CN116211256B - Non-contact sleep breathing signal acquisition method and device - Google Patents
Non-contact sleep breathing signal acquisition method and device Download PDFInfo
- Publication number
- CN116211256B CN116211256B CN202310254781.6A CN202310254781A CN116211256B CN 116211256 B CN116211256 B CN 116211256B CN 202310254781 A CN202310254781 A CN 202310254781A CN 116211256 B CN116211256 B CN 116211256B
- Authority
- CN
- China
- Prior art keywords
- respiratory
- breathing
- signal
- state
- sleep
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000029058 respiratory gaseous exchange Effects 0.000 title claims abstract description 342
- 230000007958 sleep Effects 0.000 title claims abstract description 130
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000013145 classification model Methods 0.000 claims abstract description 36
- 238000009423 ventilation Methods 0.000 claims abstract description 36
- 230000008569 process Effects 0.000 claims abstract description 23
- 230000000241 respiratory effect Effects 0.000 claims description 286
- 208000008784 apnea Diseases 0.000 claims description 67
- 230000033001 locomotion Effects 0.000 claims description 44
- 206010021079 Hypopnoea Diseases 0.000 claims description 29
- 238000010586 diagram Methods 0.000 claims description 25
- 238000001914 filtration Methods 0.000 claims description 23
- 238000012549 training Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 15
- 230000009467 reduction Effects 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 11
- 230000004622 sleep time Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000009826 distribution Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 3
- 125000004122 cyclic group Chemical group 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 claims description 3
- 230000036541 health Effects 0.000 abstract description 7
- 238000012544 monitoring process Methods 0.000 abstract description 7
- 230000007774 longterm Effects 0.000 abstract description 4
- 230000009471 action Effects 0.000 description 8
- 238000001514 detection method Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 238000012360 testing method Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 3
- 230000003247 decreasing effect Effects 0.000 description 3
- 201000010099 disease Diseases 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 230000003187 abdominal effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 208000001797 obstructive sleep apnea Diseases 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 230000003860 sleep quality Effects 0.000 description 2
- 208000011580 syndromic disease Diseases 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 206010021133 Hypoventilation Diseases 0.000 description 1
- 206010041235 Snoring Diseases 0.000 description 1
- 206010042434 Sudden death Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000006806 disease prevention Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003434 inspiratory effect Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000008667 sleep stage Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4818—Sleep apnoea
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Abstract
The invention discloses a non-contact sleep breathing signal acquisition method and device, the method comprises the steps of acquiring a vibration signal of sleep breathing, converting the vibration signal into a digital electric signal, extracting a breathing signal corresponding to the sleep breathing from the digital electric signal, determining a plurality of breathing characteristic parameters corresponding to the breathing signal, inputting the breathing characteristic parameters into a preset breathing signal classification model, obtaining a classification result, calculating the times of low ventilation state and the times of breathing pause state in the sleep breathing process, and judging the risk of breathing events in the sleep breathing. By adopting the technical scheme, the signal acquisition module can be arranged in the mattress, so that sleep breathing information of a user can be conveniently acquired in a non-contact mode in daily work and life, long-term continuous monitoring can be realized, a personal health file is established for the user, and the user experience is good; the vibration sensor collects sleep breathing signals, and accuracy and reliability of the collected signals can be improved.
Description
Technical Field
The invention relates to the technical field of vital sign monitoring, in particular to a non-contact sleep breathing signal acquisition method and device.
Background
Along with the development of social and economic technologies and the improvement of living standard, people have higher and higher attention to self health and have higher and higher requirements for daily health monitoring. Respiration is the most basic vital sign of a human body, and the pathological sign of the human body is often reflected on abnormal sleep respiration, so that the realization of daily real-time monitoring of sleep respiration is of great significance to human health assessment and disease prevention.
Sleep apnea hypopnea syndrome refers to the occurrence of apnea (cessation of oral-nasal airflow in a sleep state) and hypopnea (reduction of oral-nasal airflow intensity below 30% of normal value in a sleep state) exceeding a certain range within a specified period of time. The severity of the disease can be assessed by the Apnea Hypopnea Index (AHI), the number of respiratory events occurring per hour. The disease has high incidence rate but is not easy to be perceived, and under the condition that diagnosis cannot be timely confirmed and necessary medical intervention is obtained, normal sleep quality is affected slightly, other serious diseases are induced and even sudden death occurs.
Currently, existing respiratory event detection methods include nasal airflow detection, chest and abdominal strap detection, sleep image recognition, snore detection, and the like. Nasal airflow and chest and abdominal strap detection are relatively accurate, however, due to the need to wear the device on the user, this affects the sleep of the user, and at the same time, the location and degree of fixation of the worn device seriously affects the stability of the physiological signal acquisition; the latter two methods do not need to wear equipment, do not influence the sleeping of a user, but are greatly influenced by the actual sleeping environment, and the detection result is seriously influenced by background color, background light and environmental noise. Therefore, there is an urgent need for a respiratory event detection method that does not affect the sleep of a user and can continuously and reliably work during sleep.
Disclosure of Invention
In view of the foregoing, embodiments of the present invention are directed to providing a non-contact sleep respiratory signal acquisition method and a corresponding non-contact sleep respiratory signal acquisition apparatus that overcome or at least partially solve the foregoing problems.
In order to solve the above problems, on the one hand, an embodiment of the present invention discloses a non-contact sleep respiratory signal acquisition method, which includes:
acquiring a vibration signal of sleep respiration, and converting the vibration signal into a digital electric signal; wherein the vibration signal includes a respiration signal, a body movement signal, a heartbeat signal, and noise;
extracting a respiratory signal corresponding to sleep respiration from the digital electric signal, and obtaining a waveform diagram corresponding to the respiratory signal;
analyzing and calculating the oscillogram through a preset sliding window, and determining a plurality of breathing characteristic parameters corresponding to the breathing signals according to peaks and troughs in the sliding window, wherein the breathing characteristic parameters comprise breathing duration, area under a breathing curve, half-width of the breathing curve, deviation of the breathing curve, respiration peak valley value, inspiration duration, area under the inspiration curve, inspiration average slope, expiration duration, ratio of inspiration to expiration duration and breathing amplitude reduction parameters;
Inputting the breathing characteristic parameters into a preset breathing signal classification model, classifying the breathing signals through the breathing signal classification model, and obtaining a classification result, wherein the classification result comprises a breathing state of each breathing cycle corresponding to the breathing signals, and the breathing state comprises a normal breathing state, a low ventilation state and an apnea state;
and calculating the times of the low ventilation state and the times of the apnea state in the sleeping respiration process, and judging the risk of respiratory events in sleeping according to the times of the low ventilation state and the times of the apnea state.
Optionally, the classifying the respiratory signal by the respiratory signal classification model includes:
marking the respiratory signal with a state label according to the sliding step length, wherein the state label comprises a normal respiratory state label N, a low ventilation state label H and an apnea state label A; wherein, if the sliding window contains an apnea state, the state label is A, if the sliding window does not contain an apnea state but contains a hypopnea state, the state label is H, and if the sliding window does not contain an apnea state and does not contain a hypopnea state, the state label is N;
classifying the continuous respiration signals according to the sliding step length of the sliding window, and further obtaining corresponding continuous respiration state labels;
Labels that continuously label respiratory signals as a or H during sleep breathing are combined.
