[0046] Performing event detection analysis on the cardiovascular dynamics prediction signal and the respiratory cardiovascular feature prediction signal to obtain the predicted sleep respiratory cardiovascular event and identify the event occurrence trend.

[0047] More preferably, the trend prediction method comprises at least any one of exponential smoothing, Holt-Winters, AR, MA, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA, VARMAX, machine learning.

[0048] More preferably, the event occurrence trend comprises at least event occurrence type, event occurrence probability, and event occurrence intensity; and the event occurrence trend is obtained by machine learning, data modeling, and analysis calculation on the sleep respiratory cardiovascular event related cardiovascular dynamics signal and respiratory cardiovascular features of a large sample of sleep users.

[0049] More preferably, the specific steps of generating a sleep respiratory auxiliary adjustment strategy according to the sleep respiratory cardiovascular event and the event occurrence trend, combining a sleep respiratory knowledge base and a user sleep respiratory database, and sending the sleep respiratory adjustment device through a signal interface further comprise:

[0050] generating the sleep respiratory auxiliary adjustment strategy according to the sleep respiratory cardiovascular event and the event occurrence trend, combining a sleep respiratory knowledge base and a user sleep respiratory database;

[0051] sending the sleep respiratory auxiliary adjustment strategy to the sleep respiratory adjustment device through a signal interface, optimizing the sleep respiratory adjustment device operation control, to realize dynamic auxiliary adjustment of user sleep respiration;

[0052] generating and outputting a sleep respiratory cardiovascular evaluation and auxiliary adjustment report according to a preset report period.

[0053] More preferably, the sleep respiratory knowledge base mainly comes from the knowledge and experience of sleep respiratory related health management and clinical medicine, and at least includes sleep respiratory rules, sleep cardiovascular rules, common sleep respiratory-cardiovascular event characteristics, and common sleep respiratory adjustment methods, i.e. scene intervention parameter guidance; and the user sleep respiratory database at least includes the sleep respiratory cardiovascular dynamics signal, the sleep respiratory cardiovascular feature, the sleep respiratory cardiovascular event, the event occurrence trend, and the sleep respiratory auxiliary adjustment strategy.

[0054] More preferably, the sleep respiratory auxiliary adjustment strategy at least includes respiratory frequency target value, respiratory depth target value, adjustment method, adjustment time point, duration, and device control parameter.

[0055] More preferably, the sleep respiratory adjustment device at least includes any one of a ventilator, an odor stimulation device, an electric stimulation device, a tactile stimulation device, an environmental temperature and humidity control device, and an environmental CO2 concentration control device.

[0056] More preferably, the sleep respiratory cardiovascular evaluation and auxiliary adjustment report comprises at least statistical analysis of the sleep respiratory cardiovascular events, key feature index curves in the sleep respiratory cardiovascular features, sleep phase curves, respiratory detection and auxiliary adjustment summary, and sleep respiratory optimization suggestions.

[0057] According to the purposes of the present application, the present application proposes a cardiovascular-based sleep respiratory evaluation and auxiliary adjustment system, comprising the following modules:

[0058] An event detection and identification module is configured to collect and process sleep cardiovascular signals of a user, perform dynamic analysis, extract sleep respiratory cardiovascular dynamic signals, identify sleep respiratory cardiovascular events and event intensity, and perform trend prediction on the sleep respiratory cardiovascular dynamic signals and the sleep respiratory cardiovascular features to identify the sleep respiratory cardiovascular events and event occurrence trends.

[0059] An event intensity correction module is configured to perform feature analysis on the sleep cardiovascular signals and the sleep respiratory cardiovascular dynamic signals, obtain sleep respiratory cardiovascular features, compare relative changes in the features during occurrence of the sleep respiratory cardiovascular events, and correct the event intensity.

[0060] An event occurrence prediction module is configured to perform trend prediction on the sleep respiratory cardiovascular dynamic signals and the sleep respiratory cardiovascular features to identify the sleep respiratory cardiovascular events and event occurrence trends.

[0061] A strategy auxiliary regulation module is configured to generate sleep respiratory auxiliary adjustment strategies based on the sleep respiratory cardiovascular events and the event occurrence trends, in combination with a sleep respiratory knowledge base and a user sleep respiratory database, and send sleep respiratory adjustment equipment through a signal interface.

[0062] A data operation management module is configured to visually manage, uniformly store, and operationally manage all process data of the system.

[0063] More preferably, the event detection and identification module further comprises the following functional units:

[0064] A cardiovascular signal monitoring unit is configured to collect and process cardiovascular signals of a user during sleep to obtain the sleep cardiovascular signals.

[0065] A dynamic analysis unit is configured to perform dynamic analysis on the sleep cardiovascular signals to obtain the sleep respiratory cardiovascular dynamic signals.

[0066] An event detection and analysis unit is configured to perform event detection and analysis on the sleep respiratory cardiovascular dynamic signals to identify and extract the sleep respiratory cardiovascular events and the event intensity.

[0067] More preferably, the event intensity correction module further comprises the following functional units:

[0068] a cardiovascular feature analysis unit configured to perform feature analysis on the sleep cardiovascular signal and the sleep respiratory cardiovascular dynamics signal to obtain sleep respiratory cardiovascular features;

[0069] a sleep phase recognition unit configured to recognize sleep phase staging according to the sleep respiratory cardiovascular features, and generate a sleep phase curve;

[0070] an event feature comparison unit configured to compare and analyze relative changes in features during occurrence of the sleep respiratory cardiovascular events according to the sleep respiratory cardiovascular events and the sleep respiratory cardiovascular features, to obtain sleep respiratory cardiovascular event feature change amounts;

[0071] an event intensity correction unit configured to correct the event intensity of the sleep respiratory cardiovascular events according to the sleep respiratory cardiovascular event feature change amounts and the sleep phase staging.

[0072] More preferably, the event occurrence prediction module further comprises the following functional units:

[0073] a dynamics signal prediction unit configured to perform trend prediction on the sleep respiratory cardiovascular dynamics signal to obtain a cardiovascular dynamics prediction signal;

[0074] a cardiovascular feature prediction unit configured to perform trend prediction on the sleep respiratory cardiovascular features to obtain a respiratory cardiovascular feature prediction signal;

[0075] an event trend prediction unit configured to perform event detection analysis on the cardiovascular dynamics prediction signal and the respiratory cardiovascular feature prediction signal to obtain predicted sleep respiratory cardiovascular events, and identify the event occurrence trend.

[0076] More preferably, the strategy assisted regulation module further comprises the following functional units:

[0077] an assisted strategy generation unit configured to generate the sleep respiratory assisted regulation strategy according to the sleep respiratory cardiovascular events and the event occurrence trend, in combination with a sleep respiratory knowledge base and a user sleep respiratory database;

[0078] an assisted strategy sending unit configured to send the sleep respiratory assisted regulation strategy to the sleep respiratory regulation device through a signal interface, and optimize operation control of the sleep respiratory regulation device to achieve dynamic assisted regulation of user sleep respiration;

[0079] a report generation and output unit configured to generate and output a sleep respiratory cardiovascular evaluation and assisted regulation report according to a preset report period.

