CN109259733A - Apnea detection method, apparatus and detection device in a kind of sleep - Google Patents
Apnea detection method, apparatus and detection device in a kind of sleep Download PDFInfo
- Publication number
- CN109259733A CN109259733A CN201811252525.9A CN201811252525A CN109259733A CN 109259733 A CN109259733 A CN 109259733A CN 201811252525 A CN201811252525 A CN 201811252525A CN 109259733 A CN109259733 A CN 109259733A
- Authority
- CN
- China
- Prior art keywords
- blood oxygen
- oxygen saturation
- sleep
- preset duration
- feature
- 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.)
- Pending
Links
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
- A61B5/02—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
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/02405—Determining heart rate variability
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14542—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring blood gases
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7282—Event detection, e.g. detecting unique waveforms indicative of a medical condition
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Public Health (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Cardiology (AREA)
- Physiology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Pulmonology (AREA)
- Optics & Photonics (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present embodiments relate to apnea detection method, apparatus and detection devices in a kind of sleep, which comprises obtains heart operation sampled data and the blood oxygen saturation sampled data in preset duration；Feature extraction is carried out based on heart operation sampled data, obtains the heart operation characteristic in preset duration；Feature extraction is carried out based on blood oxygen saturation sampled data, obtains the blood oxygen saturation feature in preset duration；To the heart operation characteristic and blood oxygen saturation feature progress Fusion Features in same preset duration, sleep characteristics are obtained；The sleep characteristics are identified based on default sleep apnea model, whether sleep apnea occurs in the determination preset duration.The embodiment of the present invention has merged the sleep characteristics of heart operation characteristic and blood oxygen saturation feature by obtaining, and is modeled and identified using sleep characteristics, and recognition accuracy is high.And medical supply and medical practitioner assessment without profession, it is easy to use, be suitable for home use.
Description
Technical field
The present embodiments relate to a kind of apnea detection method in physiologic information detection technique more particularly to sleep,
Device and detection device.
Background technique
Sleep apnea syndrome (sleep apnea syndrome, SAS) is a kind of sleeping for sleep apnea
Dormancy obstacle, SAS seriously threaten the health of the mankind, are easy to cause multiple complications even to cause overworked dead.The detection of SAS is more at present
It is detected using polysomnogram (Polysolnogram, PSG) method, still, polysomnogram method needs patient to wear for a long time
It wears expensive Medical Devices to be monitored, and needs to be equipped with professional and diagnose, it is inconvenient to use, it is difficult to popularize and answer
With.
Summary of the invention
One purpose of the embodiment of the present invention is to provide a kind of easy to use, temporary conducive to breathing in universal and application sleep
Stop detection method, device and detection device.
In a first aspect, the embodiment of the invention provides a kind of apnea detection method in sleep, the method is applied to
Apnea detection equipment in sleep, which comprises
Obtain heart operation sampled data and the blood oxygen saturation sampled data in preset duration；
Feature extraction is carried out based on heart operation sampled data, the heart operation obtained in the preset duration is special
Sign；
Feature extraction is carried out based on the blood oxygen saturation sampled data, obtains the blood oxygen saturation in the preset duration
Feature；
To the heart operation characteristic and blood oxygen saturation feature progress Fusion Features in same preset duration, obtain
Obtain sleep characteristics；
The sleep characteristics are identified based on default sleep apnea model, to be in the determination preset duration
No generation sleep apnea.
In some embodiments, the method also includes:
The default sleep apnea model is obtained in advance.
It is in some embodiments, described to obtain the default sleep apnea model in advance, comprising:
Obtain heart operation sampled data and the blood oxygen saturation sampled data at least two preset durations；
Feature extraction is carried out based on heart operation sampled data, the heart operation obtained in the preset duration is special
Sign；
Feature extraction is carried out based on the blood oxygen saturation sampled data, obtains the blood oxygen saturation in the preset duration
Feature；
To the heart operation characteristic and blood oxygen saturation feature progress Fusion Features in same preset duration, obtain
Obtain sleep characteristics；
Using the sleep apnea label in the sleep characteristics and each preset duration in each preset duration as input,
Model of the training based on machine learning obtains default sleep apnea model.
In some embodiments, described that feature extraction is carried out based on heart operation sampled data, it obtains described default
Heart operation characteristic in duration, comprising:
The heart rate variability in the preset duration is obtained based on the heart operation sampled data in the preset duration
Property information.
In some embodiments, described that feature extraction is carried out based on the blood oxygen saturation sampled data, it obtains described pre-
If the blood oxygen saturation feature in duration, comprising:
The blood oxygen saturation in the preset duration is obtained based on the blood oxygen saturation sampled data in the preset duration
The sum of, at least one of blood oxygen saturation minimum value, blood oxygen saturation average value, blood oxygen saturation maximum fall ginseng
Number.
In some embodiments, described that feature extraction is carried out based on the blood oxygen saturation sampled data, it obtains described pre-
If the blood oxygen saturation feature in duration, comprising:
It is sparse from encoding model using presetting to the blood oxygen saturation sampled data in the preset duration
Reason, extraction is described to preset the sparse hiding layer data from encoding model as blood oxygen saturation feature.
In some embodiments, it is saturated in the heart operation characteristic in same preset duration and the blood oxygen
It spends feature and carries out Fusion Features, after obtaining sleep characteristics, the method also includes:
Feature Selection is carried out to the sleep characteristics based on forward, backward algorithm；
Global normalization's processing is carried out to the sleep characteristics after Feature Selection.
In some embodiments, the heart operation characteristic and the blood oxygen saturation in same preset duration
Feature carries out Fusion Features, obtains sleep characteristics, comprising:
By in same preset duration the heart operation characteristic and the blood oxygen saturation feature connect, form institute
State the sleep characteristics vector of sleep characteristics.
Second aspect, the embodiment of the invention also provides apnea detection device in a kind of sleep, described device applications
The apnea detection equipment in sleep, described device include:
Data acquisition module, for obtaining the operation sampled data of the heart in preset duration and blood oxygen saturation hits
According to；
Heart operation characteristic obtains module, for carrying out feature extraction based on heart operation sampled data, obtains institute
State the heart operation characteristic in preset duration；
Blood oxygen saturation feature obtains module, for carrying out feature extraction based on the blood oxygen saturation sampled data, obtains
Obtain the blood oxygen saturation feature in the preset duration；
Fusion Features module, for in same preset duration the heart operation characteristic and the blood oxygen saturation it is special
Sign carries out Fusion Features, obtains sleep characteristics；
Feature recognition module, for being identified based on default sleep apnea model to the sleep characteristics, with true
Whether sleep apnea occurs in the fixed preset duration.
The third aspect, the embodiment of the invention also provides apnea detection equipment in a kind of sleep, comprising:
Heart unit, for obtaining heart operation data；
Blood oxygen saturation unit, for obtaining blood oxygen saturation data；
Control processing unit, for handling the heart operation data and heart saturation data, the control
Processing unit includes:
At least one processor and the memory being connect at least one described processor communication, the memory storage
There is the instruction that can be executed by least one described processor, described instruction is executed by least one described processor, so that described
At least one processor is able to carry out above-mentioned method.
Fourth aspect, the embodiment of the invention also provides a kind of non-volatile computer readable storage medium storing program for executing, the calculating
Machine readable storage medium storing program for executing is stored with computer executable instructions, when the computer executable instructions are examined by apnea in sleep
When measurement equipment executes, apnea detection equipment in the sleep is made to execute above-mentioned method.
Apnea detection method, apparatus and detection device in sleep provided in an embodiment of the present invention, by obtaining human body
Heart operation sampled data and blood oxygen saturation data in sleep, then run sampled data to heart respectively and blood oxygen are saturated
Degree obtains the sleep characteristics for having merged heart operation characteristic and blood oxygen saturation feature according to feature extraction and fusion is carried out.And
It is modeled and is identified using sleep characteristics, recognition accuracy is high.And medical supply and medical practitioner assessment without profession, make
With convenience, it is suitable for home use.
Detailed description of the invention
One or more embodiments are illustrated by the picture in corresponding attached drawing, these exemplary theorys
The bright restriction not constituted to embodiment, the element in attached drawing with same reference numbers label are expressed as similar element, remove
Non- to have special statement, composition does not limit the figure in attached drawing.
Fig. 1 is the application scenarios schematic diagram of apnea detection method and apparatus in present invention sleep；
Fig. 2 is the hardware structural diagram of one embodiment of apnea detection equipment in present invention sleep；
Fig. 3 is the flow chart of one embodiment of apnea detection method in present invention sleep；
Fig. 4 is the flow chart of one embodiment of apnea detection method in present invention sleep；
Fig. 5 is the default sleep apnea of acquisition in present invention sleep in one embodiment of apnea detection method
The flow chart of model step；
Fig. 6 is the structural schematic diagram of one embodiment of apnea detection device in present invention sleep；
Fig. 7 is the structural schematic diagram of one embodiment of apnea detection device in present invention sleep；
Fig. 8 is default sleep apnea model in present invention sleep in one embodiment of apnea detection device
Obtain the structural schematic diagram of module；
Fig. 9 is the hardware structural diagram of apnea detection equipment in sleep provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
Currently, PSG method is mostly used to detect the apnea in sleep, still, PSG method needs patient to wear
Complicated, accurate equipment realizes sleep apnea monitoring the whole night, and needs to be equipped with professional technician, is chiefly used in medicine
Monitoring.The present invention provides apnea detection method, apparatus and detection device in a kind of sleep, does not need the medical supply of profession
It is assessed with medical practitioner, can monitor automatically and patient's state of an illness is effectively assessed, cannot be only used for medical attendance can be used for
The routine healthcare of family.
Apnea detection method, apparatus and detection device are suitable for shown in Fig. 1 in sleep provided in an embodiment of the present invention
Application environment.The application environment includes apnea detection equipment 20 in user 10 and sleep.Apnea is examined in sleep
Measurement equipment 20 is used to obtain the heart operation data and blood oxygen saturation data of user 10, and by the heart operation data with
Blood oxygen saturation data judges whether user 10 occurs apnea in sleep.
Specifically, as shown in Fig. 2, apnea detection equipment 20 includes heart unit 21, blood oxygen saturation list in sleep
Member 22 and control processing unit 23.Wherein, heart unit 21 is used to detect the heart operation data of user 10, blood oxygen saturation list
Member 22 is used for for detecting the blood oxygen saturation data of user 10, control processing unit 23 to the heart operation data and blood oxygen
Saturation data is handled, and heart operation sampled data and blood oxygen saturation sampled data in preset duration are obtained.And point
Not Ji Yu heart operation sampled data and blood oxygen saturation sampled data carry out feature extraction, extract the heart in preset duration
Then operation characteristic and blood oxygen saturation feature carry out Fusion Features to heart operation characteristic and blood oxygen saturation feature and are slept
Dormancy feature, and modeled and identified using sleep characteristics, determine whether user 10 occurs to breathe in sleep in preset duration
Pause.
In practical application, the heart operation data and blood oxygen saturation in the available certain time of processing unit 23 are controlled
Degree evidence, then (i.e. preset duration is set as 1 minute) is split and samples to data as unit of per minute, obtains each one
Heart operation sampled data and blood oxygen saturation sampled data in minute, and sampled data is run based on the heart in one minute
Determine whether user 10 occurs sleep apnea within the minute with blood oxygen saturation sampled data.When so can get this section
Between the sleep apnea situation per minute of user 10 in section, further, apnea detection equipment 20 can be in sleep
One hour is unit, counts sleep apnea number in one hour, and sentence according to the sleep apnea number in one hour
The health condition of disconnected user 10.In further embodiments, the sleep apnea number in multiple one hours can also be obtained,
Then it is averaged, obtains a hourly average sleep apnea number, and according to a hourly average sleep apnea number
Judge the health status of user 10.For example, a hourly average sleep apnea number is indicated with y, the user if y is greater than 30
10 be sleep apnea syndrome severe patient；If y is greater than 15 less than 30, user 10 is sleep apnea syndrome
Moderate patient；If y is greater than 5 less than 15, user 10 is sleep apnea syndrome patients with mild；It is used if y is less than 5
10 health of family.The value of preset duration can carry out value according to practical situations, and in the present embodiment, preset duration is 1 point
Clock, in some other embodiment, preset duration may be other times.
Wherein, heart runs sampled data such as heart impact signal (Ballistocardiogram, BCG), electrocardiogram
(Electrocardiograph, ECG) data, photoplethysmographic (Photoplethysmography, PPG) data etc..
Corresponding, heart unit 21 can be micro-tremor signal sensor (such as piezoelectric film sensor, acceleration transducer etc.), electrocardio
Map sensor, photoelectric sensor etc..Blood oxygen saturation data such as blood oxygen saturation (oxygen saturatio, SpO2) is right
It answers, blood oxygen saturation unit 22 is BOLD contrast.
By the modeling of the sleep characteristics of fusion heart operation characteristic and blood oxygen saturation feature and identify sleep apnea,
It is divided into two stages, respectively default sleep apnea model training stage, and utilize default sleep apnea model
Sleep characteristics are identified, to determine whether that the cognitive phase of sleep apnea occurs.Default sleep apnea model
Training stage is mainly heart operation sampled data and the blood oxygen saturation sampled data acquired in multiple preset durations, is then divided
Not Ji Yu heart operation sampled data and blood oxygen saturation sampled data carry out feature extraction, obtain in multiple preset durations
Heart operation characteristic and blood oxygen saturation feature, to the heart operation characteristic and blood oxygen saturation feature in same preset duration
It carries out Fusion Features and obtains the sleep characteristics in multiple preset durations.Sleep characteristics in multiple preset durations are preset with each
For sleep apnea label in duration as input, model of the training based on machine learning obtains default sleep apnea
Model.
Wherein, default sleep apnea model, which can be to be pre-configured with, is stored in apnea detection equipment 20 in sleep
In, it is also possible to what apnea detection equipment 20 in sleep was obtained by running certain method and step in use.
Wherein, sleep apnea label can be so that whether certain sign flag occurs sleep apnea, for example,
It indicates that sleep apnea occurs with Y, indicates that sleep apnea does not occur with N, the sleep apnea mark in preset duration
Label can be obtained based on the experience of doctor.Medical practitioner can run sampled data according to heart in seconds and blood oxygen is saturated
Whether degree sampled data label occurs sleep apnea each second.In order to match with default sleep apnea model, need
The label marked with each second is changed into the label marked with preset duration.If can be provided according to practical situations pre-
If the number of seconds that sleep apnea occurs in duration reaches certain numerical value, then sleep apnea has occurred in the preset duration.
For example, if preset duration is one minute, if 10 seconds or more time, to be marked as sleep-respiratory temporary in regulation one minute
Stop, then this minute is labeled as sleep apnea.In this way, the sleep apnea mark in each preset duration can be obtained
Label.
Wherein, in the model based on machine learning, machine learning algorithm can also use nothing using there is supervision algorithm
The algorithm of supervision, for example, support vector machines (Support Vector Machine, SVM) method, be based on depth confidence network
(Deep Belief Network, DBN) algorithm, hidden Markov model (Hidden Markov Model, HMM) algorithm, ladder
Degree promotes decision tree (Gradient Boosting Decision Tree, GBDT) algorithm etc., and the embodiment of the present invention does not limit
System.
After obtaining default sleep apnea model, the heart acquired in preset duration in the same way runs hits
Preset duration is obtained according to blood oxygen saturation sampled data and based on heart operation sampled data and blood oxygen saturation sampled data
Then interior sleep characteristics identify the sleep characteristics using default sleep apnea model, to determine preset duration
Inside whether sleep apnea occurs.Since heart sampled data and blood oxygen saturation sampled data can accurately embody user's 10
The state of an illness is modeled and is identified using the sleep characteristics that sampled data and blood oxygen saturation sampled data obtain are run based on heart
Sleep apnea, recognition accuracy are high.
Fig. 3 is the flow diagram of apnea detection method in sleep provided in an embodiment of the present invention, and the method can
It is executed by the control processing unit 23 in apnea detection equipment 20 in the sleep in Fig. 1 or Fig. 2, as shown in figure 3, the side
Method includes:
101: obtaining heart operation sampled data and the blood oxygen saturation sampled data in preset duration.
Sampled data is run based on heart and blood oxygen saturation sampled data obtains sleep characteristics and modeled and identified, is known
Other accuracy rate is high.In other embodiments of the invention, can also only with heart operation sampled data obtain sleep characteristics into
Row modeling and identification still cannot embody completely due to heart operation sampled data and accurately reflect the state of an illness of user 10, identification
When error is larger, recognition accuracy is low.
102: feature extraction being carried out based on heart operation sampled data, obtains the heart operation in the preset duration
Feature.
Wherein, heart operation sampled data such as BCG data, ECG data, PPG data etc., obtains the heart in preset duration
Dirty operation characteristic can be the heart rate variability (Heart obtained in preset duration based on BCG data, ECG data, PPG data etc.
Rate Variability, HRV) information.HRV is gradually the Minor variations of heart rate between heart beat cycle, is also referred between RR interphase
Minor change.In practical application, it can be obtained between RR by carrying out the detection of R wave to BCG data, ECG data, PPG data etc.
Every then to the interval RR progress linear and nonlinear data characteristics extraction acquisition HRV information.Specifically, HRV information includes following
At least one of parameter:
(1) reflect RR interphase average value (mean of the RR intervals, Mean of heart rate variability average level
RR)；
(2) average value (mean suecessive differenees, MSD) of adjacent R R interphase absolute value of the difference；
(3) RR interphase standard deviation mean value (RR interphase sd means, Mean SD)；
RR interphase standard error of the mean (Standard deviation of the average of in (4) 5 minutes
NN intervals in alls minutes of the entire recording, SDANN)；
(5) root mean square (the The root mean square of difference between of adjacent R R interphase difference
Adjacent NN intervals, R MSSD)；
(6) sinus property adjacent R R interphase difference is more than 50 milliseconds of heart rate and the ratio (Percent of the total heart rate of RR interphase
Of NN 50in the total number of RR intervals, PNN50)；
(7) standard deviation (the Standard deviation of Successive Difference of whole RR interphase difference
Between adjacent cycles, SDSD)；
(8) (i.e. in the histogram of whole RR interphase, RR interphase sum is divided by between the maximum RR of accounting example for HRV triangle index
Issue)；
(9) interim between whole RR, the difference of adjacent RR interphase is greater than heart rate (the number of pairs of of 50ms
Adjacent normal to normal intervals differing by more than 50ms, NN50)；
(10) NN50 divided by total RR interphase number percentage (50 in the total of Percent of NN
Number of RR intervals, PNN50)；
(11) ultralow frequency (Ultra Low Frequency, ULF)；
(12) very low frequencies (very low frequencies, VLF)；
(13) high frequency (High Frequency, HF)；
(14) low frequency (Low frequency, LF).
103: feature extraction being carried out based on the blood oxygen saturation sampled data, the blood oxygen obtained in the preset duration is full
With degree feature.
In some embodiments, blood oxygen saturation feature can be the sum of blood oxygen saturation in the preset duration, blood
Oxygen saturation minimum value, blood oxygen saturation average value, at least one parameter in blood oxygen saturation maximum fall.
In further embodiments, can use preset it is sparse from encoding model to blood oxygen saturation sampled data at
Reason, extraction preset the sparse hidden layer data vector from encoding model as blood oxygen saturation feature.Wherein, preset it is sparse from
Encoding model can be obtained based on the training of sparse autocoding algorithm.Specifically, being to be adopted in 1 minute, one minute with preset duration
For 60 blood oxygen saturation sampled datas of sample, then the input layer quantity of sparse autoencoder network is 60, output layer mind
Quantity through member is also 60, and the neuronal quantity of hidden layer can optimize situation according to the loss function of network and determine, such as can
Thought for 15 (hidden layer neuron quantity is lower than input layer quantity, can carry out dimensionality reduction to input layer data).By right
Sparse autoencoder network is iterated training, can be obtained preset it is sparse from encoding model.Wherein, the number of iterations is according to loss letter
Several optimum results determine, can choose the situation of error minimum, or when loss function is with the increase of the number of iterations, no longer
There is the case where significant change.Wherein, objective function hw,b(x) ≈ x, in some embodiments, loss function are as follows:Sensitivity table during the network optimization is shown as:
104: in same preset duration the heart operation characteristic and the blood oxygen saturation feature carry out feature melt
It closes, obtains sleep characteristics.
Wherein, in some embodiments, heart operation characteristic and blood oxygen saturation feature are merged i.e. by heart operation characteristic
It connects with blood oxygen saturation feature, forms sleep characteristics vector.
In sleep characteristics after Fusion Features, some features are little for identification sleep apnea effect or even can rise anti-
It acts on, differs greatly between some features.In order to remove little to recognition reaction or even reactive feature, and avoid feature
Between difference it is excessive, in some embodiments of the invention, the method also includes:
Feature Selection is carried out to the sleep characteristics based on forward, backward algorithm；
Global normalization's processing is carried out to by the sleep characteristics of Feature Selection.
Feature Selection is carried out using forward, backward algorithm to fused sleep characteristics and is carried out at global normalization
Reason, can be further improved recognition accuracy.
105: the sleep characteristics being identified based on default sleep apnea model, with the determination preset duration
Inside whether sleep apnea occurs.
The embodiment of the present invention runs sampled data and blood oxygen saturation data by the heart obtained in sleep quality, then
Sampled data is run to heart respectively and blood oxygen saturation data carries out feature extraction and fusion, it is special that heart operation has been merged in acquisition
The sleep characteristics of blood oxygen saturation of seeking peace feature.And modeled and identified using sleep characteristics, recognition accuracy is high.And it is not necessarily to
Medical supply and the medical practitioner assessment of profession, it is easy to use, be suitable for home use.
Wherein, default sleep apnea model, which can be to be pre-configured with, is stored in apnea detection equipment 20 in sleep
In, it is also possible to what apnea detection equipment 20 in sleep was obtained by operation following methods step in use, it may be assumed that
201: obtaining heart operation sampled data and the blood oxygen saturation sampled data at least two preset durations.
Heart operation sampled data and blood oxygen saturation data are temporary for establishing default sleep-respiratory as sample data
Stop model.
202: feature extraction being carried out based on heart operation sampled data, obtains the heart operation in the preset duration
Feature.
203: feature extraction being carried out based on the blood oxygen saturation sampled data, the blood oxygen obtained in the preset duration is full
With degree feature.
204: in same preset duration the heart operation characteristic and the blood oxygen saturation feature carry out feature melt
It closes, obtains sleep characteristics.
Wherein, the particular technique content of step 201,202,203 and 204 can respectively refer to above-mentioned steps 101,102,
103 and 104, details are not described herein.
205: using the sleep apnea label in the sleep characteristics and each preset duration in each preset duration as
Input, model of the training based on machine learning obtain default sleep apnea model.
Wherein, in the model based on machine learning, machine learning algorithm for example SVM method, DBN algorithm, HMM algorithm,
GBDT algorithm etc..By the sleep characteristics and the input of sleep apnea label in each preset duration, obtained based on above-mentioned algorithm
Parameters in network model, to obtain default sleep apnea model.
Correspondingly, being breathed in the sleep the embodiment of the invention also provides apnea detection device in a kind of sleep
Suspend detection device for apnea detection equipment 20 in sleep shown in fig. 1 or fig. 2, as shown in fig. 6, breathing is temporary in sleep
Stopping detection device 600 includes:
Data acquisition module 601, for obtaining the operation sampled data of the heart in preset duration and blood oxygen saturation sampling
Data.
Heart operation characteristic obtains module 602, for carrying out feature extraction based on heart operation sampled data, obtains
Heart operation characteristic in the preset duration.
Blood oxygen saturation feature obtains module 603, for carrying out feature extraction based on the blood oxygen saturation sampled data,
Obtain the blood oxygen saturation feature in the preset duration.
Fusion Features module 604, for the heart operation characteristic and blood oxygen saturation in same preset duration
It spends feature and carries out Fusion Features, obtain sleep characteristics.
Feature recognition module 605, for being identified based on default sleep apnea model to the sleep characteristics, with
It determines in the preset duration and whether sleep apnea occurs.
The embodiment of the present invention runs sampled data and blood oxygen saturation data by the heart obtained in sleep quality, then
Sampled data is run to heart respectively and blood oxygen saturation data carries out feature extraction and fusion, it is special that heart operation has been merged in acquisition
The sleep characteristics of blood oxygen saturation of seeking peace feature.And modeled and identified using sleep characteristics, recognition accuracy is high.And it is not necessarily to
Medical supply and the medical practitioner assessment of profession, it is easy to use, be suitable for home use.
In sleep in other embodiments of apnea detection device 600, Fig. 7, apnea in sleep are please referred to
Detection device 600 further include:
Default sleep apnea model obtains module 606, for obtaining the default sleep apnea model in advance.
Wherein, in sleep in some embodiments of apnea detection device 600, as shown in figure 8, default sleep-respiratory
Pause model obtains module 606
Data acquisition module 601, it is full for obtaining the operation sampled data of the heart at least two preset durations and blood oxygen
With degree sampled data.
Heart operation characteristic obtains module 602, for carrying out feature extraction based on heart operation sampled data, obtains
Heart operation characteristic in the preset duration.
Blood oxygen saturation feature obtains module 603, for carrying out feature extraction based on the blood oxygen saturation sampled data,
Obtain the blood oxygen saturation feature in the preset duration.
Fusion Features module 604, for the heart operation characteristic and blood oxygen saturation in same preset duration
It spends feature and carries out Fusion Features, obtain sleep characteristics.
Default sleep apnea model training module 6065, for by sleep characteristics in each preset duration and each
For sleep apnea label in preset duration as input, model of the training based on machine learning obtains default sleep-respiratory
Suspend model.
Wherein, in sleep in some embodiments of apnea detection device 600, heart operation characteristic obtains module
602 are specifically used for:
The heart rate variability in the preset duration is obtained based on the heart operation sampled data in the preset duration
Property information.
In sleep in some embodiments of apnea detection device 600, blood oxygen saturation feature, which obtains module 603, to be had
Body is used for:
The blood oxygen saturation in the preset duration is obtained based on the blood oxygen saturation sampled data in the preset duration
The sum of, at least one of blood oxygen saturation minimum value, blood oxygen saturation average value, blood oxygen saturation maximum fall ginseng
Number.
In sleep in other embodiments of apnea detection device 600, blood oxygen saturation feature obtains module 603
It is specifically used for:
It is sparse from encoding model using presetting to the blood oxygen saturation sampled data in the preset duration
Reason, extraction is described to preset the sparse hiding layer data from encoding model as blood oxygen saturation feature.
In sleep in other embodiments of apnea detection device 600, Fig. 7, apnea in sleep are please referred to
Detection device 600 further include:
Feature Selection module 607, for carrying out Feature Selection to the sleep characteristics based on forward, backward algorithm；
Global normalization's module 608, for being carried out at global normalization to the sleep characteristics after Feature Selection
Reason.
In sleep in other embodiments of apnea detection device 600, Fusion Features module 604 is specifically used for:
By in same preset time the heart operation characteristic and the blood oxygen saturation feature connect, form institute
State the sleep characteristics vector of sleep characteristics.
It should be noted that sleep provided by the embodiment of the present invention can be performed in apnea detection device in above-mentioned sleep
Middle apnea detection method has the corresponding functional module of execution method and beneficial effect.Not in sleep, apnea is examined
The technical detail of detailed description in Installation practice is surveyed, reference can be made to apnea detection in sleep provided by the embodiment of the present invention
Method.
The embodiment of the invention also provides a kind of sleeps of apnea detection method and apparatus application in present invention sleep
Middle apnea detection equipment, Fig. 9 show the specific hardware structure of apnea detection equipment 20 in sleep.It is breathed in sleep
Suspending detection device 20 includes heart unit 21, blood oxygen saturation unit 22 and control processing unit 23.Wherein, heart unit 21
For obtaining heart operation data, blood oxygen saturation unit 22 is used for obtaining blood oxygen saturation data, control processing unit 23
It is handled in the heart operation data and heart saturation data.Specifically, control processing unit 23 includes:
One or more processors 231 and memory 232, in Fig. 9 by taking a processor 231 as an example.
Processor 231 can be connected with memory 232 by bus or other modes, to be connected by bus in Fig. 9
For.
Memory 232 is used as a kind of non-volatile computer readable storage medium storing program for executing, can be used for storing non-volatile software journey
Sequence, non-volatile computer executable program and module, such as apnea detection method in the sleep in the embodiment of the present invention
Corresponding program instruction/module (for example, attached data acquisition module shown in fig. 6 601, heart operation characteristic obtain module 602,
Blood oxygen saturation feature obtains module 603, Fusion Features module 604 and feature recognition module 605).Processor 231 passes through operation
Non-volatile software program, instruction and the module being stored in memory 232, set thereby executing apnea detection in sleep
Standby various function application and data processing, i.e. apnea detection method in the sleep of realization above method embodiment.
Memory 232 may include storing program area and storage data area, wherein storing program area can store operation system
Application program required for system, at least one function；Storage data area can be stored according to apnea detection device in sleep
Use created data etc..In addition, memory 232 may include high-speed random access memory, it can also include non-volatile
Property memory, a for example, at least disk memory, flush memory device or other non-volatile solid state memory parts.Some
In embodiment, it includes the memory remotely located relative to processor 231 that memory 232 is optional, these remote memories can be with
Pass through apnea detection device in network connection extremely sleep.The example of above-mentioned network includes but is not limited to internet, in enterprise
Portion's net, local area network, mobile radio communication and combinations thereof.
One or more of modules are stored in the memory 232, when by one or more of processors
When 231 execution, apnea detection method in the sleep in above-mentioned any means embodiment is executed, for example, executing above description
Fig. 3 in method and step 201 of the method and step 101 to step 105, in method and step 101a, 101-105 in Fig. 4, Fig. 5
To step 205；Realize the function of module 601-604 in module 601-605, Fig. 7 in Fig. 6 in module 601-608, Fig. 8,6065
Energy.
Method provided by the embodiment of the present invention can be performed in the said goods, has the corresponding functional module of execution method and has
Beneficial effect.The not technical detail of detailed description in the present embodiment, reference can be made to method provided by the embodiment of the present invention.
The embodiment of the invention provides a kind of non-volatile computer readable storage medium storing program for executing, the computer-readable storage mediums
Matter is stored with computer executable instructions, which is executed by one or more processors, such as in Fig. 9
One processor 231 may make said one or multiple processors can be performed to exhale in the sleep in above-mentioned any means embodiment
Pause detection method is inhaled, for example, the method and step for executing method and step 101 in Fig. 3 described above to step 105, in Fig. 4
101a, 101-105, method and step 201 in Fig. 5 to step 205；Realize module 601- in module 601-605, Fig. 7 in Fig. 6
608, the module 601-604 in Fig. 8,6065 function.
The apparatus embodiments described above are merely exemplary, wherein described, unit can as illustrated by the separation member
It is physically separated with being or may not be, component shown as a unit may or may not be physics list
Member, it can it is in one place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of the modules achieve the purpose of the solution of this embodiment.
Through the above description of the embodiments, those of ordinary skill in the art can be understood that each embodiment
The mode of general hardware platform can be added to realize by software, naturally it is also possible to pass through hardware.Those of ordinary skill in the art can
With understand all or part of the process realized in above-described embodiment method be can be instructed by computer program it is relevant hard
Part is completed, and the program can be stored in a computer-readable storage medium, the program is when being executed, it may include as above
State the process of the embodiment of each method.Wherein, the storage medium can be magnetic disk, CD, read-only memory (Read-
Only Memory, ROM) or random access memory (RandomAccessMemory, RAM) etc..
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；At this
It under the thinking of invention, can also be combined between the technical characteristic in above embodiments or different embodiment, step can be with
It is realized with random order, and there are many other variations of different aspect present invention as described above, for simplicity, they do not have
Have and is provided in details；Although the present invention is described in detail referring to the foregoing embodiments, the ordinary skill people of this field
Member is it is understood that it is still possible to modify the technical solutions described in the foregoing embodiments, or to part of skill
Art feature is equivalently replaced；And these are modified or replaceed, each reality of the present invention that it does not separate the essence of the corresponding technical solution
Apply the range of a technical solution.
Claims (11)
1. a kind of apnea detection method in sleep, the method is applied to apnea detection equipment in sleep, feature
It is, which comprises
Obtain heart operation sampled data and the blood oxygen saturation sampled data in preset duration；
Feature extraction is carried out based on heart operation sampled data, obtains the heart operation characteristic in the preset duration；
Feature extraction is carried out based on the blood oxygen saturation sampled data, the blood oxygen saturation obtained in the preset duration is special
Sign；
To the heart operation characteristic and blood oxygen saturation feature progress Fusion Features in same preset duration, slept
Dormancy feature；
The sleep characteristics are identified based on default sleep apnea model, whether to be sent out in the determination preset duration
Raw sleep apnea.
2. the method according to claim 1, wherein the method also includes:
The default sleep apnea model is obtained in advance.
3. according to the method described in claim 2, it is characterized in that, described obtain the default sleep apnea mould in advance
Type, comprising:
Obtain heart operation sampled data and the blood oxygen saturation sampled data at least two preset durations；
Feature extraction is carried out based on heart operation sampled data, obtains the heart operation characteristic in the preset duration；
Feature extraction is carried out based on the blood oxygen saturation sampled data, the blood oxygen saturation obtained in the preset duration is special
Sign；
To the heart operation characteristic and blood oxygen saturation feature progress Fusion Features in same preset duration, slept
Dormancy feature；
Using the sleep apnea label in the sleep characteristics and each preset duration in each preset duration as input, training
Model based on machine learning obtains default sleep apnea model.
4. method according to claim 1 to 3, which is characterized in that described to run hits based on the heart
According to feature extraction is carried out, the heart operation characteristic in the preset duration is obtained, comprising:
The letter of the heart rate variability in the preset duration is obtained based on the heart operation sampled data in the preset duration
Breath.
5. method according to claim 1 to 3, which is characterized in that described to be sampled based on the blood oxygen saturation
Data carry out feature extraction, obtain the blood oxygen saturation feature in the preset duration, comprising:
Obtained based on the blood oxygen saturation sampled data in the preset duration the sum of blood oxygen saturation in the preset duration,
Blood oxygen saturation minimum value, blood oxygen saturation average value, at least one parameter in blood oxygen saturation maximum fall.
6. method according to claim 1 to 3, which is characterized in that described to be sampled based on the blood oxygen saturation
Data carry out feature extraction, obtain the blood oxygen saturation feature in the preset duration, comprising:
To the blood oxygen saturation sampled data in the preset duration, using preset it is sparse handled from encoding model,
It extracts and described presets the sparse hiding layer data from encoding model as blood oxygen saturation feature.
7. method according to claim 1 to 3, which is characterized in that in the institute in same preset duration
It states heart operation characteristic and the blood oxygen saturation feature carries out Fusion Features, after obtaining sleep characteristics, the method is also wrapped
It includes:
Feature Selection is carried out to the sleep characteristics based on forward, backward algorithm；
Global normalization's processing is carried out to the sleep characteristics after Feature Selection.
8. method according to claim 1 to 3, which is characterized in that described to described in same preset duration
Heart operation characteristic and the blood oxygen saturation feature carry out Fusion Features, obtain sleep characteristics, comprising:
By in same preset duration the heart operation characteristic and the blood oxygen saturation feature connect, slept described in composition
The sleep characteristics vector of dormancy feature.
9. apnea detection device in a kind of sleep, described device is applied to apnea detection equipment in sleep, feature
It is, described device includes:
Data acquisition module, for obtaining the operation sampled data of the heart in preset duration and blood oxygen saturation sampled data；
Heart operation characteristic obtains module, for carrying out feature extraction based on heart operation sampled data, obtains described pre-
If the heart operation characteristic in duration；
Blood oxygen saturation feature obtains module, for carrying out feature extraction based on the blood oxygen saturation sampled data, obtains institute
State the blood oxygen saturation feature in preset duration；
Fusion Features module, for in same preset duration the heart operation characteristic and the blood oxygen saturation feature into
Row Fusion Features obtain sleep characteristics；
Feature recognition module, for being identified based on default sleep apnea model to the sleep characteristics, to determine
It states in preset duration and whether sleep apnea occurs.
10. apnea detection equipment in a kind of sleep characterized by comprising
Heart unit, for obtaining heart operation data；
Blood oxygen saturation unit, for obtaining blood oxygen saturation data；
Processing unit is controlled, for handling the heart operation data and heart saturation data, the control processing
Unit includes:
At least one processor and the memory connecting at least one described processor communication, the memory is stored with can
The instruction executed by least one described processor, described instruction executed by least one described processor so that it is described at least
One processor is able to carry out the described in any item methods of claim 1-8.
11. a kind of non-volatile computer readable storage medium storing program for executing, which is characterized in that the computer-readable recording medium storage has
Computer executable instructions make described when the computer executable instructions are executed by apnea detection equipment in sleep
Apnea detection equipment perform claim requires the described in any item methods of 1-8 in sleep.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811252525.9A CN109259733A (en) | 2018-10-25 | 2018-10-25 | Apnea detection method, apparatus and detection device in a kind of sleep |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811252525.9A CN109259733A (en) | 2018-10-25 | 2018-10-25 | Apnea detection method, apparatus and detection device in a kind of sleep |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109259733A true CN109259733A (en) | 2019-01-25 |
Family
ID=65194296
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811252525.9A Pending CN109259733A (en) | 2018-10-25 | 2018-10-25 | Apnea detection method, apparatus and detection device in a kind of sleep |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109259733A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110742621A (en) * | 2019-10-31 | 2020-02-04 | 京东方科技集团股份有限公司 | Signal processing method and computer equipment |
| CN111374819A (en) * | 2020-03-16 | 2020-07-07 | 深圳市值米科技有限公司 | Snore stopper, snore recognition method thereof, snore recognition device and storage medium |
| CN111803038A (en) * | 2020-09-03 | 2020-10-23 | 平安国际智慧城市科技股份有限公司 | Method, system, terminal and storage medium for monitoring sleep apnea event |
| CN111814830A (en) * | 2020-06-08 | 2020-10-23 | 珠海格力电器股份有限公司 | Sleep state detection model construction method, sleep state detection method and device |
| CN112914506A (en) * | 2021-01-19 | 2021-06-08 | 青岛歌尔智能传感器有限公司 | Sleep quality detection method, device and computer readable storage medium |
| CN113243890A (en) * | 2021-05-10 | 2021-08-13 | 清华大学深圳国际研究生院 | Sleep apnea syndrome recognition device |
| CN113576401A (en) * | 2021-06-11 | 2021-11-02 | 广东工业大学 | Sleep apnea syndrome rapid diagnosis device based on convolutional neural network |
| TWI750531B (en) * | 2019-11-12 | 2021-12-21 | 國立勤益科技大學 | Detection device and detection method for apnea based on chest respiratory signal |
| US11324950B2 (en) | 2016-04-19 | 2022-05-10 | Inspire Medical Systems, Inc. | Accelerometer-based sensing for sleep disordered breathing (SDB) care |
| US11738197B2 (en) | 2019-07-25 | 2023-08-29 | Inspire Medical Systems, Inc. | Systems and methods for operating an implantable medical device based upon sensed posture information |
| CN117017233A (en) * | 2023-10-08 | 2023-11-10 | 华南师范大学 | Sleep apnea detection method, device, equipment and storage medium |
Citations (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101496716A (en) * | 2009-02-26 | 2009-08-05 | 周洪建 | Measurement method for detecting sleep apnoea with ECG signal |
| CN203468619U (en) * | 2013-08-15 | 2014-03-12 | 徐州工程学院 | Obstructive sleep apnea syndrome monitoring device |
| CN104107036A (en) * | 2014-07-09 | 2014-10-22 | 中南民族大学 | Household portable monitoring device for sleep apnea |
| CN104545818A (en) * | 2015-01-29 | 2015-04-29 | 吉林大学 | Sleep apnea syndrome detection method based on pulse and blood oxygen signals |
| CN104778671A (en) * | 2015-04-21 | 2015-07-15 | 重庆大学 | Image super-resolution method based on SAE and sparse representation |
| CN104809469A (en) * | 2015-04-21 | 2015-07-29 | 重庆大学 | Indoor scene image classification method facing service robot |
| CN105243154A (en) * | 2015-10-27 | 2016-01-13 | 武汉大学 | Remote sensing image retrieval method and system based on significant point characteristics and spare self-encodings |
| CN105606198A (en) * | 2016-01-28 | 2016-05-25 | 吉林大学 | Fiber vibration sensing system two-order signal feature extraction and determining method |
| CN106175695A (en) * | 2016-08-31 | 2016-12-07 | 北京首科长昊医疗科技有限公司 | A kind of detecting system of sleep apnea syndrome |
| CN106361277A (en) * | 2016-08-26 | 2017-02-01 | 中山大学 | Sleep apnea syndrome assessment method based on electrocardiogram signals |
| CN106473700A (en) * | 2016-08-26 | 2017-03-08 | 复旦大学 | Detecting system based on the OSAS of HRV frequency-domain analysis |
| CN107038421A (en) * | 2017-04-17 | 2017-08-11 | 杭州电子科技大学 | Modulation Types recognition methods based on sparse storehouse own coding |
| CN107292885A (en) * | 2017-08-08 | 2017-10-24 | 广东工业大学 | A kind of product defects classifying identification method and device based on autocoder |
| CN107403191A (en) * | 2017-07-03 | 2017-11-28 | 杭州电子科技大学 | A kind of semi-supervised learning machine sorting technique that transfinites with depth structure |
| CN107578007A (en) * | 2017-09-01 | 2018-01-12 | 杭州电子科技大学 | A kind of deep learning face identification method based on multi-feature fusion |
| CN107807917A (en) * | 2017-09-27 | 2018-03-16 | 风变科技（深圳）有限公司 | Method for extracting content of text, device, system and storage medium |
| CN108038414A (en) * | 2017-11-02 | 2018-05-15 | 平安科技（深圳）有限公司 | Character personality analysis method, device and storage medium based on Recognition with Recurrent Neural Network |
| CN108052889A (en) * | 2017-12-08 | 2018-05-18 | 上海壹账通金融科技有限公司 | Emotion identification method, apparatus and storage medium |
| CN108108723A (en) * | 2018-01-19 | 2018-06-01 | 深圳市恩钛控股有限公司 | A kind of face feature extraction method based on deep learning |
| CN108113647A (en) * | 2016-11-28 | 2018-06-05 | 深圳先进技术研究院 | A kind of electrocardiosignal sorter and method |
| CN108309277A (en) * | 2018-02-01 | 2018-07-24 | 深圳市禹欣鑫电子有限公司 | The Intelligent monitoring device of sleep apnea based on brain electricity and blood oxygen saturation |
| CN108523907A (en) * | 2018-01-22 | 2018-09-14 | 上海交通大学 | The fatigue state recognition method and system of sparse autoencoder network are shunk based on depth |
-
2018
- 2018-10-25 CN CN201811252525.9A patent/CN109259733A/en active Pending
Patent Citations (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101496716A (en) * | 2009-02-26 | 2009-08-05 | 周洪建 | Measurement method for detecting sleep apnoea with ECG signal |
| CN203468619U (en) * | 2013-08-15 | 2014-03-12 | 徐州工程学院 | Obstructive sleep apnea syndrome monitoring device |
| CN104107036A (en) * | 2014-07-09 | 2014-10-22 | 中南民族大学 | Household portable monitoring device for sleep apnea |
| CN104545818A (en) * | 2015-01-29 | 2015-04-29 | 吉林大学 | Sleep apnea syndrome detection method based on pulse and blood oxygen signals |
| CN104778671A (en) * | 2015-04-21 | 2015-07-15 | 重庆大学 | Image super-resolution method based on SAE and sparse representation |
| CN104809469A (en) * | 2015-04-21 | 2015-07-29 | 重庆大学 | Indoor scene image classification method facing service robot |
| CN105243154A (en) * | 2015-10-27 | 2016-01-13 | 武汉大学 | Remote sensing image retrieval method and system based on significant point characteristics and spare self-encodings |
| CN105606198A (en) * | 2016-01-28 | 2016-05-25 | 吉林大学 | Fiber vibration sensing system two-order signal feature extraction and determining method |
| CN106473700A (en) * | 2016-08-26 | 2017-03-08 | 复旦大学 | Detecting system based on the OSAS of HRV frequency-domain analysis |
| CN106361277A (en) * | 2016-08-26 | 2017-02-01 | 中山大学 | Sleep apnea syndrome assessment method based on electrocardiogram signals |
| CN106175695A (en) * | 2016-08-31 | 2016-12-07 | 北京首科长昊医疗科技有限公司 | A kind of detecting system of sleep apnea syndrome |
| CN108113647A (en) * | 2016-11-28 | 2018-06-05 | 深圳先进技术研究院 | A kind of electrocardiosignal sorter and method |
| CN107038421A (en) * | 2017-04-17 | 2017-08-11 | 杭州电子科技大学 | Modulation Types recognition methods based on sparse storehouse own coding |
| CN107403191A (en) * | 2017-07-03 | 2017-11-28 | 杭州电子科技大学 | A kind of semi-supervised learning machine sorting technique that transfinites with depth structure |
| CN107292885A (en) * | 2017-08-08 | 2017-10-24 | 广东工业大学 | A kind of product defects classifying identification method and device based on autocoder |
| CN107578007A (en) * | 2017-09-01 | 2018-01-12 | 杭州电子科技大学 | A kind of deep learning face identification method based on multi-feature fusion |
| CN107807917A (en) * | 2017-09-27 | 2018-03-16 | 风变科技（深圳）有限公司 | Method for extracting content of text, device, system and storage medium |
| CN108038414A (en) * | 2017-11-02 | 2018-05-15 | 平安科技（深圳）有限公司 | Character personality analysis method, device and storage medium based on Recognition with Recurrent Neural Network |
| CN108052889A (en) * | 2017-12-08 | 2018-05-18 | 上海壹账通金融科技有限公司 | Emotion identification method, apparatus and storage medium |
| CN108108723A (en) * | 2018-01-19 | 2018-06-01 | 深圳市恩钛控股有限公司 | A kind of face feature extraction method based on deep learning |
| CN108523907A (en) * | 2018-01-22 | 2018-09-14 | 上海交通大学 | The fatigue state recognition method and system of sparse autoencoder network are shunk based on depth |
| CN108309277A (en) * | 2018-02-01 | 2018-07-24 | 深圳市禹欣鑫电子有限公司 | The Intelligent monitoring device of sleep apnea based on brain electricity and blood oxygen saturation |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11324950B2 (en) | 2016-04-19 | 2022-05-10 | Inspire Medical Systems, Inc. | Accelerometer-based sensing for sleep disordered breathing (SDB) care |
| US11738197B2 (en) | 2019-07-25 | 2023-08-29 | Inspire Medical Systems, Inc. | Systems and methods for operating an implantable medical device based upon sensed posture information |
| CN110742621A (en) * | 2019-10-31 | 2020-02-04 | 京东方科技集团股份有限公司 | Signal processing method and computer equipment |
| TWI750531B (en) * | 2019-11-12 | 2021-12-21 | 國立勤益科技大學 | Detection device and detection method for apnea based on chest respiratory signal |
| CN111374819A (en) * | 2020-03-16 | 2020-07-07 | 深圳市值米科技有限公司 | Snore stopper, snore recognition method thereof, snore recognition device and storage medium |
| CN111814830A (en) * | 2020-06-08 | 2020-10-23 | 珠海格力电器股份有限公司 | Sleep state detection model construction method, sleep state detection method and device |
| CN111803038B (en) * | 2020-09-03 | 2021-01-01 | 平安国际智慧城市科技股份有限公司 | Method, system, terminal and storage medium for monitoring sleep apnea event |
| CN111803038A (en) * | 2020-09-03 | 2020-10-23 | 平安国际智慧城市科技股份有限公司 | Method, system, terminal and storage medium for monitoring sleep apnea event |
| CN112914506A (en) * | 2021-01-19 | 2021-06-08 | 青岛歌尔智能传感器有限公司 | Sleep quality detection method, device and computer readable storage medium |
| CN113243890A (en) * | 2021-05-10 | 2021-08-13 | 清华大学深圳国际研究生院 | Sleep apnea syndrome recognition device |
| CN113576401A (en) * | 2021-06-11 | 2021-11-02 | 广东工业大学 | Sleep apnea syndrome rapid diagnosis device based on convolutional neural network |
| CN117017233A (en) * | 2023-10-08 | 2023-11-10 | 华南师范大学 | Sleep apnea detection method, device, equipment and storage medium |
| CN117017233B (en) * | 2023-10-08 | 2024-01-09 | 华南师范大学 | Sleep apnea detection method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109259733A (en) | Apnea detection method, apparatus and detection device in a kind of sleep | |
| US20230181077A1 (en) | Machine differentiation of abnormalities in bioelectromagnetic fields | |
| CN102973253B (en) | Method and system for monitoring human physiological indexes by using visual information | |
| Fischer et al. | An algorithm for real-time pulse waveform segmentation and artifact detection in photoplethysmograms | |
| WO2019161607A1 (en) | Electrocardiogram information dynamic monitoring method and dynamic monitoring system | |
| WO2019161609A1 (en) | Method for analyzing multi-parameter monitoring data and multi-parameter monitor | |
| US10278595B2 (en) | Analysis and characterization of patient signals | |
| CN114376564B (en) | Sleep staging method, system, device and medium based on ballistocardiogram signals | |
| CN108577830A (en) | A kind of user oriented sign information dynamic monitor method and dynamic monitor system | |
| CN113317794B (en) | Vital sign analysis method and system | |
| CN107296593A (en) | A kind of hemodynamic parameter acquisition methods and device | |
| US20220233077A1 (en) | Wearable health monitoring device | |
| US11617545B2 (en) | Methods and systems for adaptable presentation of sensor data | |
| CN108091391A (en) | Illness appraisal procedure, terminal device and computer-readable medium | |
| CN110200619A (en) | It is a kind of to establish major cardiovascular disease risks Early-warning Model method and device | |
| US20160135704A1 (en) | Matrix-Based Patient Signal Analysis | |
| CN109171703A (en) | The acquisition and processing method of electrocardiosignal based on wearable body-sensing net | |
| Mai et al. | Non-contact heartbeat detection based on ballistocardiogram using UNet and bidirectional long short-term memory | |
| Zhang et al. | Deep learning-based signal quality assessment for wearable ECGs | |
| CN108305688A (en) | Illness appraisal procedure, terminal device and computer-readable medium | |
| CN111278353A (en) | Method and system for detecting vital sign signal noise | |
| EP4031003A1 (en) | System and method for determining sleep stages based on non-cardiac body signals | |
| Khavas et al. | Robust heartbeat detection using multimodal recordings and ECG quality assessment with signal amplitudes dispersion | |
| CN110279411A (en) | A kind of cardiac data method for detecting abnormality, device, storage medium and terminal device | |
| CN114469004A (en) | Human sleep health monitoring method based on fiber bragg grating sensor |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190125 |
|
| RJ01 | Rejection of invention patent application after publication |