Optionally, after merging the labels that continuously mark the respiratory signal as a or H during sleep respiration, the method further comprises:
calculating the breathing duration of the combined continuous breathing state labels corresponding to each sliding window;
judging whether the breathing duration of the continuous breathing state label is longer than a preset duration;
if the respiration duration is smaller than the preset duration, correcting and judging the combined continuous respiration state label as a label N of normal respiration;
and if the respiratory time length is greater than or equal to the preset time length, marking a section of respiratory signal corresponding to the combined continuous respiratory state label as a respiratory event, and calculating the respiratory event time length T_event.
Optionally, the preset duration is 10 seconds.
Optionally, the sliding window length is 120 seconds, and the sliding step length is 1 second.
Optionally, extracting a respiratory signal corresponding to sleep respiration from the digital electrical signal includes:
setting a hard threshold in the digital electric signal, combining time-frequency distribution in short-time Fourier transform to identify a body movement signal, and filtering the body movement signal;
And carrying out smooth filtering on the digital electric signal after the body movement signal is filtered, filtering out heartbeat and noise, obtaining a respiratory signal corresponding to sleep respiration, and calculating respiratory characteristic parameters through a preset sliding window.
Optionally, the inputting the respiratory feature parameter into a preset respiratory signal classification model, classifying the respiratory signal by using the respiratory signal classification model, and obtaining a classification result includes:
calculating the breathing duration T_ Bre, the area S_ Bre under a breathing curve, the full width at half maximum FWHM_ Bre, the breathing curve skewness Skew_ Bre, the breathing peak valley PP_ Bre, the inspiration duration T_Inh, the area S_Inh under the inspiration curve, the average slope K_Inh of the inspiration curve, the expiration duration T_Exh, the Ratio of inspiration to expiration duration Ratio In2Ex and the breathing amplitude reduction parameter drop_PP of each breathing cycle In a window;
optionally, the method further comprises: calculating sleep time t_sleep corresponding to sleep respiration, wherein t_sleep=total sleep time-time with all body movement interval time less than 2 minutes-duration of all body movements per se;
the determining the status of the sleep respiratory event according to the low ventilation state times and the apnea state times comprises:
The total number of events with hypopnea is denoted as H_event_sum, and the total number of events with apnea is denoted as A_event_sum;
calculating an apneic-hypopneas index AHI corresponding to sleep breathing;
wherein ahi= (h_event_sum+a_event_sum)/t_sleep;
if AHI is less than 5, sleeping respiratory status is normal;
if the AHI is less than or equal to 5 and less than 15, sleeping and breathing are low risk;
if the AHI is less than or equal to 15 and less than or equal to 30, the risk in sleeping and breathing is reduced;
if AHI is more than or equal to 30, sleeping and breathing are high risk.
On the other hand, the embodiment of the invention discloses a non-contact sleep respiratory signal acquisition device, which comprises:
the signal acquisition module is used for acquiring a vibration signal of sleep respiration and converting the vibration signal into a digital electric signal; wherein the vibration signal includes a respiration signal, a body movement signal, a heartbeat signal, and noise;
the respiratory signal acquisition module is used for extracting respiratory signals corresponding to sleep respiration from the digital electrical signals and obtaining waveform diagrams corresponding to the respiratory signals;
the respiratory signal processing module is used for analyzing and calculating the oscillogram through a preset sliding window, and determining a plurality of respiratory characteristic parameters corresponding to the respiratory signal according to peaks and troughs in the sliding window, wherein the respiratory characteristic parameters comprise respiratory duration, area under a respiratory curve, half-width of the respiratory curve, respiratory curve deflection, respiratory peak valley value, inspiration duration, area under the inspiration curve, inspiration average slope, expiration duration, ratio of inspiration to expiration duration and respiratory amplitude reduction parameters;
The respiratory state classification module is used for inputting the respiratory characteristic parameters into a preset respiratory signal classification model, classifying the respiratory signals through the respiratory signal classification model, and obtaining classification results, wherein the classification results comprise respiratory states of each respiratory cycle corresponding to the respiratory signals, and the respiratory states comprise a normal respiratory state, a low ventilation state and an apnea state;
the sleep respiratory quality judging module is used for calculating the times of the low ventilation state and the times of the apnea state in the sleep respiratory process and judging the risk of respiratory events in sleep according to the times of the low ventilation state and the times of the apnea state.
Optionally, the respiratory state classification module includes:
the first state label marking submodule is used for marking the breathing signal with a state label according to the sliding step length, wherein the state label comprises a normal breathing state label N, a low ventilation state label H and an apnea state label A; wherein, if the sliding window contains an apnea state, the state label is A, if the sliding window does not contain an apnea state but contains a hypopnea state, the state label is H, and if the sliding window does not contain an apnea state and does not contain a hypopnea state, the state label is N;
The second state label marking sub-module is used for classifying the continuous respiration signals according to the sliding step length of the sliding window so as to obtain corresponding continuous respiration state labels;
and the third state label marking submodule is used for combining labels for continuously marking respiratory signals as A or H in the sleeping respiratory process.
Optionally, the apparatus further includes:
the breath time length statistics module is used for calculating the breath time length of the combined continuous breath state labels corresponding to each sliding window;
the breath time length judging module is used for judging whether the breath time length of the continuous breath state label is longer than a preset time length;
the respiration state label correction module is used for correcting and judging the combined continuous respiration state label as a label N of normal respiration if the respiration duration is smaller than a preset duration;
and the respiratory event marking module is used for marking a section of respiratory signal corresponding to the combined continuous respiratory state label as a respiratory event if the respiratory time length is greater than or equal to the preset time length, and calculating the respiratory event time length T_event.
Optionally, the preset duration is 10 seconds.
Optionally, the sliding window length is 120 seconds, and the sliding step length is 1 second.
Optionally, the respiratory signal acquisition module includes:
the body movement signal filtering sub-module is used for setting a hard threshold in the digital electric signal, combining time-frequency distribution in short-time Fourier transformation to identify a body movement signal and filtering the body movement signal;
and the signal filtering sub-module is used for carrying out smooth filtering on the digital electric signal after the body movement signal is filtered, filtering out heartbeat and noise, and obtaining a respiratory signal corresponding to sleep respiration.
Optionally, the respiratory signal processing module is further configured to calculate a respiratory duration t_ Bre, a respiratory curve lower area s_ Bre, a respiratory curve full width at half maximum fwhm_ Bre, a respiratory curve skewness skew_ Bre, a respiratory peak valley pp_ Bre, an inspiration duration t_inh, an inspiratory curve lower area s_inh, an inspiration curve average slope k_inh, an expiration duration t_exh, an inspiration-to-expiration duration Ratio ratio_in2ex, and a respiratory amplitude reduction parameter drop_pp of each respiratory cycle In the window;
optionally, the device further comprises a sleep breathing time calculation module, configured to calculate a sleep time t_sleep corresponding to sleep breathing, where t_sleep=total sleep time-time with all body movements being less than 2 minutes, and duration of all body movements;
The sleep respiratory quality judging module comprises:
the respiratory event statistics sub-module is used for recording the total number of times of occurrence of the H_event as H_event_sum and recording the total number of times of occurrence of the A_event as A_event_sum;
an apnea-hypopnea index calculation sub-module for calculating an apnea-hypopnea index AHI corresponding to sleep breathing;
wherein ahi= (h_event_sum+a_event_sum)/t_sleep;
if AHI is less than 5, sleeping respiratory status is normal;
if the AHI is less than or equal to 5 and less than 15, sleeping and breathing are low risk;
if the AHI is less than or equal to 15 and less than or equal to 30, the risk in sleeping and breathing is reduced;
if AHI is more than or equal to 30, sleeping and breathing are high risk.