[0080] More preferably, the data operation management module further comprises the following functional units:

[0081] a user information management unit for registering, inputting, editing, querying, outputting and deleting user information;

[0082] a visualization management unit for visualizing and managing all data in the system;

[0083] a data storage management unit for uniformly storing and managing all data in the system;

[0084] a data operation management unit for backing up, migrating and exporting all data in the system.

[0085] According to the purpose of the present application, the present application provides a cardiovascular-based sleep respiration evaluation and auxiliary adjustment device, comprising the following modules:

[0086] an event detection and identification module for collecting and processing sleep cardiovascular signals of a user and performing dynamic analysis, extracting sleep respiration cardiovascular dynamic signals, identifying sleep respiration cardiovascular events and event intensity;

[0087] an event intensity correction module for performing feature analysis on the sleep cardiovascular signals and the sleep respiration cardiovascular dynamic signals, obtaining sleep respiration cardiovascular features and comparing the relative changes of the features during the occurrence of the sleep respiration cardiovascular events, and correcting the event intensity;

[0088] an event occurrence prediction module for performing trend prediction on the sleep respiration cardiovascular dynamic signals and the sleep respiration cardiovascular features, identifying the sleep respiration cardiovascular events and event occurrence trends;

[0089] a strategy auxiliary regulation module for generating sleep respiration auxiliary adjustment strategies according to the sleep respiration cardiovascular events and the event occurrence trends, combining a sleep respiration knowledge base and a user sleep respiration database, and sending sleep respiration adjustment equipment through a signal interface;

[0090] a data operation management module for visualizing, uniformly storing and managing all process data of the device.

[0091] The application provides a cardiovascular-based sleep respiration evaluation and auxiliary adjustment method, system and device, comprehensively considers the mutual coupling and mutual influence relationship of sleep cardiovascular and sleep respiration in the sleep process, utilizes the characteristics that sleep cardiovascular signals have good sleep respiration behavior dynamic characteristic representation ability, such as derived respiration or heart rate curves, identifies sleep respiration-cardiovascular event information and accurately quantifies and corrects event intensity through fusion analysis of sleep cardiovascular signals and sleep respiration-cardiovascular dynamic signals, and the sleep respiration auxiliary adjustment strategy based on sleep respiration-cardiovascular events and event occurrence trend (occurrence type, occurrence probability and occurrence intensity) can dynamically optimize and improve the efficiency and effectiveness of sleep respiration regulation equipment, so that the integration of scientific detection and evaluation and dynamic auxiliary adjustment of user sleep respiration is realized, and the user sleep is efficiently assisted.

[0092] In addition, in the prior art, only the occurrence of sleep respiration events is measured by the basic waveform characteristics of respiration dynamics, while in the application, the intensity of sleep respiration-cardiovascular events is further corrected by the cardiovascular change amplitude and sleep phase staging and other key sleep physiological state characteristics, so as to observe, analyze and quantify sleep respiration-cardiovascular events from the whole physiological system level, which is more in line with the unity, synergy and complex network dynamic characteristics of human physiology, and can more truly, accurately, scientifically and effectively quantify and evaluate user sleep respiration-cardiovascular events and event intensity. The application provides a new collaborative framework for sleep respiration event quantitative evaluation and auxiliary adjustment based on cardiovascular dynamics. In actual application scenarios, sleep respiration-related detection and regulation systems or devices can integrate all or part of the technical points or functions provided by the technical solutions of the application in a whole or partial manner, better meeting the needs of different user service scenarios. In addition, the technical solutions of the application can be quickly expanded to the observation, analysis and auxiliary adjustment of the multi-scene task of the cardiopulmonary coupling working system in various scenes such as task state or resting state, wakefulness or sleep, and the like.

[0093] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent from the description, or can be learned by practice of the present application. The objects and other advantages of the present application will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings. BRIEF DESCRIPTION OF DRAWINGS

[0094] The accompanying drawings are included to provide a further understanding of the technical solutions of the application, and constitute a part of the specification, and are used together with the embodiments of the present application to explain the technical solutions of the present application, and do not constitute a limitation on the technical solutions of the present application.

[0095] Figure 1 is a flow step schematic diagram of a cardiovascular-based sleep respiration evaluation and auxiliary adjustment method provided by an embodiment of the present application;

[0096] Figure 2 This is a schematic diagram of the module composition of a cardiovascular-based sleep breathing assessment and auxiliary regulation system provided in an embodiment of the present invention;

[0097] Figure 3 This is a schematic diagram of the module structure of a cardiovascular-based sleep breathing assessment and auxiliary regulation device provided in an embodiment of the present invention. Detailed Implementation

[0098] To more clearly illustrate the objectives and technical solutions of this invention, the invention will be further described below with reference to the accompanying drawings of the embodiments of this application. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort should fall within the protection scope of this invention. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other.

[0099] The applicant discovered that, similar to respiratory behavior, human cardiac physiological (ECG / pulse) signals differ significantly between sleep and wakefulness, exhibiting entirely different characteristics: First, sleep cardiovascular behavior and sleep respiratory behavior are a highly synergistic and mutually influential process. Sleep cardiovascular dynamic characteristic signals, such as sleep ECG-derived respiratory signals or heart rate variability signals, are very similar to sleep respiratory signals and are easier to collect and extract. Furthermore, sleep ECG / pulse signals, sleep ECG-derived respiratory signals, and heart rate variability signals can provide a more comprehensive description and accurate characterization of sleep respiratory events, especially in the analysis and prediction of breathing patterns and events; the greater the intensity of sleep respiratory events, the more intense the oscillations in the heart rate curve. Second, different sleep phases possess different sleep respiratory behavior characteristics; light sleep, deep sleep, and REM sleep, among others, have a significant impact on respiratory behavior patterns and the intensity of sleep respiratory events.

[0100] Therefore, this invention will use sleep cardiovascular signals as a basis to achieve scientific and comprehensive detection, analysis and quantitative evaluation of sleep breathing and sleep breathing cardiovascular events, and further complete the predictive analysis of sleep breathing cardiovascular events and the real-time generation of sleep breathing regulation strategies, thereby realizing the integration of scientific detection, evaluation and dynamic auxiliary regulation of users' sleep breathing, effectively assisting users' sleep breathing and improving sleep quality.

[0101] Combination Figure 1 As shown in the figure, an embodiment of the present invention provides a method for evaluating and assisting in the regulation of sleep breathing based on cardiovascular function, comprising the following steps:

[0102] P100: Collecting, processing and dynamics analysis of sleep cardiovascular signals of a user, extracting sleep respiratory cardiovascular dynamics signals, identifying sleep respiratory cardiovascular events and event intensity.

[0103] The first step is to collect and process cardiovascular signals of a user during sleep to obtain sleep cardiovascular signals.

[0104] In this embodiment, the cardiovascular signals include at least any of electrocardiogram signals and pulse signals. The collection and processing of the cardiovascular signals include at least collection and acquisition, analog-to-digital conversion, resampling, re-reference, de-artifacting, noise reduction, notch filtering, band-pass filtering, mean filtering, smoothing processing, and signal time window segmentation.

[0105] In this embodiment, a single-channel collection and monitoring of sleep electrocardiogram signals of a user is performed by a single-lead electrocardiogram monitor, the collection position is above the left chest, and the sampling rate is 1024 Hz. The electrocardiogram signals are processed by resampling (256 Hz), de-artifacting, correction, wavelet noise reduction, and 0.5-40 Hz band-pass filtering to obtain sleep cardiovascular signals.

[0106] The second step is to perform dynamics analysis on the sleep cardiovascular signals to obtain sleep respiratory cardiovascular dynamics signals.

[0107] In this embodiment, the dynamics analysis includes at least extraction of electrocardiogram-derived respiratory signals, extraction of heart rate curves, and identification and extraction of electrocardiogram PQRST waveforms. The sleep respiratory cardiovascular dynamics signals include at least any of electrocardiogram-derived respiratory signals, heart rate signals, electrocardiogram P wave signals, electrocardiogram QRS wave signals, and electrocardiogram T wave signals.

[0108] In this embodiment, electrocardiogram-derived respiratory (EDR) signals and heart rate signals are extracted from sleep electrocardiogram signals as sleep respiratory cardiovascular dynamics signals. The sampling rates of the above-mentioned signals are all 256 Hz. In actual application scenarios, a person skilled in the art can extract electrocardiogram-derived respiratory (EDR) signals and heart rate signals by means of basic general technologies, various mature algorithms or software in the field of electrocardiogram signal processing, while extracting and identifying P, QRS, and T waves in electrocardiogram signals. These signals, like traditional sleep respiratory signals (oral and nasal pressure, oral and nasal thermosensitive, and chest and abdominal belt pressure, etc.), have good sleep respiratory dynamics description characteristics.

[0109] The third step is to perform event detection analysis on the sleep respiratory cardiovascular dynamics signals to identify and extract sleep respiratory cardiovascular events and event intensity.

[0110] In this embodiment, the event detection analysis specifically refers to waveform feature recognition and label extraction of sleep respiratory cardiovascular dynamics signals according to a preset sleep respiratory event knowledge base and / or a machine learning model, to obtain relevant information of sleep respiratory cardiovascular events. The sleep respiratory cardiovascular events at least include event type, start time, end time, duration, peak-to-valley value, peak-to-valley time, event intensity, and event level.

[0111] In this embodiment, the sleep respiratory cardiovascular event detection model is obtained by learning and training of electrocardio-derived respiration (EDR) signals and event label information through machine learning technology, and data modeling. Further, the sleep respiratory cardiovascular event detection model can quickly and accurately detect and identify sleep respiratory cardiovascular events, and further extract various information of sleep respiratory cardiovascular events through waveform feature analysis.

[0112] In actual application scenarios, the joint use of electrocardio-derived respiration signals and heart rate signals can also well identify and extract sleep respiratory cardiovascular events. First, the start time, end time, and duration of sleep respiratory event occurrence are extracted from the waveform changes of electrocardio-derived respiration signals; second, the peak-to-valley value and peak-to-valley time are extracted from the heart rate curve signal by comparing the start time and end time, and the event intensity is preliminarily calculated; finally, the event type and event level are preliminarily determined according to the sleep respiratory clinical rules through the waveform changes of electrocardio-derived respiration signals and heart rate curve signals. In the subsequent combination of sleep respiratory cardiovascular feature changes, the event intensity is further corrected and updated. It is worth noting that the stress changes of different users and different states of heart rate for sleep respiratory cardiovascular events are not the same, plus the different basic heart rates, so the basic situation and current situation of the user need to be considered when the event level is preliminarily set according to the heart rate.

[0113] In this embodiment, the event intensity is specifically determined by the duration of sleep respiratory cardiovascular events, the peak-to-valley value and peak-to-valley time of dynamics signals, and the amplitude changes of sleep respiratory cardiovascular features and sleep phase staging. In this embodiment, the generation and calculation method of event intensity is specifically:

[0114] 1) Obtain the start time, end time, duration, peak-to-valley value, peak-to-valley time of sleep respiratory cardiovascular events, and signal feature threshold value;

[0115] 2) Calculate the relative change amount of the peak-to-valley value and the signal feature threshold value of sleep respiratory cardiovascular events to obtain the peak-to-valley relative value;

[0116] 3) Calculate the linear slope according to the start time, peak-to-valley value, and peak-to-valley time of sleep respiratory cardiovascular events to obtain the peak-to-valley front slope;

[0117] 4) Calculate the linear slope according to the end time, peak value and valley value of the sleep respiratory cardiovascular event, to obtain the peak-valley trailing edge slope;

[0118] 5) Calculate the event intensity according to the numerical combination of the duration, peak-valley relative value, peak-valley leading edge slope and peak-valley trailing edge slope of the sleep respiratory cardiovascular event.

[0119] In actual application scenarios, the numerical combination calculation method of the event intensity is selected according to the actual situation of different user scenarios. Two calculation formulas are provided in this embodiment to state the construction method.

[0120] The first calculation formula method of the event intensity is specifically:

[0121]

[0122] The second calculation formula method of the event intensity is specifically:

[0123] eti=Td*Pv*(||LSp||+||RSp||)

[0124] Wherein, eti is the event intensity, Td is the duration of the sleep respiratory event, Pv is the peak-valley relative value, LSp is the peak-valley leading edge slope, RSp is the peak-valley trailing edge slope, and || || is the absolute value operation.

[0125] P200: performing feature analysis on the sleep cardiovascular signal and the sleep respiratory cardiovascular dynamics signal to obtain sleep respiratory cardiovascular features and comparing the relative changes of the features during the occurrence of the sleep respiratory cardiovascular event to correct the event intensity.

[0126] Step 1, performing feature analysis on the sleep cardiovascular signal and the sleep respiratory cardiovascular dynamics signal to obtain sleep respiratory cardiovascular features.

[0127] In this embodiment, the feature analysis at least includes numerical feature analysis, envelope feature analysis, time-frequency feature analysis and nonlinear feature analysis; wherein the nonlinear feature at least includes entropy feature, fractal feature and complexity feature.

[0128] In this embodiment, the sleep respiratory cardiovascular features at least include numerical features, envelope features, time-frequency features and nonlinear features; the numerical features at least include dynamics signal waveform features and heart rate variability features, and the dynamics signal waveform features at least include wave type, amplitude, period and time course; 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 frequency band power, frequency band power ratio and frequency band center frequency.

[0129] It is worth mentioning that in actual application scenarios, if the signal (sleep cardiovascular signal or sleep respiratory cardiovascular dynamics signal) is too long, the signal is divided into several small signals, the signal characteristics of each small signal are obtained, and finally the average value of all small segment corresponding signal characteristics is calculated as the sleep respiratory cardiovascular characteristics, especially when calculating the nonlinear characteristics. But need to consider the influence of the start and end point interval of sleep respiratory event.

[0130] The second step is to identify the sleep phase stage according to the sleep respiratory cardiovascular characteristics, and generate a sleep phase curve.

[0131] In this embodiment, the sleep phase stage at least includes wakefulness period, light sleep period, deep sleep period and rapid eye movement sleep period; the sleep phase stage and the generation method of the sleep phase curve are specifically:

[0132] 1) Through machine learning, the sleep respiratory cardiovascular characteristics of the large-scale sleep user sample and the corresponding sleep stage data are learned and trained and data modeling is performed to obtain a sleep phase stage model;

[0133] 2) The sleep respiratory cardiovascular characteristics of the current user are input into the sleep phase stage model to obtain the corresponding sleep phase stage;

[0134] 3) The sleep phase stage values of all signal time windows are extracted in time sequence to obtain a sleep phase curve.

[0135] The third step is to compare and analyze the relative changes of the characteristics during the occurrence of the sleep respiratory cardiovascular event according to the sleep respiratory cardiovascular event and the sleep respiratory cardiovascular characteristics, and obtain the sleep respiratory cardiovascular event characteristic change amount.

[0136] In this embodiment, the heart rate variability characteristics (heart rate, heart rate variability coefficient, RR interval and NN interval) are selected as the key characteristics of the sleep respiratory cardiovascular characteristics. The sleep respiratory cardiovascular characteristics in the middle interval of the start time of the current sleep respiratory cardiovascular event and the end time of the last sleep respiratory cardiovascular event are taken as the comparison interval of the current sleep respiratory cardiovascular event. The sleep respiratory cardiovascular characteristics during the sleep respiratory cardiovascular event and the sleep respiratory cardiovascular characteristics in the comparison interval are compared and analyzed to extract the relative change amount and obtain the sleep respiratory cardiovascular event characteristic change amount.

[0137] In this embodiment, the calculation formula of the relative change of the characteristics is:

[0138]

[0139] Wherein, iFea is the relative change of the characteristics, FE i , FB i are the characteristic values during the occurrence and the comparison interval respectively, and || || is the absolute value calculation.

[0140] In actual application scenarios, the distance, correlation, and other relationship characteristic values of the sleep respiratory cardiovascular features in the event period and the contrast interval of the sleep respiratory cardiovascular event are used to replace the relative change amount to realize feature comparison and analysis in the event period and the contrast interval, and also can evaluate the occurrence features and state changes of the sleep respiratory cardiovascular event.

[0141] Fourthly, the event intensity of the sleep respiratory cardiovascular event is corrected according to the sleep respiratory cardiovascular event feature change amount and the sleep phase staging.

[0142] In this embodiment, the correction method of the event intensity is specifically as follows:

[0143] 1) The sleep respiratory cardiovascular event feature change amount is obtained, the feature change amount of the target feature is determined and selected, and the event feature relative change coefficient is obtained through numerical weighting calculation;

[0144] 2) The sleep phase staging is obtained, and the sleep phase correction coefficient is extracted according to the preset sleep phase-correction coefficient table;

[0145] 3) The signal source of the sleep respiratory cardiovascular dynamics signal is identified, and the signal source correction coefficient is extracted according to the preset signal source-correction coefficient table;

[0146] 4) The product of the event feature relative change coefficient, the sleep phase correction coefficient, and the dynamics signal source correction coefficient, and the numerical value of the event intensity is used as the corrected event intensity of the sleep respiratory cardiovascular event.

[0147] In this embodiment, the feature change amount of the heart rate variability feature (heart rate, heart rate variability coefficient, RR interval, and NN interval) in the sleep respiratory cardiovascular event feature change amount is obtained through numerical weighting calculation to obtain the event feature relative change coefficient.

[0148] In this embodiment, the comparison relationship in the preset sleep phase-correction coefficient table is as follows: wake-up period-0.80, light sleep period-0.95, deep sleep period-0.90, and rapid eye movement sleep period-1.0.

[0149] In this embodiment, the comparison relationship in the preset signal source-correction coefficient table is as follows: electrocardiogram-derived respiratory signal-1.0, heart rate signal-0.95, electrocardiogram P wave signal-0.80, electrocardiogram QRS wave signal-0.85, and electrocardiogram T wave signal-0.70.

[0150] In actual application scenarios, there are great differences in sleep breathing behavior and cardiovascular behavior in different sleep phase stages, and the sleep breathing behavior, disorder occurrence probability and intensity will also have great differences, so it is necessary to combine the cardiovascular physiological state and sleep phase information to re-correct the intensity of sleep breathing behavior or event.

[0151] P300: trend prediction is performed on the sleep breathing cardiovascular dynamics signal and the sleep breathing cardiovascular feature, and the sleep breathing cardiovascular event and event occurrence trend are identified.

[0152] In this embodiment, the method of trend prediction at least includes any one of exponential smoothing method, Holt-Winters method, AR, MA, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA, VARMAX, machine learning.

[0153] The first step is to perform trend prediction on the sleep breathing cardiovascular dynamics signal to obtain a cardiovascular dynamics prediction signal.

[0154] In this embodiment, the ARMA method is selected to perform trend prediction on the sleep breathing cardiovascular dynamics signal.

[0155] The second step is to perform trend prediction on the sleep breathing cardiovascular feature to obtain a breathing cardiovascular feature prediction signal.

[0156] In this embodiment, the ARMA method is selected to perform trend prediction on the sleep breathing cardiovascular feature.

[0157] The third step is to perform event detection analysis on the cardiovascular dynamics prediction signal and the breathing cardiovascular feature prediction signal to obtain a predicted sleep breathing cardiovascular event and identify the event occurrence trend.

[0158] In this embodiment, first, the aforementioned event detection analysis method is used to input the cardiovascular dynamics prediction signal into the constructed sleep breathing cardiovascular event detection model to identify and extract the sleep breathing cardiovascular event and event intensity; second, the aforementioned event intensity correction method is used to complete the re-correction of the event intensity; and finally, the key information of the predicted sleep breathing cardiovascular event is obtained.

[0159] In addition, in this embodiment, the event occurrence trend at least includes event occurrence type, event occurrence probability and event occurrence intensity. The event occurrence trend is obtained by machine learning on the cardiovascular dynamics signal and breathing cardiovascular feature related to the sleep breathing cardiovascular event of the large-scale sleep user sample data after learning training, data modeling and analysis calculation.

[0160] P400: generating a sleep respiration auxiliary adjustment strategy according to the sleep respiration cardiovascular event and the event occurrence trend, combining a sleep respiration knowledge base and a user sleep respiration database, and sending the sleep respiration adjustment device through a signal interface.

[0161] The first step is to generate a sleep respiration auxiliary adjustment strategy according to the sleep respiration cardiovascular event and the event occurrence trend, combining a sleep respiration knowledge base and a user sleep respiration database.

[0162] In this embodiment, the sleep respiration knowledge base mainly comes from the knowledge and experience of sleep respiration related health management and clinical medicine, and at least includes sleep respiration rules, sleep cardiovascular rules, common sleep respiration-cardiovascular event characteristics, and common sleep respiration adjustment methods, i.e. scene intervention parameter guidance; the user sleep respiration database at least includes sleep respiration cardiovascular dynamics signals, sleep respiration cardiovascular characteristics, sleep respiration cardiovascular events, event occurrence trends, and sleep respiration auxiliary adjustment strategies.

[0163] In this embodiment, the sleep respiration auxiliary adjustment strategy at least includes a respiration frequency target value, a respiration depth target value, an adjustment mode, an adjustment time point, a duration, and a device control parameter.

[0164] In this embodiment, by means of a pre-set sleep respiration knowledge base and a pre-constructed expert system, the sleep respiration cardiovascular event and the event occurrence trend, the respiration frequency target value, and the respiration depth target value are input into the expert system, and the expert system can output parameters such as the adjustment mode, the adjustment time point, the duration, and the device control parameter. It is worth mentioning that the expert system can be a traditional knowledge retrieval and application system, or a machine learning model system.

[0165] The second step is to send the sleep respiration auxiliary adjustment strategy to the sleep respiration adjustment device through a signal interface, and optimize the sleep respiration adjustment device operation control to realize dynamic auxiliary adjustment of user sleep respiration.

[0166] In this embodiment, the sleep respiration adjustment device at least includes any one of a respirator, an odor stimulation device, an electric stimulation device, a tactile stimulation device, an environmental temperature and humidity control device, and an environmental CO2 concentration control device.

[0167] In this embodiment, the environmental temperature and humidity control device (smart air conditioner and smart fresh air system) is selected as the sleep respiration adjustment device. In actual application scenarios, the sleep respiration adjustment device only needs to be able to be connected to the network, receive and analyze parameters, and be remotely controlled, etc. to meet the basic requirements of user sleep respiration cardiovascular auxiliary adjustment. In addition, under the premise of ensuring the control effect, the mode and device that interfere less with the user's sleep should be preferred.

[0168] Thirdly, a sleep respiration cardiovascular evaluation and auxiliary adjustment report is generated and output according to a preset report period.

[0169] In this embodiment, the sleep respiration cardiovascular evaluation and auxiliary adjustment report at least includes statistical analysis of sleep respiration cardiovascular events, key feature index curves in sleep respiration cardiovascular features, sleep phase curves, respiration detection and auxiliary adjustment summary, and sleep respiration optimization suggestions.

[0170] In actual application scenarios, different generation periods and different output modes of the sleep respiration cardiovascular evaluation and auxiliary adjustment report need to be formulated to meet the needs of users in different scenarios.

[0171] In combination with Figure 2 As shown in the figure, the sleep respiration evaluation and auxiliary adjustment system based on the cardiovascular system provided by the embodiment of the present application comprises the following modules:

[0172] The event detection and identification module S100 is used for collecting and processing sleep cardiovascular signals of a user and performing dynamic analysis, extracting sleep respiration cardiovascular dynamic signals, identifying sleep respiration cardiovascular events and event intensity;

[0173] The event intensity correction module S200 is used for performing feature analysis on sleep cardiovascular signals and sleep respiration cardiovascular dynamic signals, obtaining sleep respiration cardiovascular features, and comparing relative changes in features during occurrence of sleep respiration cardiovascular events to correct event intensity;

[0174] The event occurrence prediction module S300 is used for performing trend prediction on sleep respiration cardiovascular dynamic signals and sleep respiration cardiovascular features, identifying sleep respiration cardiovascular events and event occurrence trends;

[0175] The strategy auxiliary regulation module S400 is used for generating sleep respiration auxiliary adjustment strategies according to sleep respiration cardiovascular events and event occurrence trends, combining a sleep respiration knowledge base and a user sleep respiration database, and sending sleep respiration adjustment equipment through a signal interface;

[0176] The data operation management module S500 is used for visually managing, uniformly storing, and operationally managing all process data of the system.

[0177] In this embodiment, the event detection and identification module S100 further comprises the following functional units:

[0178] The cardiovascular signal monitoring unit is used for collecting and processing cardiovascular signals of a user during sleep to obtain sleep cardiovascular signals;

[0179] The dynamic analysis unit is used for performing dynamic analysis on sleep cardiovascular signals to obtain sleep respiration cardiovascular dynamic signals;

[0180] an event detection analysis unit configured to perform event detection analysis on the sleep cardiorespiratory dynamics signal, identify and extract sleep cardiorespiratory events and event intensity.

[0181] In this embodiment, the event intensity correction module S200 further comprises the following functional units:

[0182] a cardiorespiratory feature analysis unit configured to perform feature analysis on the sleep cardiorespiratory signal and the sleep cardiorespiratory dynamics signal to obtain sleep cardiorespiratory features;

[0183] a sleep phase identification unit configured to identify sleep phase staging according to the sleep cardiorespiratory features, and generate a sleep phase curve;

[0184] an event feature comparison unit configured to compare and analyze relative changes in features during occurrence of the sleep cardiorespiratory events according to the sleep cardiorespiratory events and the sleep cardiorespiratory features, and obtain sleep cardiorespiratory event feature change amounts;

[0185] an event intensity correction unit configured to correct event intensity of the sleep cardiorespiratory events according to the sleep cardiorespiratory event feature change amounts and the sleep phase staging.

[0186] In this embodiment, the event occurrence prediction module S300 further comprises the following functional units:

[0187] a dynamics signal prediction unit configured to perform trend prediction on the sleep cardiorespiratory dynamics signal to obtain a cardiorespiratory dynamics prediction signal;

[0188] a cardiorespiratory feature prediction unit configured to perform trend prediction on the sleep cardiorespiratory features to obtain a cardiorespiratory feature prediction signal;

[0189] an event trend prediction unit configured to perform event detection analysis on the cardiorespiratory dynamics prediction signal and the cardiorespiratory feature prediction signal to obtain predicted sleep cardiorespiratory events and identify event occurrence trends.

[0190] In this embodiment, the strategy assisted regulation module S400 further comprises the following functional units:

[0191] an assisted strategy generation unit configured to generate sleep respiration assisted regulation strategies according to the sleep cardiorespiratory events and the event occurrence trends, in combination with a sleep respiration knowledge base and a user sleep respiration database;

[0192] an assisted strategy sending unit configured to send the sleep respiration assisted regulation strategies to a sleep respiration regulation device through a signal interface, and optimize operation control of the sleep respiration regulation device to achieve dynamic assisted regulation of user sleep respiration;

[0193] A report generation output unit is configured to generate and output a sleep respiration cardiovascular evaluation and auxiliary adjustment report according to a preset report period.

[0194] In this embodiment, the data operation management module S500 further comprises the following functional units:

[0195] A user information management unit is configured to register, input, edit, query, output and delete user information.

[0196] A visualization management unit is configured to visually display and manage all data in the system.

[0197] A data storage management unit is configured to uniformly store and manage all data in the system.

[0198] A data operation management unit is configured to back up, migrate and export all data in the system.

[0199] The system is configured to correspondingly execute each step in the method Figure 1 , which will not be described here again.

[0200] In combination with the Figure 3 , the embodiment of the present application provides a sleep respiration cardiovascular evaluation and auxiliary adjustment device, which comprises the following modules:

[0201] An event detection and identification module M100 is configured to collect and process sleep cardiovascular signals of a user, perform dynamic analysis, extract sleep respiration cardiovascular dynamic signals, identify sleep respiration cardiovascular events and event intensity.

[0202] An event intensity correction module M200 is configured to perform feature analysis on sleep cardiovascular signals and sleep respiration cardiovascular dynamic signals, obtain sleep respiration cardiovascular features, compare relative changes in features during occurrence of sleep respiration cardiovascular events, and correct event intensity.

[0203] An event occurrence prediction module M300 is configured to perform trend prediction on sleep respiration cardiovascular dynamic signals and sleep respiration cardiovascular features, identify sleep respiration cardiovascular events and event occurrence trends.

[0204] A strategy auxiliary regulation module M400 is configured to generate sleep respiration auxiliary adjustment strategies according to sleep respiration cardiovascular events and event occurrence trends, in combination with a sleep respiration knowledge base and a user sleep respiration database, and send sleep respiration adjustment equipment through a signal interface.

[0205] A data operation management module M500 is configured to visually manage, uniformly store and operationally manage all process data of the device.

[0206] The device is configured to correspondingly execute each step in the method Figure 1The steps in the method will not be elaborated here.

[0207] The present invention also provides various programmable processors (FPGA, ASIC or other integrated circuits) for running programs, wherein the programs execute the steps in the above embodiments.

[0208] The present invention also provides a corresponding computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the memory executes the program to implement the steps in the above embodiments.

[0209] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art can make any modifications, changes, or equivalent substitutions in the form and details of the implementation without departing from the spirit and principles disclosed in this invention; all such modifications and changes fall within the scope of protection of this invention. Therefore, the scope of patent protection for this invention shall be determined by the scope defined in the appended claims.

Claims

1. A cardiovascular-based sleep disordered breathing evaluation and assisted adjustment method, characterized by, The method comprises the following steps: Collecting and processing sleep cardiovascular signals of a user, and performing dynamic analysis to extract sleep respiratory cardiovascular dynamic signals, identify sleep respiratory cardiovascular events and event intensity; Performing feature analysis on the sleep cardiovascular signals and the sleep respiratory cardiovascular dynamic signals to obtain sleep respiratory cardiovascular features and compare the relative changes in the features during the occurrence of the sleep respiratory cardiovascular events to correct the event intensity; Performing trend prediction on the sleep respiratory cardiovascular dynamic signals and the sleep respiratory cardiovascular features to identify the sleep respiratory cardiovascular events and event occurrence trends; According to the sleep respiratory cardiovascular events and the event occurrence trends, combining a sleep respiratory knowledge base and a user sleep respiratory database, generating a sleep respiratory auxiliary adjustment strategy and sending a sleep respiratory adjustment device through a signal interface; The event intensity is determined by the duration of the sleep respiratory cardiovascular events, the peak-to-valley value of the dynamic signals, the relative time at the peak-to-valley value, the amplitude change of the sleep respiratory cardiovascular features, and the sleep phase staging. The generation and calculation method of the event intensity comprises: 1) obtaining the start time, end time, duration, peak-to-valley value, time at the peak-to-valley value, and signal feature threshold of the sleep respiratory cardiovascular events; 2) calculating the relative change amount of the peak-to-valley value and the signal feature threshold of the sleep respiratory cardiovascular events to obtain a peak-to-valley relative value; 3) calculating the linear slope according to the start time, peak-to-valley value, and time at the peak-to-valley value of the sleep respiratory cardiovascular events to obtain a peak-to-valley front slope; 4) calculating the linear slope according to the end time, peak-to-valley value, and time at the peak-to-valley value of the sleep respiratory cardiovascular events to obtain a peak-to-valley rear slope; 5) calculating the event intensity according to the numerical combination of the duration of the sleep respiratory cardiovascular events, the peak-to-valley relative value, the peak-to-valley front slope, and the peak-to-valley rear slope.

2. The method of claim 1, wherein, The specific steps of collecting and processing sleep cardiovascular signals of a user, extracting sleep respiratory cardiovascular dynamic signals, and identifying sleep respiratory cardiovascular events and event intensity further comprise: Collecting and processing cardiovascular signals during the sleep process of a user to obtain the sleep cardiovascular signals; Performing dynamic analysis on the sleep cardiovascular signals to obtain the sleep respiratory cardiovascular dynamic signals; Performing event detection analysis on the sleep respiratory cardiovascular dynamic signals to identify and extract the sleep respiratory cardiovascular events and the event intensity.

3. The method of claim 2, wherein The cardiovascular signals at least include any one of electrocardiogram signals and pulse signals.

4. The method of claim 2, wherein The collection and processing at least include collection and acquisition, analog-to-digital conversion, resampling, re-reference, artifact removal, noise reduction, notch filtering, band-pass filtering, mean filtering, smoothing processing, and signal time window segmentation.

5. The method of claim 1 or 2, wherein The dynamic analysis at least includes electrocardiogram-derived respiratory signal extraction, heart rate curve extraction, electrocardiogram PQRST wave type identification and extraction; the sleep respiratory cardiovascular dynamic signals at least include any one of electrocardiogram-derived respiratory signals, heart rate signals, electrocardiogram P wave signals, electrocardiogram QRS wave signals, and electrocardiogram T wave signals.

6. The method of claim 2, wherein, The event detection analysis specifically comprises waveform feature recognition and label extraction on the sleep respiratory cardiovascular dynamics signal according to a preset sleep respiratory event knowledge base and / or a machine learning model, to obtain relevant information of the sleep respiratory cardiovascular event.

7. The method of claim 1 or 2, wherein, The sleep respiratory cardiovascular event at least comprises an event type, a start time, an end time, a duration, a peak-valley value, a time at the peak-valley value, the event intensity, and an event level.

8. The method of claim 1, wherein, The specific steps of the feature analysis on the sleep cardiovascular signal and the sleep respiratory cardiovascular dynamics signal to obtain sleep respiratory cardiovascular features and comparative analysis of relative changes in features during occurrence of the sleep respiratory cardiovascular event to correct the event intensity further comprise: The feature analysis on the sleep cardiovascular signal and the sleep respiratory cardiovascular dynamics signal to obtain the sleep respiratory cardiovascular features; According to the sleep respiratory cardiovascular features, sleep phase staging is recognized, and a sleep phase curve is generated; According to the sleep respiratory cardiovascular event and the sleep respiratory cardiovascular features, comparative analysis of relative changes in features during occurrence of the sleep respiratory cardiovascular event is performed to obtain a sleep respiratory cardiovascular event feature change amount; According to the sleep respiratory cardiovascular event feature change amount and the sleep phase staging, the event intensity of the sleep respiratory cardiovascular event is corrected.

9. The method of claim 8, wherein, The feature analysis comprises at least one of numerical feature analysis, envelope feature analysis, time-frequency feature analysis, and nonlinear feature analysis; wherein the nonlinear feature comprises at least one of entropy feature, fractal feature, and complexity feature.

10. The method of claim 8, wherein, The sleep respiratory cardiovascular features comprise at least one of numerical features, envelope features, time-frequency features, and nonlinear features; the numerical features comprise at least one of dynamics signal waveform features and heart rate variability features, the dynamics signal waveform features comprise at least one of wave type, amplitude, period, and time course; the numerical features comprise at least one of mean value, root mean square, maximum value, minimum value, variance, standard deviation, coefficient of variation, kurtosis, and skewness; the time-frequency features comprise at least one of frequency band power, frequency band power ratio, and frequency band center frequency.

11. The method of claim 8, wherein, The sleep phase staging at least comprises a wake period, a light sleep period, a deep sleep period, and a rapid eye movement sleep period; the sleep phase staging and the generation method of the sleep phase curve are specifically as follows: 1) Machine learning is used to learn and train the sleep respiratory cardiovascular features of a large-scale sleep user sample and corresponding sleep staging data, to obtain a sleep phase staging model; 2) The sleep respiratory cardiovascular features of a current user are input into the sleep phase staging model, to obtain corresponding sleep phase staging; 3) Numerical values of the sleep phase staging of all signal time windows are extracted in time sequence, to obtain the sleep phase curve.

12. The method of claim 8 or 11, wherein, The correction method of the event intensity comprises: 1) The sleep respiratory cardiovascular event feature change amount is obtained, a feature change amount of a target feature is determined and selected, an event feature relative change coefficient is obtained through numerical weighting calculation; 2) The sleep phase staging is obtained, and a sleep phase correction coefficient is extracted according to a preset sleep phase-correction coefficient reference table. 3) identifying the signal source of the sleep respiratory cardiovascular dynamics signal, extracting the signal source correction coefficient according to the preset signal source-correction coefficient table; 4) using the product of the event feature relative change coefficient, the sleep phase correction coefficient, the dynamics signal source correction coefficient, and the numerical value of the event intensity as the corrected event intensity of the sleep respiratory cardiovascular event.

13. The method of claim 1, wherein, The specific steps of the trend prediction of the sleep respiratory cardiovascular dynamics signal and the sleep respiratory cardiovascular feature, the identification of the sleep respiratory cardiovascular event and the event occurrence trend further include: Trend prediction of the sleep respiratory cardiovascular dynamics signal to obtain a cardiovascular dynamics prediction signal; Trend prediction of the sleep respiratory cardiovascular feature to obtain a respiratory cardiovascular feature prediction signal; Event detection analysis of the cardiovascular dynamics prediction signal and the respiratory cardiovascular feature prediction signal to obtain the predicted sleep respiratory cardiovascular event and identify the event occurrence trend.

14. The method of claim 13, wherein, The trend prediction method at least includes any one of exponential smoothing, Holt-Winters, AR, MA, ARMA, ARIMA, SARIMA, SARIMAX, VAR, VARMA, VARMAX, machine learning.

15. The method of claim 13, wherein, The event occurrence trend at least includes event occurrence type, event occurrence probability, and event occurrence intensity; the event occurrence trend is obtained by machine learning on the sleep respiratory cardiovascular event-related cardiovascular dynamics signal and respiratory cardiovascular feature of the large-scale sleep user sample data after learning training, data modeling, and analysis calculation.

16. The method of claim 1, wherein, The specific steps of generating the sleep respiratory auxiliary adjustment strategy according to the sleep respiratory cardiovascular event and the event occurrence trend, combining the sleep respiratory knowledge base and the user sleep respiratory database, and sending the sleep respiratory adjustment device through the signal interface further include: Generating the sleep respiratory auxiliary adjustment strategy according to the sleep respiratory cardiovascular event and the event occurrence trend, combining the sleep respiratory knowledge base and the user sleep respiratory database; Sending the sleep respiratory auxiliary adjustment strategy to the sleep respiratory adjustment device through the signal interface, optimizing the sleep respiratory adjustment device operation control to realize dynamic auxiliary adjustment of the user sleep respiratory; Generating and outputting the sleep respiratory cardiovascular evaluation and auxiliary adjustment report according to the preset report period.

17. The method of claim 16, wherein, The sleep respiratory knowledge base comes from the knowledge and experience of sleep respiratory-related health management and clinical medicine, at least including any one of sleep respiratory rules, sleep cardiovascular rules, common sleep respiratory-cardiovascular event features, and commonly used sleep respiratory adjustment methods, i.e., scene intervention parameter guidance.

18. The method of claim 16, wherein, The sleep respiratory auxiliary adjustment strategy includes at least one of respiratory frequency target value, respiratory depth target value, adjustment method, adjustment time point, duration, and device control parameter. The sleep respiratory auxiliary adjustment strategy includes at least one of respiratory frequency target value, respiratory depth target value, adjustment method, adjustment time point, duration, and device control parameter.

19. The method of claim 16 or 17, wherein, The sleep respiratory regulation device at least comprises any one of a breathing machine, an odor stimulation device, an electric stimulation device, a tactile stimulation device, an environmental temperature and humidity regulation device and an environmental concentration regulation device.

20. The method of claim 16, wherein, The sleep respiration cardiovascular evaluation and auxiliary adjustment report comprises at least one of statistical analysis of the sleep respiration cardiovascular event, key feature index curve in the sleep respiration cardiovascular feature, sleep phase curve, respiration detection and auxiliary adjustment summary, and sleep respiration optimization suggestion.

21. A cardiovascular-based sleep disordered breathing evaluation and assisted adjustment system, comprising: The system comprises the following modules: An event detection and identification module, which is configured to collect and process sleep cardiovascular signals of a user, perform dynamic analysis, extract sleep respiration cardiovascular dynamic signals, identify sleep respiration cardiovascular events and event intensity, and output the sleep respiration cardiovascular dynamic signals and the sleep respiration cardiovascular events and event intensity. An event intensity correction module, which is configured to perform feature analysis on the sleep cardiovascular signals and the sleep respiration cardiovascular dynamic signals, obtain sleep respiration cardiovascular features, compare relative changes in the features during occurrence of the sleep respiration cardiovascular events, and correct the event intensity. An event occurrence prediction module, which is configured to perform trend prediction on the sleep respiration cardiovascular dynamic signals and the sleep respiration cardiovascular features, identify the sleep respiration cardiovascular events and event occurrence trend. A strategy auxiliary adjustment module, which is configured to generate sleep respiration auxiliary adjustment strategies according to the sleep respiration cardiovascular events and the event occurrence trend, and send the sleep respiration adjustment strategies to sleep respiration adjustment equipment through a signal interface. A data operation management module, which is configured to visually manage, uniformly store, and operationally manage all process data of the system. The event intensity is determined by a duration of the sleep respiration cardiovascular event, peak and valley values of dynamic signals, relative time at the peak and valley values, amplitude changes in the sleep respiration cardiovascular features, and sleep phase staging. The event intensity is generated by the following method: 1) obtaining a start time, an end time, a duration, a peak and valley value, a time at the peak and valley values, and a signal feature threshold value of the sleep respiration cardiovascular event; 2) calculating a relative change amount of the peak and valley value and the signal feature threshold value of the sleep respiration cardiovascular event to obtain a peak and valley relative value; 3) calculating a linear slope according to the start time, the peak and valley value, and the time at the peak and valley values of the sleep respiration cardiovascular event to obtain a peak and valley front slope; 4) calculating a linear slope according to the end time, the peak and valley value, and the time at the peak and valley values of the sleep respiration cardiovascular event to obtain a peak and valley rear slope; 5) calculating the event intensity according to a numerical combination of the duration, the peak and valley relative value, the peak and valley front slope, and the peak and valley rear slope of the sleep respiration cardiovascular event.

22. The system of claim 21, wherein, The event detection and identification module further comprises the following functional units: A cardiovascular signal monitoring unit, which is configured to collect and process cardiovascular signals of a user during sleep to obtain the sleep cardiovascular signals; A dynamic analysis unit, which is configured to perform dynamic analysis on the sleep cardiovascular signals to obtain the sleep respiration cardiovascular dynamic signals; An event detection and analysis unit, which is configured to perform event detection and analysis on the sleep respiration cardiovascular dynamic signals to identify and extract the sleep respiration cardiovascular events and the event intensity.

23. The system of claim 21 or 22, wherein, The event intensity correction module further comprises the following functional units: a cardiovascular feature analysis unit configured to perform feature analysis on the sleep cardiovascular signal and the sleep respiratory cardiovascular dynamics signal to obtain sleep respiratory cardiovascular features; a sleep phase recognition unit configured to recognize sleep phase staging according to the sleep respiratory cardiovascular features, and generate a sleep phase curve; an event feature comparison unit configured to compare and analyze relative changes in features during occurrence of the sleep respiratory cardiovascular events according to the sleep respiratory cardiovascular features and the sleep respiratory cardiovascular dynamics signal, to obtain sleep respiratory cardiovascular event feature change amounts; an event intensity correction unit configured to correct the event intensity of the sleep respiratory cardiovascular events according to the sleep respiratory cardiovascular event feature change amounts and the sleep phase staging.

24. The system of claim 23, wherein, The event occurrence prediction module further includes the following functional units: a dynamics signal prediction unit configured to perform trend prediction on the sleep respiratory cardiovascular dynamics signal to obtain a cardiovascular dynamics prediction signal; a cardiovascular feature prediction unit configured to perform trend prediction on the sleep respiratory cardiovascular features to obtain respiratory cardiovascular feature prediction signals; an event trend prediction unit configured to perform event detection analysis on the cardiovascular dynamics prediction signal and the respiratory cardiovascular feature prediction signals to obtain predicted sleep respiratory cardiovascular events, and identify the event occurrence trend.

25. The system of claim 21 or 24, wherein, The strategy assisted regulation module further includes the following functional units: an assisted strategy generation unit configured to generate the sleep respiratory assisted regulation strategy according to the sleep respiratory cardiovascular events and the event occurrence trend, in combination with a sleep respiratory knowledge base and a user sleep respiratory database; an assisted strategy sending unit configured to send the sleep respiratory assisted regulation strategy to the sleep respiratory regulation device through a signal interface, and optimize operation control of the sleep respiratory regulation device to achieve dynamic assisted regulation of user sleep respiration; a report generation and output unit configured to generate and output a sleep respiratory cardiovascular evaluation and assisted regulation report according to a preset report period.

26. The system of claim 25, wherein, The data operation management module further includes the following functional units: a user information management unit configured to perform registration input, editing, query, output and deletion of user information; a visualization management unit configured to perform visualization display management of all data in the system; a data storage management unit configured to perform unified storage management of all data in the system; a data operation management unit configured to perform backup, migration and export of all data in the system.

27. A cardiovascular-based sleep disordered breathing evaluation and assist adjustment device, comprising: The following modules are included: an event detection and recognition module configured to perform collection and processing of user sleep cardiovascular signals, and dynamics analysis, to extract sleep respiratory cardiovascular dynamics signals, recognize sleep respiratory cardiovascular events and event intensity; an event intensity correction module configured to perform feature analysis on the sleep cardiovascular signal and the sleep respiratory cardiovascular dynamics signal to obtain sleep respiratory cardiovascular features, and compare relative changes in features during occurrence of the sleep respiratory cardiovascular events, to correct the event intensity; An event occurrence prediction module is configured to trend predict the sleep respiratory cardiovascular dynamics signal and the sleep respiratory cardiovascular feature, identify the sleep respiratory cardiovascular event and event occurrence trend; A strategy assisted regulation module is configured to generate a sleep respiratory assisted regulation strategy according to the sleep respiratory cardiovascular event and the event occurrence trend, and send a sleep respiratory regulation device through a signal interface in combination with a sleep respiratory knowledge base and a user sleep respiratory database; A data operation management module is configured to visually manage, uniformly store and operationally manage all process data of the device; The event intensity is determined by the duration of the sleep respiratory cardiovascular event, the peak-to-valley value of the dynamics signal, the relative time at the peak-to-valley value, the amplitude change of the sleep respiratory cardiovascular feature and the sleep phase staging; The generation and calculation method of the event intensity comprises: 1) obtaining the start time, end time, duration, peak-to-valley value, time at the peak-to-valley value and signal feature threshold value of the sleep respiratory cardiovascular event; 2) calculating the relative change amount of the peak-to-valley value and the signal feature threshold value of the sleep respiratory cardiovascular event to obtain a peak-to-valley relative value; 3) calculating a linear slope according to the start time, peak-to-valley value and time at the peak-to-valley value of the sleep respiratory cardiovascular event to obtain a peak-to-valley front slope; 4) calculating a linear slope according to the end time, peak-to-valley value and time at the peak-to-valley value of the sleep respiratory cardiovascular event to obtain a peak-to-valley rear slope; 5) calculating the event intensity according to the duration of the sleep respiratory cardiovascular event, the peak-to-valley relative value, the peak-to-valley front slope and the peak-to-valley rear slope.

Images