CN104706318B - A kind of sleep analysis method and device - Google Patents
A kind of sleep analysis method and device Download PDFInfo
- Publication number
- CN104706318B CN104706318B CN201310687525.2A CN201310687525A CN104706318B CN 104706318 B CN104706318 B CN 104706318B CN 201310687525 A CN201310687525 A CN 201310687525A CN 104706318 B CN104706318 B CN 104706318B
- Authority
- CN
- China
- Prior art keywords
- monitoring period
- period
- activity
- threshold
- complexity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
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 invention discloses a kind of sleep analysis method and device, including：Obtain more axle acceleration datas of the person to be monitored gathered according to preset sample frequency within every height monitoring period；More axle acceleration datas of the person to be monitored gathered within every height monitoring period are based respectively on, determine activity of the person to be monitored within every height monitoring period；And it is directed to the period is monitored per height respectively, according to the activity in the son monitoring period that period window corresponding to the sub- monitoring period includes, activity characteristic value of the person to be detected within every height monitoring period is determined, and determines the dynamic threshold of activity characteristic value in the monitoring period；Respectively by the activity characteristic value in every height monitoring period compared with the dynamic threshold, it is sleep state or the first sleep analysis result of waking state to obtain the person to be monitored within every height monitoring period.Using method provided in an embodiment of the present invention, the accuracy rate for classification of keeping alert while in bed is improved.
Description
Technical field
The present invention relates to signal analysis field, more particularly to a kind of sleep analysis method and device.
Background technology
Sleep study is hypnosphy and the important component of electroencephalography, and in the world today scientific research focus
One of.Polysomnography is current internationally recognized sleep monitor " goldstandard ", by being attached to person to be monitored
Electrode, the indexs such as dynamic, dynamic, the brain electricity of leg of the blood oxygen of tester, electrocardio, eye are recorded, to judge the sleep quality of person to be monitored.It is but more
Lead hypnogram monitoring device to involve great expense, with the development of science and technology the research to sleep monitor is gradually to miniaturization and family oriented
Direction develop.By more axle acceleration datas during gathering user's sleep, the acceleration during being slept using user is less than
Acceleration this feature when clear-headed, carries out analysis of keeping alert while in bed.
In existing technical scheme, more axle acceleration datas of person to be monitored are gathered by acceleration transducer, this is more
Axle acceleration data is divided into multiple subdata sections, will be right in subdata section by being judged to obtain to more axle acceleration datas
The number of activities in period answered, as the activity in the period, and when the activity in the period is more than one
During fixed threshold, it is waking state within the period to determine person to be monitored, is sleep state, to each subdata section otherwise
Analyzed, finally obtain sleep analysis result of the person to be monitored within the entirely monitoring period.Can also be to the sleep analysis
As a result carry out subsequent treatment, such as, when person to be monitored occur in prolonged sleep procedure it is of short duration clear-headed, then it is this is short
Temporarily clear-headed state is judged to sleep, when person to be monitored it is prolonged it is clear-headed during there is of short duration sleep, then it is this is short
The state temporarily slept is judged to regain consciousness.
But for different persons to be monitored, sleep habit is different, compares during presumable person's sleep to be monitored
Peace and quiet, and it is more dynamic during some possibility are slept, and it is relatively low that this results in the classification accuracy of keeping alert while in bed based on fixed threshold judgement.
The content of the invention
The embodiment of the present invention provides a kind of sleep analysis method and device, to solve present in prior art based on solid
Determine threshold decision classification accuracy of keeping alert while in bed it is relatively low the problem of.
The embodiment of the present invention provides a kind of obtain and treated according to what preset sample frequency gathered within every height monitoring period
More axle acceleration datas of human observer, more axle acceleration datas include multiple more axle accelerations, wherein, a monitoring time
Section includes more height monitoring period；
More axle acceleration datas of the person to be monitored gathered within every height monitoring period are based respectively on, determine institute
State activity of the person to be monitored within every height monitoring period；
It is directed to respectively and the period is monitored per height, the son prison included according to period window corresponding to the sub- monitoring period
The activity surveyed in the period, activity characteristic value of the person to be detected within every height monitoring period is determined, wherein, son
Monitoring period window corresponding to the period includes the sub- monitoring period and its front and rear some height monitoring period；
According to the activity characteristic value in the multiple sub- monitoring period in the monitoring period, the monitoring time is determined
The dynamic threshold of activity characteristic value in section；
Respectively by the activity characteristic value in every height monitoring period compared with the dynamic threshold, obtain described
Person to be monitored is sleep state or the first sleep analysis result of waking state within every height monitoring period.
Using method provided in an embodiment of the present invention, based on the activity of sub- monitoring period, and the sub- monitoring time
The activity for the others son monitoring period that period window corresponding to section includes, it is determined that the activity of son monitoring period is special
Value indicative；According to the activity characteristic value of son monitoring period in the whole monitoring period, it is determined that entirely activity in the monitoring period
The dynamic threshold of measure feature value；Person to be monitored is judged according to dynamic threshold in the monitoring period sleep of every height or is regained consciousness.
Compared to prior art, the accuracy rate for classification of keeping alert while in bed is improved.
The embodiment of the present invention also provides a kind of sleep analysis device, including：
Data capture unit, for obtain according to preset sample frequency every height monitoring the period in gather it is to be monitored
More axle acceleration datas of person, more axle acceleration datas include multiple more axle accelerations, wherein, a monitoring period bag
Include more height monitoring periods；
Activity determining unit, for being based respectively on the more of the person to be monitored gathered within every height monitoring period
Axle acceleration data, determine activity of the person to be monitored within every height monitoring period；
Activity characteristic value determining unit, the period is monitored per height for being directed to respectively, according to the sub- monitoring period
The activity in the son monitoring period that corresponding period window includes, determines that the person to be detected monitors the time in every height
Section in activity characteristic value, wherein, period window corresponding to the sub- monitoring period include the sub- monitoring period and its
Front and rear some height monitoring periods；
Dynamic threshold determining unit, for special according to the activity in the multiple sub- monitoring period in the monitoring period
Value indicative, determine the dynamic threshold of activity characteristic value in the monitoring period；
Processing unit, carried out for every height to be monitored to the activity characteristic value in the period respectively with the dynamic threshold
Compare, it is sleep state or the first sleep analysis knot of waking state to obtain the person to be monitored within every height monitoring period
Fruit.
Other features and advantage will illustrate in the following description, also, partly become from specification
Obtain it is clear that or being understood by implementing the application.The purpose of the application and other advantages can be by the explanations write
Specifically noted structure is realized and obtained in book, claims and accompanying drawing.
Brief description of the drawings
Accompanying drawing is used for providing a further understanding of the present invention, and a part for constitution instruction, implements with the present invention
Example is used to explain the present invention together, is not construed as limiting the invention.In the accompanying drawings：
Fig. 1 is one of flow chart of sleep analysis method provided in an embodiment of the present invention；
Fig. 2 is the two of the flow chart of sleep analysis method provided in an embodiment of the present invention；
Fig. 3 is the flow chart that the more axle acceleration datas of low frequency provided in an embodiment of the present invention carry out sleep analysis；
Fig. 4 is the structural representation of sleep analysis device provided in an embodiment of the present invention.
Embodiment
In order to provide improve treat human observer keep alert while in bed classification accuracy rate implementation, the embodiments of the invention provide one
The preferred embodiments of the present invention are illustrated by kind sleep analysis method and device below in conjunction with Figure of description, it will be appreciated that
Preferred embodiment described herein is merely to illustrate and explain the present invention, and is not intended to limit the present invention.And do not conflicting
In the case of, the feature in embodiment and embodiment in the application can be mutually combined.
The embodiment of the present invention provides a kind of sleep analysis method, idiographic flow as shown in figure 1, including：
Step 101, the multiaxis for obtaining the person to be monitored gathered according to preset sample frequency within every height monitoring period
Acceleration information, more axle acceleration datas include multiple more axle accelerations, wherein, a monitoring period supervises including more height
Survey the period.
Step 102, the more axle acceleration datas for being based respectively on the person to be monitored gathered within every height monitoring period,
Determine activity of the person to be monitored within every height monitoring period.
Step 103, is directed to per height respectively the monitoring period, according to period window bag corresponding to the sub- monitoring period
The activity in the son monitoring period included, determines activity characteristic value of the person to be detected within every height monitoring period,
Wherein, period window corresponding to the sub- monitoring period includes the sub- monitoring period and its front and rear some height monitoring time
Section.
Step 104, according to the activity characteristic value in the monitoring period in the plurality of sub- monitoring period, determine the monitoring
The dynamic threshold of activity characteristic value in period.
Step 105, respectively by every height monitor the period in activity characteristic value compared with the dynamic threshold, obtain
To the person to be monitored the first sleep analysis result in the period for sleep state or waking state is monitored in every height.
In the embodiment of the present invention, more axle acceleration datas can be acquired by acceleration transducer, can will be right
The sleep analysis of person to be monitored the whole night monitors the period as one, and the monitoring period is divided into more height monitors the periods,
More axle acceleration datas of every height monitoring period are analyzed.
Using method provided in an embodiment of the present invention, based on the activity of sub- monitoring period, and the sub- monitoring time
The activity for the others son monitoring period that period window corresponding to section includes, it is determined that the activity of son monitoring period is special
Value indicative；According to the activity characteristic value of son monitoring period in the whole monitoring period, it is determined that entirely activity in the monitoring period
The dynamic threshold of measure feature value；Person to be monitored is judged according to dynamic threshold in the monitoring period sleep of every height or is regained consciousness.
Compared to prior art, the accuracy rate for classification of keeping alert while in bed is improved.
Below in conjunction with the accompanying drawings, method and device provided by the invention and corresponding system are retouched in detail with specific embodiment
State.Method detailed step as shown in Fig. 2 including：
Step 201, the person to be monitored to collection are monitoring the progress bandpass filtering of more axle acceleration datas in the period.With
More axle accelerations caused by the physical activity of family have a frequency range, and bandpass filtering is carried out to more axle accelerations, primarily to
Remove interference.
Step 202, by filtered data, the multiaxis in every height monitoring period is accelerated according to preset sample frequency
Degrees of data is sampled, wherein, the sub- monitoring period can be set to 1 minute.
Step 203, based on more axle acceleration datas after sampling, determine person to be monitored within every height monitoring period
Activity.
Wherein, when more axle accelerations of sampled point in the sub- monitoring period are more than predetermined acceleration threshold value, it is determined that should
Person to be monitored is movable at the time of the sampled point corresponds to, and the sampled point determines according to preset sample frequency；Will be in the sub- prison
The total degree of the person to be monitored activity determined in the period is surveyed, is defined as activity of the person to be monitored in the sub- monitoring period.
Calculating on activity has a variety of methods, can select threshold method, zero passage method, area-method etc., the present embodiment choosing
The determination of activity is carried out with threshold method.
Step 204, determine the sub activity monitored in the period that period window corresponding to the sub- monitoring period includes
Average, the variance of amount, respectively as corresponding to the sub- monitoring period in period window activity average, variance, and really
Activity is more than the number of the son monitoring period of default activity in the fixed period window.