In another aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes a processor, a memory, and a computer program stored on the memory and capable of running on the processor, where the computer program when executed by the processor implements the steps of the non-contact sleep breathing signal acquisition method.
In another aspect, embodiments of the present invention also provide a computer readable storage medium having a computer program stored thereon, which when executed by a processor, implements the steps of the non-contact sleep respiratory signal acquisition method.
The embodiment of the invention discloses a non-contact sleep breathing signal acquisition method and a non-contact sleep breathing signal acquisition device, wherein the method comprises the steps of acquiring a vibration signal of sleep breathing and converting the vibration signal into a digital electric signal; wherein the vibration signal comprises a respiration signal, a body movement signal, a heartbeat signal and noise; extracting a respiration signal corresponding to sleep respiration from the digital electrical signal, and obtaining a waveform diagram corresponding to the respiration signal; analyzing and calculating a waveform chart through a preset sliding window, determining a plurality of breathing characteristic parameters corresponding to breathing signals according to peaks and troughs in the sliding window, inputting the breathing characteristic parameters into a preset breathing signal classification model, classifying the breathing signals through the breathing signal classification model, and obtaining classification results, wherein the classification results comprise breathing states of each breathing cycle corresponding to the breathing signals, and the breathing states comprise a normal breathing state, a low ventilation state and an apnea state; and calculating the times of the low ventilation state and the times of the apnea state in the sleeping respiration process, and judging the risk of the respiratory event in the sleeping respiration according to the times of the low ventilation state and the times of the apnea state.
The technical scheme of the invention has the following beneficial effects: the signal acquisition module in the non-contact sleep breathing signal acquisition device can be built in the mattress, so that the sleep breathing information of a user can be acquired in real time in daily work and life, long-term continuous monitoring can be realized, and a personal health file is established for the user; the invention collects sleep breathing signals based on the high-sensitivity vibration sensor, and can improve the accuracy and reliability of the collected signals; the sleep breathing information is acquired in a non-contact mode, and the user experience is good.
Drawings
Fig. 1 is a flowchart illustrating steps of a non-contact sleep respiratory signal acquisition method according to an embodiment of the present invention;
fig. 2 is a block diagram of a non-contact sleep respiratory signal acquisition device according to an embodiment of the present invention;
fig. 3 is a flowchart of a non-contact sleep respiratory signal acquisition method according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a respiratory event determination according to an embodiment of the present invention;
FIG. 5 is a schematic diagram showing a comparison of a respiratory signal of a healthy person with a nasal flow signal in a gold standard Polysomnography (PSG) acquired in accordance with one embodiment of the present invention;
Fig. 6 is a schematic representation of a comparison of respiratory signals acquired during a respiratory event with a patient suffering from an apneic hypoventilation syndrome with nasal airflow signals in a PSG according to one embodiment of the present invention.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Fig. 1 is a flowchart of steps of a non-contact sleep respiratory signal acquisition method according to an embodiment of the present invention, where the method includes the following steps:
step 101, obtaining a vibration signal of sleep respiration, and converting the vibration signal into a digital electric signal; wherein the vibration signal includes a respiration signal, a body movement signal, a heartbeat signal, and noise;
in the embodiment, the sensing system based on the optical fiber sensing principle has the characteristics of high sensitivity, good comfort and electromagnetic interference resistance, and can be used for internally arranging the high-sensitivity vibration sensor in daily necessities such as mattresses or pillows, so that sleeping respiratory signals of users can be collected in daily life scenes. When the mattress is particularly used, taking the mattress as an example, after a human body touches the mattress, the vibration sensor arranged on the mattress is pressed to acquire a body vibration pressure signal, and the pressure signal is converted into a digital electric signal.
Note that the types of vibration sensors include, but are not limited to, piezoelectric, fiber-optic type vibration sensors, and the number includes, but is not limited to, 1 sensor. Because the vibration sensor and the user collect the pressure signal in a non-contact mode, the acquisition of the body vibration pressure signal of the human body can be realized only by the contact between the human body and the mattress, and the problem that the existing wearable product acquisition signal needs to directly contact a test object and restrict the action of the test object to cause inconvenience is solved. For example, the vibration sensor can adopt an optical fiber bending loss type vibration sensor, the signal sampling rate is 256Hz, the sensor is embedded into a mattress, the sensor is paved below a normal mattress, and vital sign signals in the sleeping process of a human body are monitored.
102, extracting a respiratory signal corresponding to sleep respiration from the digital electric signal, and obtaining a waveform diagram corresponding to the respiratory signal;
in extracting the respiratory signal, the user may tag the body movement signal in the digital electrical signal and calculate the body movement duration. Because the vibration signal caused by body movement is far greater than the human vital sign signal, the original signal easily exceeds the measurement range and is represented as a saturated signal, and meanwhile, the frequency components of the signal are not distributed and concentrated on the frequency band. Body movement is easily identified by combining the time-frequency distribution in the short-time Fourier transform in a manner of setting a hard threshold in the signal; the respiratory signal can be extracted by adopting a Savitzky-Golay filtering method to carry out smooth filtering, and high-frequency signals such as heartbeat and noise are filtered, so that the respiratory signal is obtained. The filtering mode has the advantages that the shape and the width of the breathing signal are unchanged, so that the accuracy of the extracted breathing signal is guaranteed, and the accuracy of identification is improved.
In some embodiments, obtaining the respiratory signal from the digital electrical signal includes: filtering the digital electric signal, separating a respiratory signal from the digital electric signal, carrying out smoothing treatment and cyclic spectrum estimation on the respiratory signal through a preset sliding window, judging the signal quality of the respiratory signal through a signal quality coefficient, removing noise components contained in the respiratory signal, and generating a corresponding respiratory waveform.
The filtering process includes removing motion artifacts and heartbeat signals from the digital electrical signal. The digital electric signals are signals processed by the sensor signal acquisition circuit, and comprise baseline drift, heartbeat signals, respiratory signals and body movement artifacts, and the digital electric signals need to be preprocessed to separate the respiratory signals from the digital electric signals. Illustratively, the preprocessing may employ a hard threshold to remove the circuit saturation signal and scale to body motion, with a band pass filter having a cutoff frequency of 0.5-15Hz for the denoising operation.
Step 103, analyzing and calculating the oscillogram through a preset sliding window, and determining a plurality of breathing characteristic parameters corresponding to the breathing signal according to the wave crest and the wave trough in the sliding window, wherein the breathing characteristic parameters comprise breathing duration, area under a breathing curve, half-width of the breathing curve, deflection of the breathing curve, respiration peak valley value, inspiration duration, area under the inspiration curve, inspiration average slope, expiration duration, ratio of inspiration to expiration duration and breathing amplitude descending parameter;
Analyzing and calculating the waveform diagram through a preset sliding window comprises the following steps: a sectional sliding window is made on the breathing signals in the preset time through a sliding window with preset duration, a time-frequency distribution diagram corresponding to the breathing signals in the preset time is obtained, and a plurality of breathing characteristic parameters corresponding to the preset time are obtained according to the time-frequency distribution diagram; calculating a breathing time length T_ Bre, a breathing curve lower area S_ Bre, a breathing curve half-width FWHM_ Bre, a breathing curve skewness Skew_ Bre, a breathing peak valley PP_ Bre, an inspiration time length T_Inh, an inspiration curve lower area S_Inh, an inspiration curve average slope K_Inh, an expiration time length T_Exh, an inspiration and expiration time length Ratio ratio_In2Ex and a breathing amplitude reduction parameter drop_PP of each breathing period In each window corresponding to each sliding window; wherein the respiratory amplitude reduction parameter is expressed as: (current respiration peak valley-previous respiration peak valley)/previous respiration peak valley.
It should be noted that, a person skilled in the art may set the preset time of the respiratory signal and the time window step of the sliding window according to the actual requirement, and the embodiment of the present application does not limit the preset time. Illustratively, the segment sliding window is a 120 second sliding time window in the middle of the data processing window, the time window step being 1 second.
104, inputting the breathing characteristic parameters into a preset breathing signal classification model, classifying the breathing signals through the breathing signal classification model, and obtaining classification results, wherein the classification results comprise breathing states of each breathing cycle corresponding to the breathing signals, and the breathing states comprise a normal breathing state, a low ventilation state and an apnea state;
the preset respiratory signal classification model can be a neural network model with a classification function, respiratory signals can be segmented according to a sliding window in a model training stage, respiratory state labels are marked in a sliding step length time period according to whether respiratory signals in the last preset time period in the sliding window contain hypopnea or apnea, and respiratory signals in the sliding window are used as a training data set, and state labels are used as a training label set to be sent into the established classification model for training. The last preset time length in the sliding window is equal to the sliding step length of the sliding window. The preset respiratory signal classification model is trained according to respiratory characteristic parameters, including but not limited to support vector machines, random forests, decision trees, deep learning and the like.
In the training process of the respiratory signal classification model, the specific change characteristics of the respiratory characteristic parameters when respiratory events occur are as follows: the breathing duration is the duration of a single breath, which is an increasing trend from short to long in time when a respiratory event occurs, and may exceed 10 seconds when an apnea occurs; the area under the breathing curve is the integral of the signal curve in the breathing period, and the trend from large to small is presented when a breathing event occurs; the half-width of a breathing curve is the time interval between half of the peak of the breathing cycle curve, typically exhibiting a trend of increasing time from short to long when a breathing event occurs; the degree of asymmetry of the breathing cycle curve is measured by the degree of deviation of the breathing curve, and the degree of asymmetry generally tends to increase when a breathing event occurs; the respiration peak-valley value measures the respiration depth, the peak-valley value is reduced to below 50% when respiration events occur, and the respiration temporary value can be reduced to below 10% when respiration occurs; the inspiration time is the duration of the inspiration action in the breathing cycle, and the inspiration time shows a gradually extending trend when a breathing event occurs; the area under the inspiration curve is the integral of the breathing signal curve of the inspiration action interval, and the trend of the area from large to small is presented when a breathing event occurs; the average slope of inspiration is the average slope of the breathing signal curve in the inspiration action interval, and the average slope of inspiration shows a trend that the slope gradually becomes smaller when a breathing event occurs; the expiration time is the duration of the expiration action in the breathing cycle, which exhibits a gradual decrease in the expiration time as the breathing event occurs; the ratio of inspiration to expiration is the ratio of inspiration to expiration in the breathing cycle, and the expiration time shows a trend of increasing significantly when a respiratory event occurs; the respiratory amplitude reduction parameter is expressed as: (current respiration peak valley-previous respiration peak valley)/previous respiration peak valley, the respiration amplitude decreasing parameter assumes a continuous negative value before the occurrence of the respiration event, and the respiration amplitude decreasing parameter may be set to a range of-0.3 to-0.1, for example. The sliding window length may be set to 120 seconds and the sliding step length may be set to 1 second. It should be noted that, a person skilled in the art may set the sliding window length and the sliding step according to actual needs, and the specific numerical values of the sliding window length and the sliding step are not limited in this embodiment.
In the training process of the respiratory signal classification model, the output result of the respiratory signal classification model can be compared with the manually marked result until the output result of the respiratory signal classification model meets the user index of the user. After the respiratory signal classification model is trained, classifying the intra-window signals of the respiratory signals according to the extracted characteristic parameters by adopting a learned respiratory signal classification model, wherein the intra-window signals are respectively in a normal respiratory state (N), a low ventilation state (H) and an apnea state (A), namely, labeling the respiratory signals according to a sliding step length; if the window contains an apnea, the state label is A, if the window does not contain an apnea but contains a hypopnea, the state label is H, otherwise, the state label is N.
After a plurality of peaks and troughs corresponding to the respiratory signals are determined according to the waveform diagram, respiratory cycle and respiratory amplitude are calculated. Peaks and valleys of the respiration waveform are located. Inhalation from peak to valley and exhalation from valley to peak. The trough is used to calculate the duration of a single breath. From this, the duration Bre _t, the inspiration duration inh_t, the expiration duration exh_t, the peak-to-valley Bre _pp of a single breath can be calculated;
and judging the respiration speed. The current respiration rate rr=60/Bre _t can be calculated by the duration of a single breath. For adults, the normal value of RR is 12-20 times per minute. Thus RR was set to be slow breathing below 10 times/min and fast breathing above 24 times/min. It should be noted that, the specific value of the respiration rate RR may be set according to the actual needs of those skilled in the art, and the specific value of the respiration rate is not specifically limited in this application.
Step 105, calculating the times of low ventilation state and the times of apnea state in the sleeping respiratory process, and judging the risk of respiratory events in sleeping according to the times of low ventilation state and the times of apnea state.
The sleep time was calculated as follows: because the embodiment of the invention is not assisted by the electroencephalogram signals, the judgment of the sleep stage is difficult. And the calculation of the AHI index requires the calculation of sleep time. Since there are dense body movement signals when the person wakes up, the present invention discriminates sleep time by detecting body movement. If the interval between two successive body movements is less than 2 minutes, the period of time is counted as awake time, and all body movement durations are counted as awake time. Finally, sleep time t_sleep = total time-awake time. Wherein T sleep = total sleep time-time with all body movements interval less than 2 minutes-duration of all body movements themselves;
and calculating an AHI index. The total number of occurrences of the H event is denoted h_event_sum and the total number of occurrences of the a event is denoted a_event_sum. Ahi= (h_event_sum+a_event_sum)/t_sleep. Judging the risk of respiratory events in sleep breathing according to the times of the low ventilation state and the times of the apnea state, wherein the method comprises the following steps of:
The total number of times of occurrence of the H_event event is recorded as H_event_sum, and the total number of times of occurrence of the A_event is recorded as A_event_sum; calculating an apneic-hypopneas index AHI corresponding to sleep breathing; wherein ahi= (h_event_sum+a_event_sum)/t_sleep; if AHI is less than 5, the sleep respiration corresponding to sleep respiration is normal; if the AHI is less than or equal to 5 and less than or equal to 15, sleeping respiration corresponding to sleeping respiration is at low risk; if the AHI is less than or equal to 15 and less than or equal to 30, the risk in sleep breathing corresponding to the sleep breathing is reduced; if the AHI is more than or equal to 30, the sleep respiration corresponding to the sleep respiration is at high risk. Illustratively, the AHI index may be used to measure the severity of sleep apnea hypopnea syndrome, and is divided into normal (AHI < 5), mild (5. Ltoreq. AHI < 15), moderate (15. Ltoreq. AHI < 30), and severe (AHI. Ltoreq.30).
The embodiment of the invention discloses a non-contact sleep breathing signal acquisition method, which comprises the steps of acquiring a vibration signal of sleep breathing and converting the vibration signal into a digital electric signal; wherein the vibration signal comprises a respiration signal, a body movement signal, a heartbeat signal and noise; extracting a respiration signal corresponding to sleep respiration from the digital electrical signal, and obtaining a waveform diagram corresponding to the respiration signal; analyzing and calculating a waveform chart through a preset sliding window, determining a plurality of breathing characteristic parameters corresponding to breathing signals according to peaks and troughs in the sliding window, inputting the breathing characteristic parameters into a preset breathing signal classification model, classifying the breathing signals through the breathing signal classification model, and obtaining classification results, wherein the classification results comprise breathing states of each breathing cycle corresponding to the breathing signals, and the breathing states comprise a normal breathing state, a low ventilation state and an apnea state; and calculating the times of the low ventilation state and the times of the apnea state in the sleeping respiration process, and judging the risk of the respiratory event in the sleeping respiration according to the times of the low ventilation state and the times of the apnea state. By adopting the technical scheme, the signal acquisition module in the non-contact sleep breathing signal acquisition device can be arranged in the mattress, so that the sleep breathing information of a user can be acquired in real time in daily work and life, long-term continuous monitoring can be realized, and a personal health file is established for the user; the invention collects sleep breathing signals based on the high-sensitivity vibration sensor, and can improve the accuracy and reliability of the collected signals; the sleep breathing information is acquired in a non-contact mode, and the user experience is good.
It should be noted that, for simplicity of description, the method embodiments are shown as a series of acts, but it should be understood by those skilled in the art that the embodiments are not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred embodiments, and that the acts are not necessarily required by the embodiments of the invention.
In order to implement the above-mentioned non-contact sleep respiratory signal acquisition method, fig. 2 is a block diagram of a non-contact sleep respiratory signal acquisition device according to an embodiment of the present invention, where the device includes:
the signal acquisition module 201 is configured to acquire a vibration signal of sleep respiration, and convert the vibration signal into a digital electrical signal; wherein the vibration signal includes a respiration signal, a body movement signal, a heartbeat signal, and noise;
a respiratory signal acquisition module 202, configured to extract a respiratory signal corresponding to sleep respiration from the digital electrical signal, and obtain a waveform diagram corresponding to the respiratory signal;
The respiratory signal processing module 203 is configured to analyze and calculate the waveform map through a preset sliding window, and determine a plurality of respiratory characteristic parameters corresponding to the respiratory signal according to peaks and troughs in the sliding window, where the respiratory characteristic parameters include respiratory duration, area under a respiratory curve, half-width of the respiratory curve, deviation of the respiratory curve, peak valley of respiration, inspiration duration, area under the inspiration curve, average slope of inspiration, expiration duration, ratio of inspiration to expiration duration, and respiratory amplitude decreasing parameter;
a respiratory state classification module 204, configured to input the respiratory feature parameter into a preset respiratory signal classification model, classify the respiratory signal by using the respiratory signal classification model, and obtain a classification result, where the classification result includes a respiratory state of each respiratory cycle corresponding to the respiratory signal, and the respiratory state includes a normal respiratory state, a hypopnea state, and an apnea state;
the sleep respiratory quality judging module 205 is configured to calculate the number of times of low ventilation status and the number of times of apnea status in the sleep respiratory process, and judge the risk of respiratory events in the sleep respiratory according to the number of times of low ventilation status and the number of times of apnea status.
In an alternative embodiment, the respiratory state classification module includes:
the first state label marking submodule is used for marking the breathing signal with a state label according to the sliding step length, wherein the state label comprises a normal breathing state label N, a low ventilation state label H and an apnea state label A; wherein, if the sliding window contains an apnea state, the state label is A, if the sliding window does not contain an apnea state but contains a hypopnea state, the state label is H, and if the sliding window does not contain an apnea state and does not contain a hypopnea state, the state label is N;
the second state label marking sub-module is used for classifying the continuous respiration signals according to the sliding step length of the sliding window so as to obtain corresponding continuous respiration state labels;
and the third state label marking submodule is used for combining labels for continuously marking respiratory signals as A or H in the sleeping respiratory process.
In an alternative embodiment, the apparatus further comprises:
the breath time length statistics module is used for calculating the breath time length of the combined continuous breath state labels corresponding to each sliding window;
the breath time length judging module is used for judging whether the breath time length of the continuous breath state label is longer than a preset time length;
The respiration state label correction module is used for correcting and judging the combined continuous respiration state label as a label N of normal respiration if the respiration duration is smaller than a preset duration;
and the respiratory event marking module is used for marking a section of respiratory signal corresponding to the combined continuous respiratory state label as a respiratory event if the respiratory time length is greater than or equal to the preset time length, and calculating the respiratory event time length T_event.
In an alternative embodiment, the predetermined duration is 10 seconds.
In an alternative embodiment, the sliding window length is 120 seconds and the sliding step length is 1 second.
Optionally, the respiratory signal acquisition module includes:
the body movement signal filtering sub-module is used for setting a hard threshold in the digital electric signal, combining time-frequency distribution in short-time Fourier transformation to identify a body movement signal and filtering the body movement signal;
and the signal filtering sub-module is used for carrying out smooth filtering on the digital electric signal after the body movement signal is filtered, filtering out heartbeat and noise, and obtaining a respiratory signal corresponding to sleep respiration.
In an alternative embodiment, the respiratory signal processing module 203 may include:
in an alternative embodiment, the device further comprises a sleep time calculation module for calculating a sleep time T sleep corresponding to sleep breathing,
The sleep quality breath judging module 205 may include:
the respiratory event statistics sub-module is used for recording the total number of times of occurrence of the H_event as H_event_sum and recording the total number of times of occurrence of the A_event as A_event_sum;
an apnea-hypopnea index calculation sub-module for calculating an apnea-hypopnea index AHI corresponding to sleep breathing;
wherein ahi= (h_event_sum+a_event_sum)/t_sleep;
if AHI is less than 5, the sleep respiration corresponding to sleep respiration is normal; if the AHI is less than or equal to 5 and less than or equal to 15, sleeping respiration corresponding to sleeping respiration is at low risk; if the AHI is less than or equal to 15 and less than or equal to 30, the risk in sleep breathing corresponding to the sleep breathing is reduced; if the AHI is more than or equal to 30, the sleep respiration corresponding to the sleep respiration is at high risk.
The embodiment of the invention discloses a non-contact sleep breathing signal acquisition method and a non-contact sleep breathing signal acquisition device, wherein the method comprises the steps of acquiring a vibration signal of sleep breathing and converting the vibration signal into a digital electric signal; wherein the vibration signal comprises a respiration signal, a body movement signal, a heartbeat signal and noise; extracting a respiration signal corresponding to sleep respiration from the digital electrical signal, and obtaining a waveform diagram corresponding to the respiration signal; analyzing and calculating a waveform chart through a preset sliding window, determining a plurality of breathing characteristic parameters corresponding to breathing signals according to peaks and troughs in the sliding window, inputting the breathing characteristic parameters into a preset breathing signal classification model, classifying the breathing signals through the breathing signal classification model, and obtaining classification results, wherein the classification results comprise breathing states of each breathing cycle corresponding to the breathing signals, and the breathing states comprise a normal breathing state, a low ventilation state and an apnea state; and calculating the times of the low ventilation state and the times of the apnea state in the sleeping respiration process, and judging the risk of the respiratory event in the sleeping respiration according to the times of the low ventilation state and the times of the apnea state. By adopting the technical scheme, the signal acquisition module in the non-contact sleep breathing signal acquisition device can be arranged in the mattress, so that the sleep breathing information of a user can be acquired in real time in daily work and life, long-term continuous monitoring can be realized, and a personal health file is established for the user; the invention collects sleep breathing signals based on the high-sensitivity vibration sensor, and can improve the accuracy and reliability of the collected signals; the sleep breathing information is acquired in a non-contact mode, and the user experience is good.
In some embodiments, fig. 3 is a flowchart of a non-contact sleep respiratory signal acquisition method according to an embodiment of the present invention, and fig. 4 is a schematic diagram of respiratory event determination according to an embodiment of the present invention. The sleep respiratory signal acquisition method adopts a support vector machine as a learned classifier to classify the sliding window into 3 classes, and distinguishes normal respiration, apnea and hypopnea. Because the respiratory signal is one-dimensional data, and the time sequence of the respiratory signal is directly learned and classified, the satisfactory effect is difficult to obtain. The specific implementation process comprises the following steps:
building a training data set: dividing a known respiratory signal into signals with equal length of a sliding window, and calculating corresponding respiratory characteristic parameters to form a corresponding data set X= { X 1 ，……,x n -comprising n samples in K-dimensional space. Alternatively, in this embodiment, k=11. The trained personnel manually marks the corresponding labels (N, A and H) of the samples in X according to the corresponding respiratory signals, and marks the labels as Y= { Y 1 ，……，y n }。
Training data: the support vector machine trains a plurality of hyperplanes w from the data set T x+b=0 separates three types of data, where w is a K-dimensional vector and b is a scalar. The training aim is to find a plurality of optimal separation hyperplanes, reduce the error rate of data classification as much as possible, and obtain an optimal solution w * And b * 。
And (3) data verification: the optimal solution w obtained in the training stage * And b * Converting into a separation hyperplane. For verification sample X test By a decision function Y predict ＝sign(w * x test +b * ) To classify it.
For a segment of respiratory signals to be classified, the respiratory characteristic parameter composition verification sample X is calculated through a sliding window test Obtaining a corresponding classification result Y predict As a respiration status label for the window. Thus, by sliding the sliding window, a continuous breath state label (N, a and H) is obtained.
FIG. 5 is an example of a comparison of respiratory signals acquired by an embodiment of the present invention for a healthy person with a gold standard, a nasal airflow signal in a Polysomnography (PSG), which is the primary reference basis for a hospital PSG to determine respiratory events. The respiratory signal collected by the invention is shown above in fig. 5, and the nasal airflow signal of the PSG is shown below in fig. 5. It can be seen that in normal breathing, the breathing cycle is uniform and the breathing amplitude is stable. Although the method for detecting respiration is different, the respiration waveforms of the two respiration waveforms are very close, and the respiration period and the respiration amplitude have a very good corresponding relationship, which is the key for judging respiratory events.
Fig. 6 is an example of a comparison of respiratory signals acquired by an embodiment of the present invention when a respiratory event occurs in a patient with an apneic hypopneas syndrome with a nasal airflow signal in a PSG, one of which is depicted as an a_event and an h_event. The respiratory signal collected by the invention is shown above in fig. 6, and the nasal airflow signal of the PSG is shown below in fig. 6. It can be seen that when a respiratory event occurs, a significant change in the cadence of respiration occurs in the respiratory signal, the most prominent change being a significant decrease or even disappearance of the respiratory wave in the amplitude of the respiration. The two contrast signals undergo nearly identical synchronous changes, which indicates that the present invention is effective for detection of respiratory events.
In another aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes a processor, a memory, and a computer program stored on the memory and capable of running on the processor, where the computer program when executed by the processor implements the steps of the non-contact sleep breathing signal acquisition method.
In another aspect, embodiments of the present invention also provide a computer readable storage medium having a computer program stored thereon, which when executed by a processor, implements the steps of the non-contact sleep respiratory signal acquisition method.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.
It will be apparent to those skilled in the art that embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
The foregoing describes a non-contact sleep respiratory signal acquisition method and a non-contact sleep respiratory signal acquisition device according to the present invention in detail, and specific examples are provided herein to illustrate the principles and embodiments of the present invention, and in light of the above-described general knowledge of one skilled in the art, the present invention should not be construed as limited to the embodiments and application ranges.
Claims (6)
1. A method for acquiring a non-contact sleep respiratory signal, comprising:
acquiring a vibration signal of sleep respiration, and converting the vibration signal into a digital electric signal; wherein the vibration signal includes a respiration signal, a body movement signal, a heartbeat signal, and noise;
extracting a respiratory signal corresponding to sleep respiration from the digital electric signal, and obtaining a waveform diagram corresponding to the respiratory signal, wherein the waveform diagram comprises the following steps:
filtering the digital electric signal, separating a respiratory signal from the digital electric signal, carrying out smoothing treatment and cyclic spectrum estimation on the respiratory signal through a preset sliding window, judging the signal quality of the respiratory signal through a signal quality coefficient, removing noise components contained in the respiratory signal, and generating a corresponding respiratory waveform;
analyzing and calculating the oscillogram through a preset sliding window, and determining a plurality of breathing characteristic parameters corresponding to the breathing signals according to peaks and troughs in the sliding window, wherein the breathing characteristic parameters comprise breathing duration, area under a breathing curve, half-width of the breathing curve, deviation of the breathing curve, respiration peak valley value, inspiration duration, area under the inspiration curve, inspiration average slope, expiration duration, ratio of inspiration to expiration duration and breathing amplitude reduction parameters; the sliding window is 120 seconds long, and the sliding step length is 1 second;
Calculating the breathing duration T_ Bre, the area S_ Bre under a breathing curve, the full width at half maximum FWHM_ Bre, the breathing curve skewness Skew_ Bre, the breathing peak valley PP_ Bre, the inspiration duration T_Inh, the area S_Inh under the inspiration curve, the average slope K_Inh of the inspiration curve, the expiration duration T_Exh, the Ratio of inspiration to expiration duration Ratio In2Ex and the breathing amplitude reduction parameter drop_PP of each breathing cycle In a window;
inputting the breathing characteristic parameters into a preset breathing signal classification model, classifying the breathing signals through the breathing signal classification model, and obtaining a classification result, wherein the classification result comprises a breathing state of each breathing cycle corresponding to the breathing signals, and the breathing state comprises a normal breathing state, a low ventilation state and an apnea state;
the respiratory signal classification model is trained according to respiratory characteristic parameters, respiratory signals are segmented according to sliding windows, whether respiratory signals contain hypopnea or apnea is judged according to whether respiratory signals in the last preset time period in the sliding windows are labeled by respiratory state labels, the respiratory signals in the sliding windows are used as training data sets, the state labels are used as training label sets and sent into the established classification model for training, and the last preset time period in the sliding windows is equal to the sliding step length of the sliding windows;
The respiratory signal classification model classifies the respiratory signal, including:
marking the respiratory signal with a state label according to the sliding step length, wherein the state label comprises a normal respiratory state label N, a low ventilation state label H and an apnea state label A; wherein, if the sliding window contains an apnea state, the state label is A, if the sliding window does not contain an apnea state but contains a hypopnea state, the state label is H, and if the sliding window does not contain an apnea state and does not contain a hypopnea state, the state label is N;
classifying the continuous respiration signals according to the sliding step length of the sliding window, and further obtaining corresponding continuous respiration state labels;
combining labels with respiratory signals continuously marked as A or H in the sleeping respiratory process;
after combining the labels that continuously label the respiratory signal as a or H during sleep respiration, the method further comprises:
calculating the breathing duration of the combined continuous breathing state labels corresponding to each sliding window;
judging whether the breathing duration of the continuous breathing state label is longer than a preset duration;
if the respiration duration is smaller than the preset duration, correcting and judging the combined continuous respiration state label as a label N of normal respiration;
If the respiration time length is greater than or equal to the preset time length, marking a section of respiration signal corresponding to the combined continuous respiration state label as a respiration event, and calculating the respiration event time length T_event;
calculating the times of low ventilation state and the times of apnea state in the sleeping respiration process, and judging the risk of respiratory events in sleeping according to the times of low ventilation state and the times of apnea state, wherein the method comprises the following steps:
the sleep time T sleep corresponding to sleep breath is calculated,
wherein T sleep = total sleep time-time with all body movements interval less than 2 minutes-duration of all body movements themselves;
if the respiratory event comprises an A label, the respiratory event is marked as A_event; otherwise, marking as H_event, judging the risk of respiratory events in sleep according to the times of the low ventilation state and the times of the apnea state, wherein the method comprises the following steps of:
the total number of times of occurrence of the H_event event is recorded as H_event_sum, and the total number of times of occurrence of the A_event is recorded as A_event_sum;
calculating an apneic-hypopneas index AHI corresponding to sleep breathing;
wherein ahi= (h_event_sum+a_event_sum)/t_sleep;
if AHI is less than 5, sleeping respiratory status is normal;
If the AHI is less than or equal to 5 and less than 15, sleeping and breathing are low risk;
if the AHI is less than or equal to 15 and less than or equal to 30, the risk in sleeping and breathing is reduced;
if AHI is more than or equal to 30, sleeping and breathing are high risk.
2. The method of claim 1, wherein the predetermined duration is 10 seconds.
3. The method of claim 1, wherein extracting a respiratory signal corresponding to sleep breathing from the digital electrical signal comprises:
setting a hard threshold in the digital electric signal, combining time-frequency distribution in short-time Fourier transform to identify a body movement signal, and filtering the body movement signal;
and carrying out smooth filtering on the digital electric signal after the body movement signal is filtered, filtering out heartbeat and noise, obtaining a respiratory signal corresponding to sleep respiration, and calculating respiratory characteristic parameters through a preset sliding window.
4. A non-contact sleep disordered breathing signal acquisition device, comprising:
the signal acquisition module is used for acquiring a vibration signal of sleep respiration and converting the vibration signal into a digital electric signal; wherein the vibration signal includes a respiration signal, a body movement signal, a heartbeat signal, and noise;
the respiratory signal acquisition module is used for extracting respiratory signals corresponding to sleep respiration from the digital electrical signals and obtaining waveform diagrams corresponding to the respiratory signals, and comprises the following steps:
Filtering the digital electric signal, separating a respiratory signal from the digital electric signal, carrying out smoothing treatment and cyclic spectrum estimation on the respiratory signal through a preset sliding window, judging the signal quality of the respiratory signal through a signal quality coefficient, removing noise components contained in the respiratory signal, and generating a corresponding respiratory waveform;
the respiratory signal processing module is used for analyzing and calculating the oscillogram through a preset sliding window, and determining a plurality of respiratory characteristic parameters corresponding to the respiratory signal according to peaks and troughs in the sliding window, wherein the respiratory characteristic parameters comprise respiratory duration, area under a respiratory curve, half-width of the respiratory curve, respiratory curve deflection, respiratory peak valley value, inspiration duration, area under the inspiration curve, inspiration average slope, expiration duration, ratio of inspiration to expiration duration and respiratory amplitude reduction parameters; the sliding window is 120 seconds long, and the sliding step length is 1 second;
the respiratory signal processing module is further used for calculating respiratory duration T_ Bre, area S_ Bre under a respiratory curve, respiratory curve full width at half maximum FWHM_ Bre, respiratory curve skewness Shew_ Bre, respiratory peak valley PP_ Bre, inspiration duration T_Inh, area S_Inh under an inspiration curve, average slope K_Inh of the inspiration curve, expiration duration T_Exh, ratio of inspiration to expiration duration Ratio Ratio_In2Ex and respiratory amplitude reduction parameter drop_PP of each respiratory cycle In the window;
The respiratory state classification module is used for inputting the respiratory characteristic parameters into a preset respiratory signal classification model, classifying the respiratory signals through the respiratory signal classification model, and obtaining classification results, wherein the classification results comprise respiratory states of each respiratory cycle corresponding to the respiratory signals, and the respiratory states comprise a normal respiratory state, a low ventilation state and an apnea state;
the respiratory signal classification model is trained according to respiratory characteristic parameters, respiratory signals are segmented according to sliding windows, whether respiratory signals contain hypopnea or apnea is judged according to whether respiratory signals in the last preset time period in the sliding windows are labeled by respiratory state labels, the respiratory signals in the sliding windows are used as training data sets, the state labels are used as training label sets and sent into the established classification model for training, and the last preset time period in the sliding windows is equal to the sliding step length of the sliding windows;
the respiratory state classification module includes:
the first state label marking submodule is used for marking the breathing signal with a state label according to the sliding step length, wherein the state label comprises a normal breathing state label N, a low ventilation state label H and an apnea state label A; wherein, if the sliding window contains an apnea state, the state label is A, if the sliding window does not contain an apnea state but contains a hypopnea state, the state label is H, and if the sliding window does not contain an apnea state and does not contain a hypopnea state, the state label is N;
The second state label marking sub-module is used for classifying the continuous respiration signals according to the sliding step length of the sliding window so as to obtain corresponding continuous respiration state labels;
a third state label marking sub-module for combining labels continuously marking the respiratory signal as A or H in the sleeping respiratory process;
the apparatus further comprises:
the breath time length statistics module is used for calculating the breath time length of the combined continuous breath state labels corresponding to each sliding window;
the breath time length judging module is used for judging whether the breath time length of the continuous breath state label is longer than a preset time length;
the respiration state label correction module is used for correcting and judging the combined continuous respiration state label as a label N of normal respiration if the respiration duration is smaller than a preset duration;
the respiratory event marking module is used for marking a section of respiratory signal corresponding to the combined continuous respiratory state label as a respiratory event if the respiratory time length is greater than or equal to a preset time length, and calculating a respiratory event time length T_event;
the sleep respiratory quality judging module is used for calculating the times of the low ventilation state and the times of the apnea state in the sleep respiratory process and judging the risk of respiratory events in sleep according to the times of the low ventilation state and the times of the apnea state;
The device also comprises a sleep breathing time calculation module, which is used for calculating sleep time T_sleep corresponding to sleep breathing, wherein T_sleep = total sleep time-time with all body movement interval time less than 2 minutes-duration of all body movements per se;
the sleep respiratory quality judging module comprises:
the respiratory event statistics sub-module is used for recording the total number of times of occurrence of the H_event as H_event_sum and recording the total number of times of occurrence of the A_event as A_event_sum;
an apnea-hypopnea index calculation sub-module for calculating an apnea-hypopnea index AHI corresponding to sleep breathing;
wherein ahi= (h_event_sum+a_event_sum)/t_sleep;
if AHI is less than 5, sleeping respiratory status is normal;
if the AHI is less than or equal to 5 and less than 15, sleeping and breathing are low risk;
if the AHI is less than or equal to 15 and less than or equal to 30, the risk in sleeping and breathing is reduced;
if AHI is more than or equal to 30, sleeping and breathing are high risk.
5. An electronic device, comprising: a processor, a memory and a computer program stored on the memory and capable of running on the processor, which when executed by the processor performs the steps of the method according to any of claims 1-3.
6. A computer readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, implements the steps of the method according to any of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310254781.6A CN116211256B (en) | 2023-03-16 | 2023-03-16 | Non-contact sleep breathing signal acquisition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310254781.6A CN116211256B (en) | 2023-03-16 | 2023-03-16 | Non-contact sleep breathing signal acquisition method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116211256A CN116211256A (en) | 2023-06-06 |
| CN116211256B true CN116211256B (en) | 2023-12-22 |
Family
ID=86591015
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310254781.6A Active CN116211256B (en) | 2023-03-16 | 2023-03-16 | Non-contact sleep breathing signal acquisition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116211256B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116343997A (en) * | 2023-04-03 | 2023-06-27 | 广州柏曼光电科技有限公司 | Sleep-aiding method and device based on sleep state |
| CN117747117A (en) * | 2024-02-21 | 2024-03-22 | 安徽星辰智跃科技有限责任公司 | Sound-based sleep respiration evaluation and auxiliary adjustment method, system and device |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1736324A (en) * | 2005-07-26 | 2006-02-22 | 中国人民解放军空军航空医学研究所 | Detection/disposal method and apparatus for obtaining respiratory disturbance index information |
| CN103263261A (en) * | 2013-05-02 | 2013-08-28 | 北京博实联创科技有限公司 | System for presuming sleep indexes and sleep stages by method for measuring physiological parameters in unconstrained manner |
| CN103892797A (en) * | 2012-12-31 | 2014-07-02 | 中国移动通信集团公司 | Signal processing method and device for sleep structure analysis |
| CN109480783A (en) * | 2018-12-20 | 2019-03-19 | 深圳和而泰智能控制股份有限公司 | A kind of apnea detection method, apparatus and calculate equipment |
| CN109620208A (en) * | 2018-12-29 | 2019-04-16 | 南京茂森电子技术有限公司 | Sleep Apnea-hypopnea Syndrome detection system and method |
| CN210228129U (en) * | 2018-12-29 | 2020-04-03 | 南京茂森电子技术有限公司 | Sleep apnea hypopnea syndrome detection device |
| CN111543942A (en) * | 2020-04-02 | 2020-08-18 | 南京润楠医疗电子研究院有限公司 | Classification and identification device and method for sleep apnea hypopnea event |
| CN113710151A (en) * | 2018-11-19 | 2021-11-26 | 瑞思迈传感器技术有限公司 | Method and apparatus for detecting breathing disorders |
| CN115040074A (en) * | 2022-05-06 | 2022-09-13 | 清华大学深圳国际研究生院 | Obstructive sleep apnea detection method and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015089274A1 (en) * | 2013-12-11 | 2015-06-18 | Oregon Health & Science University | Method and apparatus for assessment of sleep apnea |
| US20210345891A1 (en) * | 2020-05-08 | 2021-11-11 | Pacesetter, Inc. | Method and device for detecting respiration anomaly from low frequency component of electrical cardiac activity signals |
-
2023
- 2023-03-16 CN CN202310254781.6A patent/CN116211256B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1736324A (en) * | 2005-07-26 | 2006-02-22 | 中国人民解放军空军航空医学研究所 | Detection/disposal method and apparatus for obtaining respiratory disturbance index information |
| CN103892797A (en) * | 2012-12-31 | 2014-07-02 | 中国移动通信集团公司 | Signal processing method and device for sleep structure analysis |
| CN103263261A (en) * | 2013-05-02 | 2013-08-28 | 北京博实联创科技有限公司 | System for presuming sleep indexes and sleep stages by method for measuring physiological parameters in unconstrained manner |
| CN113710151A (en) * | 2018-11-19 | 2021-11-26 | 瑞思迈传感器技术有限公司 | Method and apparatus for detecting breathing disorders |
| CN109480783A (en) * | 2018-12-20 | 2019-03-19 | 深圳和而泰智能控制股份有限公司 | A kind of apnea detection method, apparatus and calculate equipment |
| CN109620208A (en) * | 2018-12-29 | 2019-04-16 | 南京茂森电子技术有限公司 | Sleep Apnea-hypopnea Syndrome detection system and method |
| CN210228129U (en) * | 2018-12-29 | 2020-04-03 | 南京茂森电子技术有限公司 | Sleep apnea hypopnea syndrome detection device |
| CN111543942A (en) * | 2020-04-02 | 2020-08-18 | 南京润楠医疗电子研究院有限公司 | Classification and identification device and method for sleep apnea hypopnea event |
| CN115040074A (en) * | 2022-05-06 | 2022-09-13 | 清华大学深圳国际研究生院 | Obstructive sleep apnea detection method and device |
Non-Patent Citations (1)
| Title |
|---|
| "Expiratory Time Constant and Sleep Apnea Severity in the Overlap Syndrome";Darunee Wiriyaporn;《Journal or clinical sleep medicine》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116211256A (en) | 2023-06-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116211256B (en) | Non-contact sleep breathing signal acquisition method and device | |
| US11751817B2 (en) | Breathing disorder identification, characterization and diagnosis methods, devices and systems | |
| US10278639B2 (en) | Method and system for sleep detection | |
| US7559903B2 (en) | Breathing sound analysis for detection of sleep apnea/popnea events | |
| JP5628147B2 (en) | Sleep / wake state evaluation method and system | |
| US9833184B2 (en) | Identification of emotional states using physiological responses | |
| EP2709525B1 (en) | Apnea and hypopnea detection using breath pattern recognition | |
| Almazaydeh et al. | A neural network system for detection of obstructive sleep apnea through SpO2 signal features | |
| US20160045161A1 (en) | Mask and method for breathing disorder identification, characterization and/or diagnosis | |
| JP7104076B2 (en) | Methods and devices for determining sleep statistics | |
| US20120071741A1 (en) | Sleep apnea monitoring and diagnosis based on pulse oximetery and tracheal sound signals | |
| US20140213937A1 (en) | Systems and methods to monitor and quantify physiological stages | |
| AU2012255590A1 (en) | OSA/CSA diagnosis using recorded breath sound amplitude profile and pitch contour | |
| CN115209801A (en) | Sleep apnea detection system and method | |
| CN112155560B (en) | Apnea detection method and system based on real-time cardiac shock signal | |
| Hwang et al. | Sleep period time estimation based on electrodermal activity | |
| Gerla et al. | Multivariate analysis of full-term neonatal polysomnographic data | |
| Espiritu et al. | Automated detection of sleep disorder-related events from polysomnographic data | |
| CA2585824A1 (en) | Breathing sound analysis for detection of sleep apnea/hypopnea events | |
| Ferri et al. | Quantifying leg movement activity during sleep | |
| JP2020151358A (en) | Sleep apnea syndrome determination apparatus, sleep apnea syndrome determination method, and sleep apnea syndrome determination program | |
| Guul et al. | Portable prescreening system for sleep apnea | |
| US20230320655A1 (en) | Systems and Methods for Using Triaxial Accelerometer Data for Sleep Monitoring | |
| Aittokallio et al. | Detection of high-frequency respiratory movements during sleep | |
| WO2024023212A1 (en) | Monitoring sleeping disorders and system thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |