CN110432865A - A kind of sleep state classification system - Google Patents
A kind of sleep state classification system Download PDFInfo
- Publication number
- CN110432865A CN110432865A CN201910670219.5A CN201910670219A CN110432865A CN 110432865 A CN110432865 A CN 110432865A CN 201910670219 A CN201910670219 A CN 201910670219A CN 110432865 A CN110432865 A CN 110432865A
- Authority
- CN
- China
- Prior art keywords
- value
- num
- wave
- state
- threshold
- 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
- A61B5/4812—Detecting sleep stages or cycles
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention relates to a kind of sleep state classification systems.Existing sleep state classification system has that device structure is complicated and it is cumbersome to calculate assorting process.Basic data acquisition device and processor of the present invention.This system has the characteristics that, calculating few using basic data type and assorting process are simple, both simplified the structure of basic data acquisition device by reducing basic data acquisition type, facilitate user to dress to use, promote data accuracy, the requirement to processor is also reduced with assorting process by interruption and constantly acquisition frame data packet and simplified calculating, improving operational speed reduces hardware cost, ensure hardware operation stability, promotes usage experience.
Description
Technical field
The present invention relates to bedding fields, and in particular to a kind of sleep state classification system.
Background technique
The quality of sleep quality directly affects the people's health even service life, and more and more people are because of rhythm of life
Accelerate and there are problems that sleep disturbance.Medical research shows that insomnia will cause second day tired and exercise not harmony once in a while, for a long time
Insomnia can then bring attention that cannot concentrate, remember and obstacle occur and the consequences such as unable to do what one wishes that work.In order to promote sleep quality,
Brain electricity bioelectrical signals are acquired by professional sleep detection equipment, then extract brain electricity bioelectrical signals by different analysis methods
Various characteristic parameters finally carry out sleep state identification using each stage of the sorting algorithm to sleep, and then by specific
Stimulation mode promotes sleep quality, then how to judge that sleep state becomes the emphasis for implementing this method.Currently, leading sleep prison more
Survey is an important new technology in sleep medicine, referred to as diagnoses " goldstandard " of sleep disturbance disease.But due to using
The parameter monitored when Polysomnography is monitored is more, and the sensor of use is very more, and these sensors are most
It is attached to the head of user and each position of body, so user is uncomfortable, more in the presence of monitoring environment of being hospitalized during monitoring
Lead pastes the disadvantages of sensitive, sensor is easy to fall off.Although Portable sleep monitor overcomes Polysomnography and leads
The disadvantages of joining less, be easy to carry, but due to its power reguirements height, price, the exact requirements to electrode point and operation
The factors such as step is various, cause ordinary consumer that can not be easy to use, and then influence to promote and apply on commercial market.
Summary of the invention
In order to solve the deficiencies in the prior art, the present invention provides a kind of sleep state classification system, by reducing basic number
Calculating classification is carried out to user's sleep state according to type and quantity, has both been effectively reduced using difficulty, also ensures that user's sleep shape
State distinguishes precision, promotes usage experience.
The present invention is accomplished by the following way, a kind of sleep state classification system, the system comprises: basic data acquisition
Device and processor.Basic data acquisition device obtains basic data by portable wearable device；Processor is received from basis
The data of data collector simultaneously obtain user's sleep state by calculating.
The system at runtime, is realized by following steps:
The first step obtains basic data, in collection process, the basic data acquisition device by basic data acquisition device
Basic data is obtained by interval acquisition mode, the basic data of single continuous acquisition forms frame data packet；
Second step, processor are filtered frame data packet, and successively obtaining frequency is respectively 8~12Hz, 18
Alpha wave waveform diagram, the Beta wave waveform diagram, Sigma of~30Hz, 12~16Hz, 0.5~3Hz, 4~7Hz and 40~50Hz
Wave waveform diagram, Delta wave waveform diagram, Theta wave waveform diagram and EMG wave waveform diagram；
Step 3: being counted to the numerical value of collection point in each waveform diagram, and successively flat by the processing that takes absolute value, data
Level-one characteristic value parameter group and supplemental characteristic value parameter group, the level-one feature value parameter are obtained by cumulative calculation after sliding processing
Group includes: Alpha value, Beta value, Sigma value, Delta value, Theta value and EMG value, the supplemental characteristic value parameter group packet
Include Num-ARI value, Num-LCZ value, Num-Spindle value, Num-Delta value, Num-Alpha value and Num-Theta value；
Secondary characteristics value parameter group is obtained step 4: calculating by level-one characteristic value parameter group, the secondary characteristics value ginseng
Array includes: SVD value, DVB value, BEV value, TVA value and TVB value；
5th step classifies to user's sleep state by the parameter obtained in third step and the 4th step, so that user
Sleep state is classified as waking state, shallowly sleeps state, gently sleeps state, deep sleep or REM state.
Specifically, the basic data for limiting type is acquired by basic data acquisition device, and passes through processing based on this
Device, which calculates, obtains level-one characteristic value parameter group, supplemental characteristic value parameter group and secondary characteristics value parameter group, and to user's sleep shape
State is determined and is classified, and provides reference frame for subsequent sleeping operation.This system has, calculating few using basic data type
The simple feature with assorting process both simplifies the structure of basic data acquisition device, side by reducing basic data acquisition type
Just user, which dresses, uses, and promotes data accuracy, also reduces the requirement to processor by simplifying the process of calculating and classification,
Improving operational speed reduces hardware cost, it is ensured that hardware operation stability promotes usage experience.
Preferably, sorting out by decision tree to user's sleep state, at least five sections are set on the decision tree
Point classifies to sleep state, specifically:
Step 1: setting threshold values Num-ARI-Threshold, at first node, as Num-ARI > Num-ARI-
When Threshold, frame data coating is determined as artificial artefact, and exports the upper frame data of processing outward by processor
Otherwise the sleep state obtained when packet is transferred to second node；
Second step sets threshold values TVA-Threshold, TVB-Threshold at second node, as TVA > TVA-
When Threshold and TVB > TVB-Threshold, it is transferred to third node, otherwise, is transferred to the 5th node；
Third step sets threshold values Num-Delta-Threshold, at third node, as Num-Delta > Num-
When Delta-Threshold, processor determines that user's sleep state is in deep sleep and conveys outward, otherwise, is transferred to the 4th
Node；
4th step sets threshold values Num-Theta-Threshold and Num-Spindle-Threshold, in fourth node
Place, as Num-Theta > Num-Theta-Threshold and Num-Spindle < Num-Spindle-Threshold, place
Reason device determines user's sleep state in shallowly sleeping state and conveying outward, and otherwise, processor determines that user's sleep state is in light
It sleeps state and conveys outward；
5th step, setting threshold values BEV-Threshold and Num-Alpha-Threshold work as BEV at the 5th node
> BEV-Threshold and when Num-Alpha < Num-Alpha-Threshold, processor determines that user's sleep state is in
REM state simultaneously conveys outward, and otherwise, processor determines that user's sleep state is in waking state and conveys outward.
Use different parameters as judgment basis on each node of decision tree, and selected parameter can be distinguished effectively
Positioned at the sleep state of two side outlet of node, both effectively simplify deterministic process, promotes judging efficiency, also assure classification accuracy.
Preferably, there are the replaceable and independent data for realizing sleep state classification, specifically at same node:
At first node, parameter Num-LCZ alternative parameter Num-ARI can use, set threshold values Num-LCZ-Threshold, when
When Num-LCZ > Num-LCZ-Threshold, frame data coating is determined as artificial artefact, and defeated outward by processor
Otherwise the sleep state obtained when handling upper frame data packet out is transferred to second node.At second node, it can use
Parameter DVB alternative parameter TVA and TVB, setting threshold values DVB-Threshold are transferred to third as DVB > DVB-Threshold
Otherwise node is transferred to the 5th node.At third node, parameter SVD alternative parameter DVB can use, set threshold values SVD-
Threshold, as Num-Delta > Num-Delta-Threshold and SVD < SVD-Threshold, processor determines to use
Family sleep state is in deep sleep and conveys outward, otherwise, is transferred to fourth node.
Preferably, being divided into A, 30s≤A≤60s between the acquisition time of the adjacent frame data packet, frame data is formed
The single continuous acquisition time needed for packet is B, B≤A, and 5s≤B≤60s, and the data of the basic data acquisition device acquire frequency
Rate is C, 200samples/s≤C≤1000samples/s, and the basic data acquisition device is carried out when forming frame data packet
Data times of collection be n, n=C*B.Basic data acquisition device times of collection is corresponding with the data bulk of formation, and single is continuous
Resulting data form a frame data packet after the completion of acquisition, then obtain Alpha wave waveform diagram, Beta respectively by filter
Wave waveform diagram, Sigma wave waveform diagram, Delta wave waveform diagram, Theta wave waveform diagram and EMG wave waveform diagram.Work as frame data
When data bulk in packet is more, for draw formed waveform diagram data point it is also more so that waveform diagram draw it is more accurate, make
The waveform diagram that must be drawn more tends to the figure of true electric wave, provides more accurate data for processor.As C < 200samples/s
When, the waveform diagram of drafting can influence waveform diagram and draw precision because of data point negligible amounts；It is right as C > 1000samples/s
It is limited but higher to hardware requirement to draw precision improvement, while more High-frequency Interferences can be introduced, property is more relatively low.
Preferably, successively respectively to Alpha wave waveform diagram, Beta wave waveform diagram, Sigma wave waveform diagram, Delta wave wave
The corresponding numerical value in each collection point carries out the processing that takes absolute value in shape figure, Theta wave waveform diagram and EMG wave waveform diagram, and obtains
Data group Xm, later, to data group XmIt carries out data smoothing processing and obtains transit data group Ym,It crosses
The data length for crossing data group is m, 1≤m≤n-0.5*C, wherein 0.5*C is the length of data smoothing processing, finally, to mistake
Cross data group YmCarry out accumulation calculating and successively obtain the Alpha value, Beta value, Sigma value, Delta value, Theta value with
And EMG value.When acquiring basic data, user obviously exceeding variation range because of various factors basic data acquisition device can occur
Bounce numerical value, causes waveform diagram transient change larger, although will not have an impact to user's sleep state, will affect processor
Judge user's sleep state, cumulative mean processing is carried out to continuous 0.5*C data, effectively reduces single bounce numerical value to processing
Device judges the dormant influence of user.In addition, data group XmIt is that 0.5*C data are added up and put down to each data and thereafter
It handles, so that transit data group YmData bulk m can reduce to n-0.5*C.
Preferably, each parameter is by obtaining corresponding Mean-Alpha divided by parameter m in level-one characteristic value parameter group
Value, Mean-Beta value, Mean-Sigma value, Mean-Delta value, Mean-Theta value and Mean-EMG value, then calculate and obtain
Obtain SVD value, DVB value, BEV value, TVA value and TVB value, wherein SVD=Mean-Sigma/Mean-Delta, DVB=Mean-
Delta/Mean-Beta, BEV=Mean-Beta/Mean-EMG, TVA=Mean-Theta/Mean-Alpha, TVB=Mean-
Theta/Mean-Beta.Above-mentioned parameter is obtained by cumulative mean, jitter parameter is effectively reduced and user's sleep state is judged
The influence of accuracy.
Preferably, maximum threshold values corresponding to the setting of each waveform diagram and minimum threshold values, and so as to form data fluctuations model
It encloses, cumulative statistics is carried out beyond the collection point number of data fluctuations range to each waveform diagram, and form Num-ARI value, parameter
Num-ARI value is counted to the quantity of the bounce numerical value beyond preset range.
Preferably, the number for passing through x-axis to curve in each waveform diagram carries out cumulative statistics, and form Num-LCZ value, wave
In shape figure curve pass through x-axis number it is related with waveform frequency, parameter Num-LCZ value is counted to waveform frequency.
Preferably, counting to the spindle wave quantity in each waveform diagram, Num-Spindle value is obtained with this, it is described
Spindle wave is frequency in the Spindle waves that 11~16Hz, duration are greater than 0.5s and waveform is big in middle part small in ends.
Preferably, the duration occurred to Delta wave in Delta wave waveform diagram counts, Num-Delta is obtained with this
Value, for embodying the power of Delta wave.
Preferably, the duration occurred to Alpha wave in Alpha wave waveform diagram counts, Num-Alpha is obtained with this
Value, for embodying the power of Alpha wave.
Preferably, the duration occurred to Theta wave in Theta wave waveform diagram counts, Num-Theta is obtained with this
Value, for embodying the power of Theta wave.
Preferably, the basic data acquisition device conveys frame data packet, institute to processor at the same time
Processor is stated to receive frame data packet and judge user's sleep state so that user's sleep state slept in waking state, shallowly state,
It gently sleeps and switches between state, deep sleep and REM state.Basic data acquisition device obtains basic data with interval sampling mode,
Shorter time period is intercepted in each time section and carries out basic data acquisition, when both having ensured that the sampling time can be with entire sleep
Section, it is ensured that detection accuracy, the sample duration also effectively shortened in each interval time section, by reducing basic data quantity
It reduces the operand of processor, and then promotes the operation efficiency of processor, reduce the requirement to processor performance, be effectively reduced
Equipment cost.Basic data acquisition device can carry out the acquisition movement of one-time continuous in each interval time section, and utilize single
The basic data that continuous acquisition obtains forms frame data packet, and processor carries out analytical calculation to frame data packet and obtain use
User sleep state of the family in interval time section provides reference frame for subsequent sleeping operation.User's sleep state is divided into
Waking state shallowly sleeps state, gently sleeps state, deep sleep and REM state, with different numbers when user is in different conditions
According to feature, the data characteristics for have after summarizing with each sleep state to the data in frame data packet is compared,
And then obtain real-time user's sleep state.REM state refers to the rapid eye movement phase.
Preferably, state smoothing module is equipped in the processor, so that user's sleep state can only be along awake
State shallowly sleeps state, gently sleeps state and the sequence of deep sleep is back and forth smoothly switched, the REM state and it is described gently
The state of sleeping is smoothly switched, and in handoff procedure, when great-leap-forward switching occurs in user's sleep state, the processor is in shape
State smoothing module switches one by one under intervening along preset order and the revised user's sleep state of outside output smoothing.Work as list
When bounce data occurs in frame data packet, it will lead to user's sleep state and improper switching occur, such as user is directly from awake shape
State switches to deep sleep, this is the state change for running counter to actual conditions, at this point, being slept by the user that frame data packet obtains
Dormancy state and the practical sleep state of non-user at this time, for this purpose, state smoothing module is arranged in processor, so that adjacent
Frame data packet switches between obtaining user's sleep state according to default rule, and carries out to the improper switching of appearance strong
System corrigendum.Specifically, default rule are as follows: waking state shallowly sleeps state, gently sleeps and can only successively cut between state and deep sleep
Change, REM state only with gently sleep state and switched each other, such as gently the state of sleeping can only switch to it is shallow sleep state or deep sleep and
Waking state cannot be directly switch into, shallowly state is slept in another example waking state can only switch to and cannot be directly switch into and gently sleep shape
State or deep sleep, also for example REM state can only gently sleep state by gently sleeping from state switching and can only switch to.
Preferably, the Alpha wave, Beta wave, Sigma wave, Delta wave and Theta wave are brain wave, the EMG
Wave is myoelectricity wave.The type of basic data is less, and the performance requirement to basic data acquisition device has both been effectively reduced, and effectively simplification is set
Standby structure, reduces cost, subsequent processing workload is also effectively reduced, and then reduce the requirement to processor.
Preferably, basic data acquisition device is the glasses with data collecting assembly, the glasses include mirror holder and set
Temple in mirror holder both ends, the data collecting assembly include splitting the temporo electricity before the left front temporo electrode at the mirror holder both ends, the right side
Pole and bridge of the nose reference electrode, after glasses are worn in place, temporo electrode and bridge of the nose reference electrode before the left front temporo electrode, the right side
It is contradicted on station to be detected respectively.Glasses have the characteristics that easy to wear low with processing cost.When in use, when glasses are dressed
After in place, synchronous conflict is on station to be detected respectively for temporo electrode and bridge of the nose reference electrode before the left front temporo electrode, the right side, really
Protecting basic data acquisition device can continue accurately to acquire data, and frame data packet is periodically sent to processor.
Preferably, the basic data acquisition device is eyeshade, eyeshade includes cover and bandage, left front temporo electrode and it is right before
Temporo electrode is split at the both ends of the cover, and bridge of the nose reference electrode is set in the middle part of cover.Eyeshade can play the work to shut out the light
With, moreover it is possible to comfort is provided for user.
Protrusion of the invention is the utility model has the advantages that this system has, calculating few using basic data type and assorting process simple
The characteristics of, both simplified the structure of basic data acquisition device by reducing basic data acquisition type, and facilitated user to dress and use,
Data accuracy is promoted, also by interruption and constantly frame data packet is obtained and simplifies and just dissipate and assorting process reduces pair
The requirement of processor, improving operational speed reduce hardware cost, it is ensured that hardware operation stability promotes usage experience.
Detailed description of the invention
Fig. 1 is the Glasses structure schematic diagram；
Fig. 2 is the decision tree structure schematic diagram；
Fig. 3 is the Eye path structure schematic diagram；
In figure: 1, glasses, 2, mirror holder, 3, temple, 4, left front temporo electrode, 5, it is right before temporo electrode, 6, bridge of the nose reference electrode, 7,
Eyeshade.
Specific embodiment
Property feature is further described for the essence of the present invention with specific embodiment with reference to the accompanying drawings of the specification.
Embodiment one:
The present embodiment provides a kind of sleep state classification systems.
The system comprises: basic data acquisition device and processor.Basic data acquisition device acquisition basic data is simultaneously transmitted
To processor, processor receives the data from basic data acquisition device and obtains user's sleep state by calculating.
In the present embodiment, basic data acquisition device is the glasses 1 (as shown in Figure 1) with data collecting assembly, the eye
Mirror 1 includes mirror holder 2 and the temple 3 set on 2 both ends of mirror holder, and the data collecting assembly includes splitting at 2 both ends of mirror holder
Left front temporo electrode 4, it is right before temporo electrode 5 and be set to 2 middle part bridge of the nose reference electrode 6 of mirror holder, after glasses 1 are worn in place, institute
State left front temporo electrode 4, it is right before temporo electrode 5 and bridge of the nose reference electrode 6 contradicted on station to be detected respectively.The glasses 1 are logical
It crosses and acquires relevant rudimentary data with temporo electrode 5 and bridge of the nose reference electrode 6 before the left front temporo electrode 4 of user's body Surface Mount conjunction, the right side, and
It is conveyed to processor.Specifically, the left front temporo electrode 4, it is right before temporo electrode 5 and bridge of the nose reference electrode 6 respectively with it is close left front
Temporo and bridge of the nose fitting before temporo, the nearly right side form a kind of single lead collector of simple measurement electrical activity of brain.Basic data acquisition device pair
The Alpha wave, Beta wave, Sigma wave, Delta wave, Theta wave, EMG wave are acquired, wherein the Alpha wave,
Beta wave, Sigma wave, Delta wave and Theta wave are brain wave, and the EMG wave is myoelectricity wave.
In the present embodiment, the system at runtime, is realized by following steps:
The first step obtains basic data, in collection process, the basic data acquisition device by basic data acquisition device
Basic data is obtained by interval acquisition mode, the basic data of single continuous acquisition forms frame data packet.
Specifically, A is divided between the acquisition time of the adjacent frame data packet, A=60s, parameter A are adjacent frame data
Packet starts time interval when acquisition, and the single continuous acquisition time needed for forming frame data packet is B, B=10s, by user
Entire sleep procedure be divided into mono- section of 60s of interval period, and the data for carrying out in each interval period 10s acquire behaviour
Make, and so as to form frame data packet.
Second step, processor are filtered frame data packet, and successively obtaining frequency is respectively 8~12Hz, 18
Alpha wave waveform diagram, the Beta wave waveform diagram, Sigma of~30Hz, 12~16Hz, 0.5~3Hz, 4~7Hz and 40~50Hz
Wave waveform diagram, Delta wave waveform diagram, Theta wave waveform diagram and EMG wave waveform diagram.
Specifically, the data acquiring frequency of the basic data acquisition device is C, C=250samples/s, the basis number
It is n, n=C*B=250samples/s*10s=according to the data times of collection that collector is carried out when forming frame data packet
It 2500 times, when parameter n is non-integer, is rounded by the mode of rounding up.Basic data acquisition device collected single frames every time
Data packet is integrated data, is filtered decomposition to integrated data using filter and obtains the Alpha with differentiation frequency
Wave waveform diagram, Beta wave waveform diagram, Sigma wave waveform diagram, Delta wave waveform diagram, Theta wave waveform diagram and EMG wave waveform
Figure.Each waveform diagram has 2500 basic points to connect to be formed along time sequencing.It is preferable to use 8 rank Butterworth filters, cut ratio
Snow husband's filter or 4 rank Butterworth filters carry out the EEG signals in denoising extraction special frequency band to frame data packet
And electromyography signal.
Step 3: being counted to the numerical value of collection point in each waveform diagram, and successively flat by the processing that takes absolute value, data
Level-one characteristic value parameter group and supplemental characteristic value parameter group, the level-one feature value parameter are obtained by cumulative calculation after sliding processing
Group includes: Alpha value, Beta value, Sigma value, Delta value, Theta value and EMG value, the supplemental characteristic value parameter group packet
Include Num-ARI value, Num-LCZ value, Num-Spindle value, Num-Delta value, Num-Alpha value and Num-Theta value.
Specifically, due to Alpha wave waveform diagram, Beta wave waveform diagram, Sigma wave waveform diagram, Delta wave waveform diagram,
Theta wave waveform diagram and EMG wave waveform diagram are connected to be formed by 2500 basic points, so passing through each waveform diagram corresponding 2500
A basic point obtains corresponding Alpha value, Beta value, Sigma value, Delta value, Theta value and EMG value to calculate.With Alpha
For wave waveform diagram, Alpha wave waveform diagram includes 2500 basic points, when calculating, firstly, the numerical value to 2500 basic points carries out
Absolute value processing, obtains the data group X including 2500 data with thism；Later, to data group XmCarry out data smoothing processing simultaneously
Obtain transit data group Ym, in processing,The length of data smoothing processing is 0.5*C, transit data
The data length of group is m, m=n-0.5*C=2375, due to data group XmIn the 2376th data when being calculated,
Data bulk is less than 125 afterwards, so transit data group YmData length be 2375；Finally, to transit data group YmIn
2375 numerical value carry out accumulation calculating and obtaining the Alpha value.And so on obtain Beta value, Sigma value, Delta
Value, Theta value and EMG value.When the length 0.5*C of data smoothing processing is non-integer, it is rounded by the mode of rounding up.
Specifically, for the corresponding maximum threshold values of each waveform diagram setting and minimum threshold values, and so as to form with each waveform diagram
Corresponding data fluctuations range unites the basic point number for exceeding data fluctuations range in 2500 basic points in each waveform diagram
It counts and adds up and obtain Num-ARI value.
Specifically, the number for passing through x-axis to curve in each waveform diagram count and add up forming Num-LCZ value.
Specifically, the spindle wave quantity in each waveform diagram is counted, Num-Spindle value, the spinning is obtained with this
Hammering wave into shape is frequency in the Spindle waves that 11~16Hz, duration are greater than 0.5s and waveform is big in middle part small in ends.
Specifically, in the entire sleep procedure of user, Alpha wave, Beta wave, Sigma wave, Delta wave, Theta wave
And EMG wave the case where lacking individually can occur because user's sleep state is different, when basic data acquisition device can not collect
When signal, corresponding basic point numerical value is 0 to the waveform this moment, leads to not to form corresponding waveform diagram, and then calculating is facilitated to correspond to
Level-one feature value parameter.The duration occurred to Delta wave in Delta wave waveform diagram counts, and obtains Num-Delta with this
Value.The duration occurred to Alpha wave in Alpha wave waveform diagram counts, and obtains Num-Alpha value with this.To Theta wave wave
The duration that Theta wave occurs in shape figure is counted, and obtains Num-Theta value with this.
Secondary characteristics value parameter group is obtained step 4: calculating by level-one characteristic value parameter group, the secondary characteristics value ginseng
Array includes: SVD value, DVB value, BEV value, TVA value and TVB value.It is calculated by level-one characteristic value parameter group and obtains Mean-
Alpha value, Mean-Beta value, Mean-Sigma value, Mean-Delta value, Mean-Theta value and Mean-EMG value, in turn
Obtain SVD value, DVB value, BEV value, TVA value and TVB value.
Specifically, Mean-Alpha value, Mean-Beta value, Mean-Sigma value, Mean-Delta value, Mean-Theta
Value and Mean-EMG value are the average value of each waveform basic point numerical value in frame data packet, for example, Mean-Alpha value is Alpha
Value divided by parameter m=2375 obtain, and so on obtain Mean-Beta value, Mean-Sigma value, Mean-Delta value, Mean-
Theta value and Mean-EMG value.
Specifically, SVD=Mean-Sigma/Mean-Delta, for indicate in frame data packet Sigma wave signal with
Strong and weak comparison between Delta wave signal, when parameter SVD value is bigger, illustrates that Sigma wave signal is better than Delta wave signal, instead
It illustrates that Sigma wave signal is weaker than Delta wave signal when parameter SVD value is smaller.
Specifically, DVB=Mean-Delta/Mean-Beta, for indicate in frame data packet Delta wave signal with
Strong and weak comparison between Beta wave signal, when parameter DVB value is bigger, illustrates that Delta wave signal is better than Beta wave signal, conversely,
When parameter DVB value is smaller, illustrate that Delta wave signal is weaker than Beta wave signal.
Specifically, BEV=Mean-Beta/Mean-EMG, for indicating Beta wave signal and EMG wave in frame data packet
Strong and weak comparison between signal, when parameter BEV value is bigger, illustrates that Beta wave signal is better than EMG wave signal, conversely, working as parameter BEV
Value gets over hour, illustrates that Beta wave signal is weaker than EMG wave signal.
Specifically, TVA=Mean-Theta/Mean-Alpha, for indicate in frame data packet Theta wave signal with
Strong and weak comparison between Alpha wave signal, when parameter TVA value is bigger, illustrates that Theta wave signal is better than Alpha wave signal, instead
It illustrates that Theta wave signal is weaker than Alpha wave signal when parameter TVA value is smaller.
Specifically, TVB=Mean-Theta/Mean-Beta, for indicate in frame data packet Theta wave signal with
Strong and weak comparison between Beta wave signal, when parameter TVB value is bigger, illustrates that Theta wave signal is better than Beta wave signal, conversely,
When parameter TVB value is smaller, illustrate that Theta wave signal is weaker than Beta wave signal.
5th step classifies to user's sleep state by the parameter obtained in third step and the 4th step, so that user
Sleep state is classified as waking state, shallowly sleeps state, gently sleeps state, deep sleep or REM state.By decision tree to user
Sleep state is sorted out, and at least five nodes are set on the decision tree and are classified (as shown in Figure 2) to sleep state, tool
Body:
At first node, threshold values Num-ARI-Threshold is set, as Num-ARI > Num-ARI-Threshold
When, frame data coating is determined as artificial artefact, and is exported by processor and obtained when handling upper frame data packet outward
Sleep state be otherwise transferred to second node.Artificial artefact refers to the interference signal for influencing electricity physiological signal, including physiology
Artefact and equipment artefact.When artificial artefact occurs, basic data will appear the bounce beyond normal range (NR), by parameter
Num-ARI value determines whether to generate artificial artefact compared with size between corresponding threshold value Num-ARI-Threshold, specifically,
As Num-ARI > Num-ARI-Threshold, frame data coating is determined as artificial artefact, which cancels,
And the sleep state that obtains when conveying upper frame data packet outward can be with conversely, then artificial artefact is not present in specification this moment
It is further analyzed.
At second node, threshold values TVA-Threshold, TVB-Threshold are set, as TVA > TVA-Threshold
And when TVB > TVB-Threshold, it is transferred to third node, otherwise, is transferred to the 5th node.When user is in REM state or regains consciousness
When state, human thinking is active and mainly issues alpha wave and beta wave, and when user, which is in, shallowly sleeps state, brain is mainly sent out
Theta wave out, when user, which is in, gently to sleep, brain mainly issues K- complex wave (sigma wave and delta wave) and frequency 11
Between~16Hz, the spindle wave (sleep spindle) of last for several seconds, when user is in sound sleep, brain mainly issues delta
Wave.Therefore, it is in REM shape that user can be distinguished by the strong or weak relation of alpha, beta and theta, delta wave signal
Any state in state and waking state is in and shallowly sleeps state, gently sleeps any state in state and deep sleep, so, when
TVA > TVA-Threshold and when TVB > TVB-Threshold, theta wave signal is stronger at this time or delta wave is stronger or
Beta Bobbi is weaker, illustrates that user's human thinking is inactive and is in state of shallowly sleeping, gently sleeps state or deeper sleep shape
State, conversely, beta wave signal or alpha wave are believed at this time as TVA≤TVA-Threshold and TVB≤TVB-Threshold
It is number stronger, illustrate that the human thinking of user is active and in REM state or waking state.
At third node, threshold values Num-Delta-Threshold is set, as Num-Delta > Num-Delta-
When Threshold, processor determines that user's sleep state is in deep sleep and conveys outward, otherwise, is transferred to fourth node.When
When user is in deep sleep, delta wave signal becomes strong, and delta wave epoch is also elongated, illustrates that user is in sound sleep
State conversely, then illustrate that the sleep state of user is not at deep sleep, and enters fourth node to shallowly sleeping state and gently sleep shape
State is determined.
At fourth node, threshold values Num-Theta-Threshold and Num-Spindle-Threshold are set, when
Num-Theta > Num-Theta-Threshold and when Num-Spindle < Num-Spindle-Threshold, processor is sentenced
User's sleep state is determined in shallowly sleeping state and conveying outward, and otherwise, processor determines that user's sleep state is in and gently sleeps state
And it conveys outward.When user, which is in, shallowly sleeps state, theta wave signal becomes strong, when user, which is in, gently sleeps state, spindle
Wave number amount increases, as Num-Theta > Num-Theta-Threshold and Num-Spindle < Num-Spindle-
When Threshold, theta wave epoch is elongated at this time and spindle wave number amount is reduced, then illustrates that user is in and shallowly sleep
State, conversely, working as Num-Theta≤Num-Theta-Threshold and Num-Spindle >=Num-Spindle-
When Threshold, theta wave epoch shortens at this time and spindle wave number amount increases, then illustrates that user is in and gently sleep
State.
At the 5th node, threshold values BEV-Threshold and Num-Alpha-Threshold are set, as BEV > BEV-
When Threshold and Num-Alpha < Num-Alpha-Threshold, processor determines that user's sleep state is in REM state
And convey outward, otherwise, processor determines that user's sleep state is in waking state and conveys outward.When user is in awake shape
When state, human thinking is active and issues beta wave and alpha wave, and EMG signal amplitude variation at this time is bigger.When user is in
When REM state, brain mainly issues theta and beta wave, and EMG wave signal becomes strong, but when amplitude variation is relatively awake changes
It is smaller, so, as BEV > BEV-Threshold and Num-Alpha < Num-Alpha-Threshold, beta wave at this time
Signal is better than EMG wave signal and Alpha wave epoch shortens, then illustrates that user is in REM state, conversely, illustrating to use
Family is in waking state.
The real-time sleep state of the corresponding user of frame data packet is judged by the above method, and provides reference to subsequent operation
With the foundation of control.
In the present embodiment, the basic data acquisition device conveys frame data to processor at the same time
Packet, the processor receives frame data packet and judges user's sleep state, so that user's sleep state is slept in waking state, shallowly
State is gently slept and is switched between state, deep sleep and REM state.
In the present embodiment, state smoothing module is equipped in the processor, so that user's sleep state can only edge
Waking state shallowly sleeps state, gently sleeps state and the sequence of deep sleep is back and forth smoothly switched, the REM state and institute
The light state of sleeping is stated to be smoothly switched, in handoff procedure, when there is great-leap-forward switching in user's sleep state, the processor
Switch simultaneously the revised user's sleep state of output smoothing outward one by one along preset order under the intervention of state smoothing module.
It is back and forth smoothly switched by limiting the relational implementation that can directly switch between each sleep state, specifically, waking state sleeps shape with shallow
Between state, shallowly sleeps state and gently sleep between state, gently sleep between state and deep sleep, gently sleeping and reciprocally switch between state and REM state.
When the sleep that processor is calculated and obtained to the No.1 frame data packet smoothly obtained along the time and No. two frame data packets
When state is not belonging to a kind of handoff relation of any of the above-described permission, then the sleep state obtained using No.1 data packet is basic point, knot
The sleep state for closing the acquisition of two number packets is that direction carries out the adaptability switching for meeting above-mentioned handoff relation, for example, when passing through
The sleep state that No.1 frame data packet obtains is shallowly to sleep state and be deep by the sleep state that No. two frame data packets obtain
When sleeping state, then it is not belonging to a kind of handoff relation of any of the above-described permission, is then basic point towards deep sleep shallowly to sleep state
Direction carries out adaptability switching, shallowly sleeps state and gently sleeps and reciprocally switches between state, then is obtained by No. two frame data packets
Sleep state is defined as gently sleeping state, and obtains whether result normally judges basic point as No. three frame data packets are judged.
It is to be appreciated that parameter A can also be 30s, 35s, 40s, 50s, 60s etc., wanted as long as meeting 30s≤A≤60s
It asks.
It is to be appreciated that parameter B can also be 5s, 7s, 11s, 20s, 30s etc., as long as meeting B≤A and 5s≤B≤60s
Requirement.
It is to be appreciated that parameter C can also for 200samples/s, 300samples/s, 500samples/s,
1000samples/s etc., as long as meeting 200samples/s≤C≤1000samples/s requirement.
It is to be appreciated that parameter m can also be 1,10,500 etc., as long as meeting 1≤m≤n-0.5*C requirement.
Embodiment two:
Compared to embodiment one, the present embodiment provides another sleep state classification systems.
By the judgment basis replacement in the 5th step at first node are as follows: at first node, set threshold values Num-LCZ-
Threshold, as Num-LCZ > Num-LCZ-Threshold, frame data coating is determined as artificial artefact, and passes through
Processor exports the sleep state obtained when handling upper frame data packet outward and is otherwise transferred to second node.When artificial artefact
When appearance, the number that waveform passes through x-axis in each waveform diagram, which will appear, to be obviously increased, by parameter Num-LCZ value and corresponding threshold
Size relatively determines whether to generate artificial artefact between value Num-LCZ-Threshold, specifically, as Num-LCZ > Num-
When LCZ-Threshold, frame data coating is determined as artificial artefact, which cancels, and conveys upper one outward
The sleep state obtained when frame data packet can further be divided conversely, then artificial artefact is not present in specification this moment
Analysis.
The other feature of sleep state classification system described in the present embodiment is consistent with embodiment one, no longer repeats.
Embodiment three:
By the judgment basis replacement in the 5th step at second node are as follows: at second node, set threshold values DVB-
Threshold is transferred to third node as DVB > DVB-Threshold, otherwise, is transferred to the 5th node.When user is in REM
When state or waking state, human thinking is active and mainly issues alpha wave and beta wave, when user, which is in, shallowly sleeps state,
Brain mainly issues theta wave, and when user, which is in, gently to sleep, brain mainly issues K- complex wave (sigma wave and delta wave)
And frequency is between 11~16Hz, the spindle wave (sleep spindle) of last for several seconds, and when user is in sound sleep, brain master
Issue delta wave.Therefore, user can be distinguished by the strong or weak relation of alpha, beta and theta, delta wave signal
It is to be in REM, waking state group, is in N1, N2, N3 phase group, so, as DVB > DVB-Threshold, at this time
Theta wave signal is stronger or delta wave is stronger or beta Bobbi is weaker, illustrates that user's human thinking is inactive and sleeps shape in shallow
State gently sleeps state or deeper sleep state, conversely, as DVB≤DVB-Threshold, at this time beta wave signal or
Alpha wave signal is stronger, illustrates that the human thinking of user is active and is in REM state or waking state.
The other feature of sleep state classification system described in the present embodiment is consistent with embodiment one, no longer repeats.
Example IV:
By the judgment basis replacement in the 5th step at third node are as follows: at third node, set threshold values SVD-
Threshold, as Num-Delta > Num-Delta-Threshold and SVD < SVD-Threshold, processor determines to use
Family sleep state is in deep sleep and conveys outward, otherwise, is transferred to fourth node.It gently sleeps state when user is in or shallowly sleeps shape
When state, sigma wave signal becomes strong, and when user is in deep sleep, delta wave signal becomes strong, so, as Num-Delta >
When Num-Delta-Threshold and SVD < DVB-Threshold, at this time delta wave signal occur time it is elongated and
Delta wave signal is better than sigma wave signal, illustrates that user is in deep sleep, conversely, working as Num-Delta≤Num-Delta-
Threshold and when SVD >=DVB-Threshold, the time that delta wave signal occurs at this time shortens and delta wave signal is weak
In sigma wave signal, illustrates that user is in and gently sleep state or shallowly sleep state, be transferred to fourth node and carry out classification judgement.
The other feature of sleep state classification system described in the present embodiment is consistent with embodiment one, no longer repeats.
Embodiment five:
Compared to embodiment one, the present embodiment provides another basic data acquisition devices.
In the present embodiment, basic data acquisition device is the eyeshade 7 (as shown in Figure 3) with data collecting assembly, the eye
Cover includes cover and bandage, and the data collecting assembly includes splitting the temporo electricity before the left front temporo electrode 4 at the cover both ends, the right side
Pole 5 and be set to bridge of the nose reference electrode 6 in the middle part of cover, after eyeshade is worn in place, the left front temporo electrode 4, it is right before temporo electrode 5
And bridge of the nose reference electrode 6 is contradicted respectively on station to be detected.
When in use, eyeshade is fixed on default station by bandage, so that cover is bonded with user's forehead.The eyeshade
7 by the left front temporo electrode 4 that is closed with user's body Surface Mount, it is right before temporo electrode 5 and bridge of the nose reference electrode 6 acquire relevant rudimentary number
According to, and conveyed to processor.Specifically, the left front temporo electrode 4, it is right before temporo electrode 5 and bridge of the nose reference electrode 6 respectively and closely
Temporo and bridge of the nose fitting before left front temporo, the nearly right side form a kind of single lead collector of simple measurement electrical activity of brain.
The other feature of sleep state classification system described in the present embodiment is consistent with embodiment one, no longer repeats.
Claims (10)
1. a kind of sleep state classification system, which is characterized in that the system comprises:
Basic data acquisition device obtains basic data by portable wearable device；
Processor receives the data from basic data acquisition device and obtains user's sleep state by calculating；
The system at runtime, is realized by following steps:
The first step obtains basic data by basic data acquisition device, and in collection process, the basic data acquisition device passes through
It is spaced acquisition mode and obtains basic data, the basic data of single continuous acquisition forms frame data packet；
Second step, processor are filtered frame data packet, and successively obtain frequency be respectively 8~12Hz, 18~
Alpha wave waveform diagram, the Beta wave waveform diagram, Sigma wave of 30Hz, 12~16Hz, 0.5~3Hz, 4~7Hz and 40~50Hz
Waveform diagram, Delta wave waveform diagram, Theta wave waveform diagram and EMG wave waveform diagram；
Step 3: counted to the numerical value of collection point in each waveform diagram, and successively by the processing that takes absolute value, data smoothing at
Level-one characteristic value parameter group and supplemental characteristic value parameter group, the level-one characteristic value parameter group packet are obtained by accumulation calculating after reason
Include: Alpha value, Beta value, Sigma value, Delta value, Theta value and EMG value, the supplemental characteristic value parameter group include
Num-ARI value, Num-LCZ value, Num-Spindle value, Num-Delta value, Num-Alpha value and Num-Theta value；
Secondary characteristics value parameter group, the secondary characteristics value parameter group are obtained step 4: calculating by level-one characteristic value parameter group
It include: SVD value, DVB value, BEV value, TVA value and TVB value；
5th step classifies to user's sleep state by the parameter obtained in third step and the 4th step, so that user sleeps
State is classified as waking state, shallowly sleeps state, gently sleeps state, deep sleep or REM state.
2. a kind of sleep state classification system according to claim 1, which is characterized in that slept by decision tree to user
State is sorted out, and at least five nodes are set on the decision tree and are classified to sleep state, specifically:
Step 1: setting threshold values Num-ARI-Threshold, at first node, as Num-ARI > Num-ARI-
When Threshold, frame data coating is determined as artificial artefact, and exports the upper frame data of processing outward by processor
Otherwise the sleep state obtained when packet is transferred to second node；
Second step sets threshold values TVA-Threshold, TVB-Threshold, at second node, as TVA > TVA-
When Threshold and TVB > TVB-Threshold, it is transferred to third node, otherwise, is transferred to the 5th node；
Third step sets threshold values Num-Delta-Threshold, at third node, as Num-Delta > Num-Delta-
When Threshold, processor determines that user's sleep state is in deep sleep and conveys outward, otherwise, is transferred to fourth node；
4th step sets threshold values Num-Theta-Threshold and Num-Spindle-Threshold, at fourth node, when
Num-Theta > Num-Theta-Threshold and when Num-Spindle < Num-Spindle-Threshold, processor is sentenced
User's sleep state is determined in shallowly sleeping state and conveying outward, and otherwise, processor determines that user's sleep state is in and gently sleeps state
And it conveys outward；
5th step sets threshold values BEV-Threshold and Num-Alpha-Threshold, at the 5th node, as BEV >
When BEV-Threshold and Num-Alpha < Num-Alpha-Threshold, processor determines that user's sleep state is in REM
State simultaneously conveys outward, and otherwise, processor determines that user's sleep state is in waking state and conveys outward.
3. a kind of sleep state classification system according to claim 2, which is characterized in that setting threshold values Num-LCZ-
Threshold, at first node, as Num-LCZ > Num-LCZ-Threshold, frame data coating determines for people
Work artefact, and export the sleep state obtained when handling upper frame data packet outward by processor and be otherwise transferred to the second section
Point；Alternatively, setting threshold values DVB-Threshold, as DVB > DVB-Threshold, is transferred to third section at second node
Otherwise point is transferred to the 5th node；Alternatively, a kind of sleep state classification system according to claim 2, which is characterized in that
Threshold values SVD-Threshold is set, at third node, as Num-Delta > Num-Delta-Threshold and SVD <
When SVD-Threshold, processor determines that user's sleep state is in deep sleep and conveys outward, otherwise, is transferred to Section four
Point.
4. a kind of sleep state classification system according to claim 1-3, which is characterized in that the adjacent single frames
A, 30s≤A≤60s are divided between the acquisition time of data packet, the single continuous acquisition time needed for forming frame data packet is B, B
≤ A, and 5s≤B≤60s, the data acquiring frequency of the basic data acquisition device are C, 200samples/s≤C≤
1000samples/s, the data times of collection that the basic data acquisition device is carried out when forming frame data packet are n, n=C*
B。
5. a kind of sleep state classification system according to claim 4, which is characterized in that successively respectively to Alpha wave wave
In shape figure, Beta wave waveform diagram, Sigma wave waveform diagram, Delta wave waveform diagram, Theta wave waveform diagram and EMG wave waveform diagram
The corresponding numerical value in each collection point carries out the processing that takes absolute value, and obtains data group Xm, later, to data group XmCarry out data smoothing
It handles and obtains transit data group Ym,The data length of transit data group be m, 1≤m≤n-0.5*C,
Wherein, 0.5*C is the length of data smoothing processing, finally, to transit data group YmCarry out accumulation calculating and successively acquisition described in
Alpha value, Beta value, Sigma value, Delta value, Theta value and EMG value.
6. a kind of sleep state classification system according to claim 5, which is characterized in that each in level-one characteristic value parameter group
Parameter is by obtaining corresponding Mean-Alpha value, Mean-Beta value, Mean-Sigma value, Mean-Delta divided by parameter m
Value, Mean-Theta value and Mean-EMG value, then calculate and obtain SVD value, DVB value, BEV value, TVA value and TVB value,
In,
SVD=Mean-Sigma/Mean-Delta,
DVB=Mean-Delta/Mean-Beta,
BEV=Mean-Beta/Mean-EMG,
TVA=Mean-Theta/Mean-Alpha,
TVB=Mean-Theta/Mean-Beta.
7. a kind of sleep state classification system according to claim 1-3, which is characterized in that set to each waveform diagram
Fixed corresponding maximum threshold values and minimum threshold values, and so as to form data fluctuations range, data fluctuations range is exceeded to each waveform diagram
Collection point number carry out cumulative statistics, and form Num-ARI value；Alternatively, to curve in each waveform diagram pass through the number of x-axis into
The cumulative statistics of row, and form Num-LCZ value；Alternatively, counting to the spindle wave quantity in each waveform diagram, Num- is obtained with this
Spindle value, the spindle wave are frequency in the shuttle that 11~16Hz, duration are greater than 0.5s and waveform is big in middle part small in ends
Shape wave；Alternatively, the duration occurred to Delta wave in Delta wave waveform diagram counts, Num-Delta value is obtained with this；Or
Person, the duration occurred to Alpha wave in Alpha wave waveform diagram count, and obtain Num-Alpha value with this；Alternatively, right
The duration that Theta wave occurs in Theta wave waveform diagram is counted, and obtains Num-Theta value with this.
8. a kind of sleep state classification system according to claim 1-3, which is characterized in that the basic data
Collector conveys frame data packet to processor at the same time, and the processor receives frame data packet and judges to use
Family sleep state, so that user's sleep state is slept state in waking state, shallowly, gently slept between state, deep sleep and REM state
Switching is equipped with state smoothing module in the processor so that user's sleep state can only along waking state, shallowly sleep shape
State gently sleeps state and the sequence of deep sleep is back and forth smoothly switched, and the REM state carries out flat with the gently state of sleeping
Sliding cutting changes, and in handoff procedure, when great-leap-forward switching occurs in user's sleep state, the processor is in state smoothing processing mould
Block switches one by one under intervening along preset order and the revised user's sleep state of outside output smoothing；Alternatively, the Alpha
Wave, Beta wave, Sigma wave, Delta wave and Theta wave are brain wave, and the EMG wave is myoelectricity wave.
9. a kind of sleep state classification system according to claim 1-3, which is characterized in that basic data acquisition
Device is equipped with data collecting assembly, before the data collecting assembly includes the left front temporo electrode (4) split, is right temporo electrode (5) with
And bridge of the nose reference electrode (6), after basic data acquisition device is worn in place, temporo electrode (5) before the left front temporo electrode (4), the right side
And bridge of the nose reference electrode (6) is contradicted respectively on station to be detected.
10. a kind of sleep state classification system according to claim 9, which is characterized in that the basic data acquisition device
For glasses (1), the glasses include mirror holder (2) and temple (3), and left front temporo electrode (4) He Youqian temporo electrode (5) splits described
The both ends of mirror holder (2), bridge of the nose reference electrode (6) are set in the middle part of mirror holder (2)；Alternatively, the basic data acquisition device is eyeshade
(7), eyeshade includes cover and bandage, and left front temporo electrode (4) He Youqian temporo electrode (5) splits in the both ends of the cover, the bridge of the nose
Reference electrode (6) is set in the middle part of cover.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910670219.5A CN110432865A (en) | 2019-07-24 | 2019-07-24 | A kind of sleep state classification system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910670219.5A CN110432865A (en) | 2019-07-24 | 2019-07-24 | A kind of sleep state classification system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110432865A true CN110432865A (en) | 2019-11-12 |
Family
ID=68431301
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910670219.5A Pending CN110432865A (en) | 2019-07-24 | 2019-07-24 | A kind of sleep state classification system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110432865A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110917464A (en) * | 2019-12-12 | 2020-03-27 | 喜临门家具股份有限公司 | Vibration sleep-aiding system for inducing Alpha wave generation |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104068849A (en) * | 2014-07-02 | 2014-10-01 | 西安交通大学 | Method for automatically identifying and extracting K complex waves in sleep brain waves |
| CN204033325U (en) * | 2014-09-09 | 2014-12-24 | 喜临门家具股份有限公司 | A kind of sleep safeguards system |
| US20140378808A1 (en) * | 2013-06-22 | 2014-12-25 | Zinus, Inc. | Comfortably Securing Physiological Sensors Onto A Person During Sleep |
| CN105496363A (en) * | 2015-12-15 | 2016-04-20 | 浙江神灯生物科技有限公司 | Method for classifying sleep stages on basis of sleep EGG (electroencephalogram) signal detection |
| CN106236088A (en) * | 2016-08-31 | 2016-12-21 | 博睿康科技（常州）股份有限公司 | A kind of electrode for encephalograms |
| CN106377251A (en) * | 2016-09-21 | 2017-02-08 | 广州视源电子科技股份有限公司 | Sleep state identification model training method and sleep state identification model training system based on electroencephalogram |
| CN106997106A (en) * | 2016-01-22 | 2017-08-01 | 周常安 | Spectacle physiological sensing, object wearing device and Glasses structure |
| CN108089451A (en) * | 2017-01-11 | 2018-05-29 | 浙江师范大学 | A kind of intelligent sleep environmental control system based on real-time neuroelectricity signal processing |
| CN109464130A (en) * | 2019-01-09 | 2019-03-15 | 浙江强脑科技有限公司 | Sleep householder method, system and readable storage medium storing program for executing |
-
2019
- 2019-07-24 CN CN201910670219.5A patent/CN110432865A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140378808A1 (en) * | 2013-06-22 | 2014-12-25 | Zinus, Inc. | Comfortably Securing Physiological Sensors Onto A Person During Sleep |
| CN104068849A (en) * | 2014-07-02 | 2014-10-01 | 西安交通大学 | Method for automatically identifying and extracting K complex waves in sleep brain waves |
| CN204033325U (en) * | 2014-09-09 | 2014-12-24 | 喜临门家具股份有限公司 | A kind of sleep safeguards system |
| CN105496363A (en) * | 2015-12-15 | 2016-04-20 | 浙江神灯生物科技有限公司 | Method for classifying sleep stages on basis of sleep EGG (electroencephalogram) signal detection |
| CN106997106A (en) * | 2016-01-22 | 2017-08-01 | 周常安 | Spectacle physiological sensing, object wearing device and Glasses structure |
| CN106236088A (en) * | 2016-08-31 | 2016-12-21 | 博睿康科技（常州）股份有限公司 | A kind of electrode for encephalograms |
| CN106377251A (en) * | 2016-09-21 | 2017-02-08 | 广州视源电子科技股份有限公司 | Sleep state identification model training method and sleep state identification model training system based on electroencephalogram |
| CN108089451A (en) * | 2017-01-11 | 2018-05-29 | 浙江师范大学 | A kind of intelligent sleep environmental control system based on real-time neuroelectricity signal processing |
| CN109464130A (en) * | 2019-01-09 | 2019-03-15 | 浙江强脑科技有限公司 | Sleep householder method, system and readable storage medium storing program for executing |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110917464A (en) * | 2019-12-12 | 2020-03-27 | 喜临门家具股份有限公司 | Vibration sleep-aiding system for inducing Alpha wave generation |
| CN110917464B (en) * | 2019-12-12 | 2022-04-12 | 喜临门家具股份有限公司 | Vibration sleep-aiding system for inducing Alpha wave generation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104720748B (en) | A kind of sleep stage determines method and system | |
| CN104720746B (en) | A kind of sleep stage determines method and system | |
| CN105054951B (en) | A kind of fatigue driving eeg monitoring method based on frequency of wink identification | |
| CN105943015A (en) | Wearable heart rate variability (HRV) monitoring device with active noise reduction function | |
| CN106333691A (en) | The method and device for judging sleep state, rest state and motion state of human body | |
| CN105105714A (en) | Sleep period separating method and system | |
| CN106236083A (en) | Sleep state removes the equipment of eye electricity artefact in analyzing | |
| CN106725382A (en) | Sleep state judgement system and method based on action and HRV measurements | |
| CN106236079A (en) | Electric and the sleep monitor eyeshield of eye electricity compound detection and sleep monitor method for brain | |
| CN110584611A (en) | Sleep monitoring system with good energy-saving effect | |
| WO2021180028A1 (en) | Method, apparatus and device for evaluating sleep quality on basis of high-frequency electroencephalography, and storage medium | |
| CN108852304A (en) | Sleeping quality analyzing device and method based on EEG signals | |
| CN107890339A (en) | A kind of sleep stage detection method and wearable sleep stage detection means | |
| CN106214129B (en) | Terminate method, equipment, sleep server and the intelligence wearing terminal of sleep inferior | |
| CN110433384A (en) | A kind of sleep music system | |
| CN106108855A (en) | Snoring system based on sound of snoring identification and classification | |
| CN107582051A (en) | A kind of animal mood brain electricity analytical equipment | |
| CN109044280A (en) | A kind of sleep stage method and relevant device | |
| CN110585551A (en) | Light awakening control system based on sleep staging | |
| CN106730232B (en) | A kind of intelligence awakening method and system | |
| CN110432865A (en) | A kind of sleep state classification system | |
| Yin et al. | Experimental research on real-time acquisition and monitoring of wearable EEG based on TGAM module | |
| CN109620158A (en) | Method, intelligent terminal and the storage device of Sleep-Monitoring | |
| CN108992040A (en) | A kind of sleep quality state monitoring method and device | |
| CN110585554B (en) | Sleep-aiding music control system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191112 |