Period window can be set to 5 minutes, the sub- monitoring period be 1 minute, then corresponding to the sub- monitoring period when
Between son monitoring period for including of section window, be the son monitoring period corresponding to each 2 minutes before and after the current son monitoring period
And the current son monitoring period, determine that this 5 sons monitor the average and variance of period activity.
Step 205, to activity is more than in the average of activity, variance and the period window in the period window
The number of the son monitoring period of default activity, is weighted summation, obtains the movable measure feature in the sub- monitoring period
Value, is referred to as PS values by activity characteristic value.
When the determination of PS values can also monitor the activity of period, current son monitoring according to current son in period window
Between the logarithm of section activity and the maximum of other sub- monitoring period activities in the period window and variable quantity etc. come
It is determined that weight coefficient is empirical value.
Step 206, determine to monitor average and variance that more height in the period monitor multiple PS values in the periods, respectively
Average and variance as PS values in the monitoring period.
Step 207, average and variance based on PS values in the monitoring period, determine the dynamic threshold of PS values.
When the average of activity characteristic value is more than the first default activity characteristic value with variance sum in the monitoring period
During threshold value, the first default eigenvalue threshold is defined as to the dynamic threshold of activity characteristic value in the monitoring period；
When the average of activity characteristic value presets movable measure feature no more than first with variance sum in the monitoring period
It is worth threshold value, and the average of activity characteristic value is less than the second default activity characteristic value threshold with variance sum in the monitoring period
During value, the second default activity eigenvalue threshold is defined as to the dynamic threshold of activity characteristic value in the monitoring period,
Wherein, the second default activity eigenvalue threshold is less than the first default activity eigenvalue threshold；
When the average of activity characteristic value presets movable measure feature no more than first with variance sum in the monitoring period
It is worth threshold value, and the average of activity characteristic value presets eigenvalue threshold not less than second with variance sum in the monitoring period
When, the average of activity characteristic value in the monitoring period and variance sum are defined as movable measure feature in the monitoring period
The dynamic threshold of value.
Step 208, by the PS values in the sub- monitoring period compared with the dynamic threshold of PS values in the monitoring period,
Obtain first sleep analysis result of the person to be monitored within the sub- monitoring period.
Wherein, the activity characteristic value of group monitoring period is more than the dynamic of activity characteristic value in the monitoring period
During threshold value, it is waking state within the sub- monitoring period to determine person to be monitored；
Dynamic threshold of the activity characteristic value of group monitoring period no more than activity characteristic value in the monitoring period
During value, it is waking state within the sub- monitoring period to determine person to be monitored.
In the method that above-described embodiment provides, other method can also be used on dynamic threshold value determination method, such as
Maximum between-cluster variance algorithm, it is a kind of adaptive threshold method, based on multiple threshold values, for each threshold value, by clear-headed and sleep
As two classifications, inter-class variance is calculated, using the maximum threshold value of the inter-class variance for causing two classes as final threshold value；Entropy threshold
Method, determine that the monitoring period is clear-headed or sleep probability, and corresponding entropy per height according to different threshold values, it is determined that energy
Enough make the maximum threshold value of entropy；Minimum error method, the method derive from Bayes minimal error sorting techniques, and Eb (T) is target class
（It is clear-headed）Mistake assigns to background classes（Sleep）Probability, Eo (T) is background classes（Sleep）Mistake assigns to target class（It is clear-headed）Probability, always
Probability of error E (T)=Eb (T)+Eo (T), E (T) is taken minimum value, as optimal classification method.
In addition, some users may read a book before sleeping, play the custom such as mobile phone, more axle acceleration numbers that sort of activity is brought
It is relatively low according to frequency, it is possible to that sleep state can be judged as only with above-mentioned processing procedure, therefore, the embodiment of the present invention also provides
It is a kind of for the more axle acceleration datas of low frequency carry out sleep analysis method, specific steps as shown in figure 3, including：
Step 301, more axle acceleration datas to person to be monitored in the monitoring period carry out LPF, obtain every height
Monitor the more axle acceleration datas of low frequency in the period.
Step 302, the complexity for determining the more axle acceleration datas of low frequency in the monitoring period per height respectively.
Wherein, the determination method of complexity can have a variety of, and this programme can first determine that low frequency is more in the sub- monitoring period
The extreme value number and the difference of adjacent maximum and minimum of axle acceleration data, the two parameters are weighted summation,
It is determined that son monitors the complexity of the more axle acceleration datas of low frequency in the period.
Step 303, according to monitoring the period in more height monitoring the period in the more axle acceleration data complexities of low frequency
Average and variance, determine the dynamic threshold of the complexity monitoring period Nei.Determination on complexity dynamic threshold can
With identical with above-mentioned PS values dynamic threshold value determination method, will not be repeated here.
Step 304, whether the complexity is more than according to the complexity of the more axle acceleration datas of low frequency in the sub- monitoring period
Dynamic threshold, it is the second sleep analysis knot of sleep state or waking state to determine person to be monitored within the sub- monitoring period
Fruit.
When the average of complexity and variance sum are more than the first default complexity threshold in the monitoring period, by this
One default complexity threshold is defined as the dynamic threshold of complexity in the monitoring period；
When the average of complexity presets complexity threshold, and the prison no more than first with variance sum in the monitoring period
When the average of complexity is less than the second default complexity threshold with variance sum in the survey period, by the second default complexity threshold
Value is defined as the dynamic threshold of complexity in the monitoring period, wherein, it is first pre- that the second default complexity threshold is less than this
If complexity threshold；
When the average of complexity presets complexity threshold, and the prison no more than first with variance sum in the monitoring period
, will be multiple in the monitoring period when average of complexity presets complexity threshold with variance sum not less than second in the survey period
The average and variance sum of miscellaneous degree are defined as the dynamic threshold of complexity in the monitoring period.
Step 305, for above-mentioned first sleep analysis result it is sleep state and the second sleep analysis result is clear-headed shape
The son monitoring period of state, the 3rd sleep analysis result for determining the sub- monitoring period is waking state.Wherein, the 3rd is slept
The dormancy analysis result sleep analysis result final within the monitoring period as person to be monitored.It is also based on the 3rd sleep analysis
As a result the sleep state for treating human observer is further analyzed.
Based on same inventive concept, the sleep analysis method provided according to the above embodiment of the present invention, correspondingly, the present invention
Another embodiment additionally provides sleep analysis device, and apparatus structure schematic diagram is as shown in figure 4, specifically include：
Data capture unit 401, treated for obtaining according to what preset sample frequency gathered within every height monitoring period
More axle acceleration datas of human observer, more axle acceleration datas include multiple more axle accelerations, wherein, a monitoring period
The period is monitored including more height；
Activity determining unit 402, for being based respectively on the person's to be monitored gathered within every height monitoring period
More axle acceleration datas, determine activity of the person to be monitored within every height monitoring period；
Activity characteristic value determining unit 403, the period is monitored per height for being directed to respectively, according to the sub- monitoring time
The activity in the son monitoring period that period window corresponding to section includes, determines that the person to be detected monitors the time in every height
Section in activity characteristic value, wherein, period window corresponding to the sub- monitoring period include the sub- monitoring period and its
Front and rear some height monitoring periods；
Dynamic threshold determining unit 404, for according to the activity in the plurality of sub- monitoring period in the monitoring period
Characteristic value, determine the dynamic threshold of activity characteristic value in the monitoring period；
Processing unit 405, enter for every height to be monitored into the activity characteristic value in the period respectively with the dynamic threshold
Row compares, and it is sleep state or the first sleep analysis knot of waking state to obtain the person to be monitored within every height monitoring period
Fruit.
Further, activity determining unit 402, accelerate specifically for the multiaxis when sampled point in the sub- monitoring period
When degree is more than predetermined acceleration threshold value, it is movable, the sampled point root at the time of sampled point corresponds to determine the person to be monitored
It is determined according to preset sample frequency；And the total degree for person's activity to be monitored being determined within the sub- monitoring period, really
It is set to activity of the person to be monitored in the sub- monitoring period.
Further, activity characteristic value determining unit 403, specifically for determining the time corresponding to the sub- monitoring period
Average, the variance for the activity in the son monitoring period that section window includes, during respectively as corresponding to the sub- monitoring period
Between in section window activity average, variance；
Determine that activity is more than the sub number for monitoring the period for presetting activity in the period window；
To activity is more than default activity in the average of activity, variance and the period window in the period window
The number of the son monitoring period of amount, is weighted summation, obtains the activity characteristic value in the sub- monitoring period；
Dynamic threshold determining unit 404, is specifically used for：It is it is determined that more in the plurality of sub- monitoring period in the monitoring period
The average and variance of individual activity characteristic value, respectively as the average and variance of activity characteristic value in the monitoring period；
When the average of activity characteristic value is more than the first default activity characteristic value with variance sum in the monitoring period
During threshold value, the first default eigenvalue threshold is defined as to the dynamic threshold of activity characteristic value in the monitoring period；
When the average of activity characteristic value presets movable measure feature no more than first with variance sum in the monitoring period
It is worth threshold value, and the average of activity characteristic value is less than the second default activity characteristic value threshold with variance sum in the monitoring period
During value, the second default activity eigenvalue threshold is defined as to the dynamic threshold of activity characteristic value in the monitoring period,
Wherein, the second default activity eigenvalue threshold is less than the first default activity eigenvalue threshold；
When the average of activity characteristic value presets movable measure feature no more than first with variance sum in the monitoring period
It is worth threshold value, and the average of activity characteristic value presets eigenvalue threshold not less than second with variance sum in the monitoring period
When, the average of activity characteristic value in the monitoring period and variance sum are defined as movable measure feature in the monitoring period
The dynamic threshold of value.
Further, processing unit 405, specifically for group monitor the period activity characteristic value be more than the monitoring when
Between in section during the dynamic threshold of activity characteristic value, it is waking state within the sub- monitoring period to determine person to be monitored；And
When the activity characteristic value of group monitoring period is not more than the dynamic threshold of activity characteristic value in the monitoring period, it is determined that
Person to be monitored is waking state within the sub- monitoring period.
Further, said apparatus, in addition to：Low-frequency data processing unit 406, for the plurality of sub- monitoring period
Interior more axle acceleration datas carry out LPF according to predeterminated frequency, and the low frequency multiaxis obtained in every height monitoring period adds
Speed data；
The complexity of the more axle acceleration datas of low frequency in the monitoring period per height is determined respectively；
According in the plurality of sub- monitoring period in the monitoring period the more axle acceleration data complexities of low frequency it is equal
Value and variance, determine the dynamic threshold of the complexity monitoring period Nei；
Whether it is more than the dynamic threshold of the complexity according to the complexity of the more axle acceleration datas of low frequency in the sub- monitoring period
Value, it is sleep state or the second sleep analysis result of waking state to determine the person to be monitored within the sub- monitoring period；
When for the first sleep analysis result be sleep state and the second sleep analysis result is the son monitoring of waking state
Between section, the 3rd sleep analysis result for determining the sub- monitoring period is waking state.
Further, low-frequency data processing unit 406, the more axle acceleration datas of low frequency in a son monitoring period are determined
Complexity, specifically for determining the extreme value number of the more axle acceleration datas of low frequency and adjacent pole in the sub- monitoring period
The difference of big value and minimum；And to the extreme value number of the more axle acceleration datas of low frequency in the sub- monitoring period and adjacent
The difference of maximum and minimum, summation is weighted, it is determined that son monitors the complexity of the more axle acceleration datas of low frequency in the period.
Further, low-frequency data processing unit 406, the dynamic threshold of complexity in the monitoring period is determined, specifically
It is for when the average of complexity and variance sum are more than the first default complexity threshold in the monitoring period, this is first pre-
If complexity threshold is defined as the dynamic threshold of complexity in the monitoring period；
When the average of complexity presets complexity threshold, and the prison no more than first with variance sum in the monitoring period
When the average of complexity is less than the second default complexity threshold with variance sum in the survey period, by the second default complexity threshold
Value is defined as the dynamic threshold of complexity in the monitoring period, wherein, it is first pre- that the second default complexity threshold is less than this
If complexity threshold；
When the average of complexity presets complexity threshold, and the prison no more than first with variance sum in the monitoring period
, will be multiple in the monitoring period when average of complexity presets complexity threshold with variance sum not less than second in the survey period
The average and variance sum of miscellaneous degree are defined as the dynamic threshold of complexity in the monitoring period.
The respective handling step that the function of above-mentioned each unit may correspond in flow shown in Fig. 1 to Fig. 3, it is no longer superfluous herein
State.
To sum up it is somebody's turn to do, scheme provided in an embodiment of the present invention, obtains and monitor the period in every height according to preset sample frequency
More axle acceleration datas of the person to be monitored of interior collection；It is based respectively on the person to be monitored gathered within every height monitoring period
More axle acceleration datas, determine the person to be monitored every height monitoring the period in activity；And it is directed to respectively per height
The period is monitored, the activity in the period is monitored according to the son that period window corresponding to the sub- monitoring period includes, really
Activity characteristic value of the fixed person to be detected within every height monitoring period；And according to the plurality of sub- prison in the monitoring period
The activity characteristic value surveyed in the period, determine the dynamic threshold of activity characteristic value in the monitoring period；Respectively will be each
Activity characteristic value in the son monitoring period obtains the person to be monitored when every height monitors compared with the dynamic threshold
Between be sleep state or the first sleep analysis result of waking state in section.Using method provided in an embodiment of the present invention, compare
In prior art, the accuracy rate for classification of keeping alert while in bed is improved.
The sleep analysis device that embodiments herein is provided can be realized by computer program.Those skilled in the art
It should be appreciated that above-mentioned Module Division mode is only one kind in numerous Module Division modes, if being divided into other moulds
Block or non-division module, all should be within the protection domain of the application as long as sleep analysis device has above-mentioned function.
The application is with reference to method, the equipment according to the embodiment of the present application（System）And the flow of computer program product
Figure and/or block diagram describe.It should be understood that can be by every first-class in computer program instructions implementation process figure and/or block diagram
Journey and/or the flow in square frame and flow chart and/or block diagram and/or the combination of square frame.These computer programs can be provided
The processors of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce
A raw machine so that produced by the instruction of computer or the computing device of other programmable data processing devices for real
The device for the function of being specified in present one flow of flow chart or one square frame of multiple flows and/or block diagram or multiple square frames.
These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which produces, to be included referring to
Make the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one square frame of block diagram or
The function of being specified in multiple square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that counted
Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, so as in computer or
The instruction performed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in individual square frame or multiple square frames.
Obviously, those skilled in the art can carry out the essence of various changes and modification without departing from the present invention to the present invention
God and scope.So, if these modifications and variations of the present invention belong to the scope of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to comprising including these changes and modification.
Claims (14)
- A kind of 1. sleep analysis method, it is characterised in that including：Obtain more axle acceleration datas of the person to be monitored gathered according to preset sample frequency within every height monitoring period, institute Stating more axle acceleration datas includes multiple more axle accelerations, wherein, a monitoring period includes more height monitoring period；More axle acceleration datas of the person to be monitored gathered within every height monitoring period are based respectively on, it is determined that described treat Activity of the human observer within every height monitoring period；It is directed to respectively and the period is monitored per height, according to during the son monitoring that period window corresponding to the sub- monitoring period includes Between activity in section, determine activity characteristic value of the person to be detected within every height monitoring period, specifically include：Really Average, the variance for the activity in the son monitoring period that period window corresponding to the fixed sub- monitoring period includes, respectively As corresponding to the sub- monitoring period in period window activity average, variance；It is movable in the period window to determine Amount is more than the number of the son monitoring period of default activity；To the average of activity, variance in the period window and it is somebody's turn to do Activity is more than the number of the son monitoring period of default activity in period window, is weighted summation, obtains the sub- prison The activity characteristic value surveyed in the period；Wherein, period window corresponding to the sub- monitoring period includes the sub- monitoring period And its front and rear some height monitoring period；According to the activity characteristic value in the multiple sub- monitoring period in the monitoring period, determine in the monitoring period The dynamic threshold of activity characteristic value；Respectively by every height monitor the period in activity characteristic value with the dynamic threshold compared with, obtain described in wait to supervise Survey person is sleep state or the first sleep analysis result of waking state within every height monitoring period.
- 2. the method as described in claim 1, it is characterised in that based on the person to be monitored gathered within the sub- monitoring period More axle acceleration datas, determine activity of the person to be monitored in the sub- monitoring period, specifically include：When more axle accelerations of sampled point in the sub- monitoring period are more than predetermined acceleration threshold value, the person to be monitored is determined It is movable at the time of the sampled point corresponds to, the sampled point is determined according to preset sample frequency；The total degree of person's activity to be monitored will be determined within the sub- monitoring period, be defined as the person to be monitored in the son Monitor the activity of period.
- 3. the method as described in claim 1, it is characterised in that according in the multiple sub- monitoring period in the monitoring period Activity characteristic value, determine it is described monitoring the period in activity characteristic value dynamic threshold, specifically include：It is determined that the average and variance of multiple activity characteristic values in the monitoring period in the multiple sub- monitoring period, respectively Average and variance as activity characteristic value in the monitoring period；When the average of activity characteristic value is more than the first default activity characteristic value threshold with variance sum in the monitoring period During value, the described first default eigenvalue threshold is defined as to the dynamic threshold of activity characteristic value in the monitoring period；When the average of activity characteristic value presets activity characteristic value no more than first with variance sum in the monitoring period Threshold value, and the average of activity characteristic value is less than the second default activity characteristic value threshold with variance sum in the monitoring period During value, the described second default activity eigenvalue threshold is defined as to the dynamic threshold of activity characteristic value in the monitoring period Value, wherein, the described second default activity eigenvalue threshold is less than the described first default activity eigenvalue threshold；When the average of activity characteristic value presets activity characteristic value no more than first with variance sum in the monitoring period Threshold value, and the average of activity characteristic value presets eigenvalue threshold not less than second with variance sum in the monitoring period When, the average of activity characteristic value in the monitoring period is defined as activity in the monitoring period with variance sum The dynamic threshold of characteristic value.
- 4. the method as described in claim 1, it is characterised in that by the activity characteristic value in the sub- monitoring period and the prison Survey the dynamic threshold of activity characteristic value in the period to be compared, obtain person to be monitored first sleeping within the sub- monitoring period Dormancy analysis result, is specifically included：When the activity characteristic value of group monitoring period is more than the dynamic threshold of activity characteristic value in the monitoring period, It is waking state within the sub- monitoring period to determine person to be monitored.
- 5. the method as described in claim 1, it is characterised in that obtain and monitor the period in every height according to preset sample frequency After more axle acceleration datas of the person to be monitored of interior collection, in addition to：LPF is carried out according to predeterminated frequency to more axle acceleration datas in the multiple sub- monitoring period, obtained each The more axle acceleration datas of low frequency in the son monitoring period；The complexity of the more axle acceleration datas of low frequency in the monitoring period per height is determined respectively；According to it is described monitoring the period in the multiple sub- monitoring period in the more axle acceleration data complexities of low frequency it is equal Value and variance, determine the dynamic threshold of the complexity in the monitoring period；Whether it is more than the dynamic threshold of the complexity according to the complexity of the more axle acceleration datas of low frequency in the sub- monitoring period, It is sleep state or the second sleep analysis result of waking state that the person to be monitored, which is determined, within the sub- monitoring period；For the first sleep analysis result be sleep state and the second sleep analysis result is that the sub of waking state monitors the period, The 3rd sleep analysis result for determining the sub- monitoring period is waking state.
- 6. method as claimed in claim 5, it is characterised in that determine the more axle acceleration numbers of low frequency in a son monitoring period According to complexity, specifically include：Determine in the sub- monitoring period extreme value number of the more axle acceleration datas of low frequency and adjacent maximum with it is minimum The difference of value；Extreme value number and adjacent maximum and minimum to the more axle acceleration datas of low frequency in the sub- monitoring period Difference, be weighted summation, it is determined that son monitoring the period in the more axle acceleration datas of low frequency complexity.
- 7. method as claimed in claim 5, it is characterised in that the dynamic threshold of complexity in the monitoring period is determined, Specifically include：When average and the variance sum of complexity in the monitoring period are more than the first default complexity threshold, by described the One default complexity threshold is defined as the dynamic threshold of complexity in the monitoring period；When the average of complexity presets complexity threshold, and the prison no more than first with variance sum in the monitoring period When the average of complexity is less than the second default complexity threshold with variance sum in the survey period, by the described second default complexity Threshold value is defined as the dynamic threshold of complexity in the monitoring period, wherein, the described second default complexity threshold is less than institute State the first default complexity threshold；When the average of complexity presets complexity threshold, and the prison no more than first with variance sum in the monitoring period When the average of complexity presets complexity threshold with variance sum not less than second in the survey period, by the monitoring period The average and variance sum of complexity are defined as the dynamic threshold of complexity in the monitoring period.
- A kind of 8. sleep analysis device, it is characterised in that including：Data capture unit, for obtaining the person's to be monitored gathered according to preset sample frequency within every height monitoring period More axle acceleration datas, more axle acceleration datas include multiple more axle accelerations, wherein, a monitoring period includes more Height monitors the period；Activity determining unit, the multiaxis for being based respectively on the person to be monitored gathered within every height monitoring period add Speed data, determine activity of the person to be monitored within every height monitoring period；Activity characteristic value determining unit, the period is monitored per height for being directed to respectively, it is corresponding according to the sub- monitoring period Period window include son monitoring the period in activity, determine the person to be detected every height monitoring the period in Activity characteristic value, the activity characteristic value determining unit, specifically for determining the time corresponding to the sub- monitoring period Average, the variance for the activity in the son monitoring period that section window includes, during respectively as corresponding to the sub- monitoring period Between in section window activity average, variance；When determining that activity is more than the son monitoring of default activity in the period window Between section number；To in the period window in the average of activity, variance and the period window activity be more than it is default The number of the son monitoring period of activity, is weighted summation, obtains the activity characteristic value in the sub- monitoring period；Its In, period window corresponding to the sub- monitoring period includes the sub- monitoring period and its front and rear some height monitoring time Section；Dynamic threshold determining unit, for according to the movable measure feature in the multiple sub- monitoring period in the monitoring period Value, determine the dynamic threshold of activity characteristic value in the monitoring period；Processing unit, compared for every height to be monitored into the activity characteristic value in the period respectively with the dynamic threshold Compared with it is sleep state or the first sleep analysis knot of waking state to obtain the person to be monitored within every height monitoring period Fruit.
- 9. device as claimed in claim 8, it is characterised in that the activity determining unit, specifically for when the sub- monitoring When more axle accelerations of sampled point are more than predetermined acceleration threshold value in period, determine that the person to be monitored is corresponding in the sampled point At the time of be movable, the sampled point is determined according to preset sample frequency；And will be within the sub- monitoring period really The total degree of fixed person's activity to be monitored, is defined as activity of the person to be monitored in the sub- monitoring period.
- 10. device as claimed in claim 8, it is characterised in that the dynamic threshold determining unit, be specifically used for：It is determined that the average and variance of multiple activity characteristic values in the monitoring period in the multiple sub- monitoring period, respectively Average and variance as activity characteristic value in the monitoring period；When the average of activity characteristic value is more than the first default activity characteristic value threshold with variance sum in the monitoring period During value, the described first default eigenvalue threshold is defined as to the dynamic threshold of activity characteristic value in the monitoring period；When the average of activity characteristic value presets activity characteristic value no more than first with variance sum in the monitoring period Threshold value, and the average of activity characteristic value is less than the second default activity characteristic value threshold with variance sum in the monitoring period During value, the described second default activity eigenvalue threshold is defined as to the dynamic threshold of activity characteristic value in the monitoring period Value, wherein, the described second default activity eigenvalue threshold is less than the described first default activity eigenvalue threshold；When the average of activity characteristic value presets activity characteristic value no more than first with variance sum in the monitoring period Threshold value, and the average of activity characteristic value presets eigenvalue threshold not less than second with variance sum in the monitoring period When, the average of activity characteristic value in the monitoring period is defined as activity in the monitoring period with variance sum The dynamic threshold of characteristic value.
- 11. device as claimed in claim 8, it is characterised in that the processing unit, the period is monitored specifically for group When activity characteristic value is more than the dynamic threshold of activity characteristic value in the monitoring period, determine person to be monitored in the sub- prison It is waking state to survey in the period.
- 12. device as claimed in claim 8, it is characterised in that described device, in addition to：Low-frequency data processing unit, is used for LPF is carried out according to predeterminated frequency to more axle acceleration datas in the multiple sub- monitoring period, obtains every height prison The more axle acceleration datas of low frequency surveyed in the period；The complexity of the more axle acceleration datas of low frequency in the monitoring period per height is determined respectively；According to it is described monitoring the period in the multiple sub- monitoring period in the more axle acceleration data complexities of low frequency it is equal Value and variance, determine the dynamic threshold of the complexity in the monitoring period；Whether it is more than the dynamic threshold of the complexity according to the complexity of the more axle acceleration datas of low frequency in the sub- monitoring period, It is sleep state or the second sleep analysis result of waking state that the person to be monitored, which is determined, within the sub- monitoring period；For the first sleep analysis result be sleep state and the second sleep analysis result is that the sub of waking state monitors the period, The 3rd sleep analysis result for determining the sub- monitoring period is waking state.
- 13. device as claimed in claim 12, it is characterised in that the low-frequency data processing unit, determine a son monitoring The complexity of the more axle acceleration datas of low frequency in period, specifically for determining that low frequency multiaxis accelerates in the sub- monitoring period The extreme value number and the difference of adjacent maximum and minimum of degrees of data；And to low frequency multiaxis in the sub- monitoring period The extreme value number and the difference of adjacent maximum and minimum of acceleration information, are weighted summation, it is determined that the son monitoring time The complexity of the more axle acceleration datas of low frequency in section.
- 14. device as claimed in claim 12, it is characterised in that the low-frequency data processing unit, when determining the monitoring Between in section complexity dynamic threshold, specifically for when in the monitoring period average of complexity and variance sum be more than the During one default complexity threshold, the described first default complexity threshold is defined as to the dynamic of complexity in the monitoring period Threshold value；When the average of complexity presets complexity threshold, and the prison no more than first with variance sum in the monitoring period When the average of complexity is less than the second default complexity threshold with variance sum in the survey period, by the described second default complexity Threshold value is defined as the dynamic threshold of complexity in the monitoring period, wherein, the described second default complexity threshold is less than institute State the first default complexity threshold；When the average of complexity presets complexity threshold, and the prison no more than first with variance sum in the monitoring period When the average of complexity presets complexity threshold with variance sum not less than second in the survey period, by the monitoring period The average and variance sum of complexity are defined as the dynamic threshold of complexity in the monitoring period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310687525.2A CN104706318B (en) | 2013-12-16 | 2013-12-16 | A kind of sleep analysis method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310687525.2A CN104706318B (en) | 2013-12-16 | 2013-12-16 | A kind of sleep analysis method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104706318A CN104706318A (en) | 2015-06-17 |
| CN104706318B true CN104706318B (en) | 2018-02-23 |
Family
ID=53406402
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310687525.2A Active CN104706318B (en) | 2013-12-16 | 2013-12-16 | A kind of sleep analysis method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104706318B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106580332B (en) * | 2015-10-20 | 2019-08-13 | 中国移动通信集团公司 | A kind of method, device and equipment of motion detection |
| CN105559751B (en) * | 2015-12-14 | 2018-06-12 | 安徽华米信息科技有限公司 | Monitor the method, apparatus and wearable device of micro- active state |
| CN105496362B (en) * | 2015-12-15 | 2018-08-31 | 成都麦杰康科技有限公司 | Sleep state monitors system and method |
| CN106913313B (en) * | 2015-12-28 | 2020-07-10 | 中国移动通信集团公司 | Sleep monitoring method and system |
| CN105640508B (en) * | 2016-03-30 | 2018-09-18 | 安徽华米信息科技有限公司 | Real-time sleep monitor method and device, intelligent wearable device |
| CN107638165B (en) * | 2016-07-20 | 2021-01-26 | 平安科技（深圳）有限公司 | Sleep detection method and device |
| CN106175696B (en) * | 2016-09-14 | 2019-02-01 | 广州视源电子科技股份有限公司 | Sleep state monitoring method and system |
| CN106388786A (en) * | 2016-12-14 | 2017-02-15 | 广东技术师范学院 | Processing method acquiring data based on intelligent ring |
| CN106774006B (en) * | 2017-01-22 | 2023-12-29 | 成都图灵创想科技有限责任公司 | Device and method for detecting firing pin of computerized flat knitting machine |
| TWI642409B (en) * | 2017-05-22 | 2018-12-01 | 研鼎智能股份有限公司 | Sleep analysis method |
| CN108095694B (en) * | 2018-01-12 | 2021-01-26 | 安徽华米智能科技有限公司 | State monitoring method and device of wearable equipment and wearable equipment |
| CN108606882B (en) * | 2018-03-23 | 2019-09-10 | 合肥工业大学 | Intelligent wheelchair control system based on myoelectricity and acceleration self adaptive control |
| CN108324251A (en) * | 2018-03-25 | 2018-07-27 | 东莞市华睿电子科技有限公司 | A kind of sleep quality of human body monitoring data processing method |
| CN109062396B (en) * | 2018-07-24 | 2020-01-21 | 百度在线网络技术（北京）有限公司 | Method and device for controlling equipment |
| CN109620158B (en) * | 2018-12-28 | 2021-10-15 | 惠州Tcl移动通信有限公司 | Sleep monitoring method, intelligent terminal and storage device |
| CN110338804A (en) * | 2019-07-02 | 2019-10-18 | 中山大学 | Human body liveness appraisal procedure based on action recognition |
| EP3995077A4 (en) * | 2019-07-05 | 2023-01-11 | The University of Tokyo | Sleep-wakefulness determination device and program |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6527715B2 (en) * | 1998-10-30 | 2003-03-04 | The United States Of America As Represented By The Secretary Of The Army | System and method for predicting human cognitive performance using data from an actigraph |
| US7189204B2 (en) * | 2002-12-04 | 2007-03-13 | Cardiac Pacemakers, Inc. | Sleep detection using an adjustable threshold |
| US7996076B2 (en) * | 2007-04-02 | 2011-08-09 | The Regents Of The University Of Michigan | Automated polysomnographic assessment for rapid eye movement sleep behavior disorder |
| US20100100004A1 (en) * | 2008-10-16 | 2010-04-22 | Koninklijke Nederlandse Akademie Van Wetenschappen | Skin Temperature Measurement in Monitoring and Control of Sleep and Alertness |
| US8280501B2 (en) * | 2010-04-16 | 2012-10-02 | Dyna Dx Corporation | Systems and methods for quantitatively characterizing slow wave activities and states in sleep |
| JP5760669B2 (en) * | 2011-05-12 | 2015-08-12 | 富士通株式会社 | State determination device and state determination method |
-
2013
- 2013-12-16 CN CN201310687525.2A patent/CN104706318B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104706318A (en) | 2015-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104706318B (en) | A kind of sleep analysis method and device | |
| CN109464130B (en) | Sleep assisting method, system and readable storage medium | |
| CN106963369B (en) | Electroencephalogram relaxation degree identification method and device based on neural network model | |
| CN108766532A (en) | Improve teaching method, device and the computer readable storage medium of attention | |
| US20070179396A1 (en) | Method and System for Detecting and Classifying Facial Muscle Movements | |
| CN110706816B (en) | Method and equipment for sleep environment regulation and control based on artificial intelligence | |
| CN106175754B (en) | Waking state detection device in sleep state analysis | |
| CN109671500A (en) | Schizophrenia auxiliary diagnosis classification method based on electroencephalogram time domain data | |
| CN104720746A (en) | Sleeping stage determination method and system | |
| JP2022536552A (en) | Seizure detection system and method based on changes in electroencephalogram (EEG) nonlinearity | |
| CN106648087B (en) | Feature EEG processing method based on consciousness task | |
| CN117064409B (en) | Method, device and terminal for evaluating transcranial direct current intervention stimulation effect in real time | |
| Rahman et al. | Mental stress recognition using K-nearest neighbor (KNN) classifier on EEG signals | |
| CN114391846A (en) | Emotion recognition method and system based on filtering type feature selection | |
| CN114190953A (en) | Training method and system of electroencephalogram signal noise reduction model for electroencephalogram acquisition equipment | |
| CN108143412A (en) | A kind of control method of children's brain electricity mood analysis, apparatus and system | |
| CN110584666A (en) | Sleep latency refining and classifying system | |
| CN107280659A (en) | The processing method and system of a kind of electrocardiosignal | |
| CN111613338B (en) | Method and system for constructing spike-slow complex wave detection model | |
| CN106175698B (en) | Sleep cycle detection device in sleep state analysis | |
| Martens et al. | Separating burst from background spikes in multichannel neuronal recordings using return map analysis | |
| Cynthia et al. | A new system for detecting fatigue and sleepiness using brain connectivity: EEG based estimation of fatigue, vigilance and sleepiness for drivers | |
| CN114010208B (en) | Zero-filling frequency domain convolutional neural network method suitable for SSVEP classification | |
| CN106361327B (en) | Waking state detection method and system in sleep state analysis | |
| CN104545911B (en) | Method and system for detection of human physiological status transition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |