CN104688229A - Method for monitoring sleep respiration based on snore signals - Google Patents
Method for monitoring sleep respiration based on snore signals Download PDFInfo
- Publication number
- CN104688229A CN104688229A CN201510043540.2A CN201510043540A CN104688229A CN 104688229 A CN104688229 A CN 104688229A CN 201510043540 A CN201510043540 A CN 201510043540A CN 104688229 A CN104688229 A CN 104688229A
- Authority
- CN
- China
- Prior art keywords
- signal
- sound
- sleep
- snoring
- breath
- 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
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
- A61B5/0826—Detecting or evaluating apnoea events
Abstract
The invention relates to a method for monitoring sleep respiration based on snore signals. The method is clear in logic and used for judging whether a sleeper has obstructive respiration or not based on the snore signals. The method comprises the following steps: firstly, acquiring continuous respiration signals of a to-be-tested sleeper; secondly, dividing the continuous respiration signals to obtain respiration signal segments, and judging whether each respiration signal segment comprises the snore signal or not respectively; thirdly, judging whether primary obstructive respiration exists or not for all the snore signals respectively; finally, judging whether obstructive respiration exists or not according to an obstructive respiration model for the snore signals with the primary obstructive respiration, and obtaining a sleep respiration situation of the to-be-tested sleeper through judgment. Therefore, the method realizes real-time monitoring for the sleeper, provides scientific and accurate sleep respiration monitoring for the sleeper, thereby paying close attention to sleep health of the sleeper in real time.
Description
Technical field
The present invention relates to a kind of sleep breath monitoring method based on sound of snoring signal.
Background technology
Breathing state is the most important vital signs of human body.To the Real-Time Monitoring of sleep quality breathing state, for some crowd, the patient of such as sleep apnea, or the baby of just birth, and some suffer from respiratory system disease, heart disease patient all very important.
The existing method to the monitoring of sleep quality breathing state mainly utilizes the EEG signals led in hypnotic instrument collection patient sleeps process, electrocardiogram, blood oxygen saturation, mouth and nose air-flow more, lower limb dynamic with position chanP etc. information, the sleep-respiratory state of comprehensive descision patient.But this method cost is high, the information gathering of strange sleep environment and contact can have influence on the sleep quality of patient.
Summary of the invention
For above-mentioned technical problem, technical problem to be solved by this invention is to provide a kind of clear logic, whether there is the sleep breath monitoring method based on sound of snoring signal of obstructive respiration based on sound of snoring signal determining sleep person.
The present invention is in order to solve the problems of the technologies described above by the following technical solutions: the present invention devises a kind of sleep breath monitoring method based on sound of snoring signal, comprises the steps:
Step 001. enrolls the continuous breath signal obtaining single sleep person to be measured continuously according to default sample frequency, enter step 002;
Step 002. is for the continuous breath signal of this sleep person to be measured, using from the 0th moment, each moment of adjacent n/2 duration is successively as each initial time, and by each initial time respectively corresponding duration be that each breath signal of n forms each breath signal section respectively, n > 0, enters step 003;
First step 003. carries out pretreatment respectively for each breath signal section of this sleep person to be measured; Then, whether judge respectively in each breath signal section containing sound of snoring signal, if each breath signal Duan Zhongjun is not containing sound of snoring signal, then judge to there is not obstructive respiration in the admission process of this sleep person to be measured continuous breath signal in step 001, monitoring method terminates; If in each breath signal section, exist at least one breath signal section and contain sound of snoring signal, then obtain the position of sound of snoring signal in this corresponding breath signal sections all, and the duration of sound of snoring signal, enter step 004;
Step 004. is for all sound of snoring signals obtained, according to the position of sound of snoring signal and the duration of sound of snoring signal, the interval judging whether to exist adjacent two sections of sound of snoring signals exceedes to preset to breathe tentative interval lower limit but be less than to preset breathes the tentative interval upper limit, and have continuous sound of snoring signal in the default judgement time before these two sections of sound of snoring signals, preset in the judgement time after these two sections of sound of snoring signals simultaneously and have continuous sound of snoring signal, if do not exist, then judge to there is not obstructive respiration in the admission process of this sleep person to be measured continuous breath signal in step 001, monitoring method terminates; If exist, then enter step 005;
Step 005. is for each group that meets condition described in step 004 adjacent sound of snoring signal, according to judging the method whether containing sound of snoring signal in breath signal section in step 003, judge whether there is breathing signal in each group of adjacent sound of snoring signal respectively, if all containing breathing signal in the adjacent sound of snoring signal of each group, then judge to there is not obstructive respiration in the admission process of this sleep person to be measured continuous breath signal in step 001, monitoring method terminates; If in the adjacent sound of snoring signal of each group, exist at least one group of adjacent sound of snoring signal not containing breathing signal, then judge that there is primary occlusion in this corresponding adjacent sound of snoring signal breathes, and carry out data merging for the repeating part in this corresponding adjacent sound of snoring signal, obtain all adjacent sound of snoring signals that there is primary occlusion and breathe, enter step 006;
Step 006. is for all adjacent sound of snoring signals that there is primary occlusion and breathe, extract the acoustic features in each adjacent sound of snoring signal respectively, and according to the obstructive respiration model established in advance, judge whether there is obstructive respiration in this each adjacent sound of snoring signal respectively, and then whether there is obstructive respiration in the admission process judging this sleep person to be measured continuous breath signal in step 001.
As a preferred technical solution of the present invention: in described step 001, when to process the breath signal of multiple sleep person to be measured simultaneously and sleep person to be measured cooperate with on one's own initiative pre-record breath signal time, first the breath signal of each sleep person to be measured is enrolled respectively, extract acoustic features wherein respectively, and training obtains the Breathiness model of each sleep person to be measured corresponding respectively; Then admission obtains the continuous breath signal of multiple sleep person to be measured continuously, and pass through the Breathiness model of each sleep person to be measured corresponding respectively, obtain the continuous Breathiness of each sleep person to be measured corresponding respectively, finally respectively for the continuous Breathiness of each sleep person to be measured, perform follow-up Overall Steps respectively.
As a preferred technical solution of the present invention: in described step 001, when enrolling the institute of the continuous breath signal of sleep person to be measured in the environment, there is other non-sleep persons or sleep person to be measured does not cooperate with on one's own initiative when pre-recording breath signal, according to the number presetting sleep person to be measured, for the continuous breath signal of admission acquisition continuously, realize each sleep person's breath signal automatic cluster to be measured, obtain the continuous Breathiness of each sleep person to be measured corresponding respectively, finally respectively for the continuous Breathiness of each sleep person to be measured, perform follow-up Overall Steps respectively.
As a preferred technical solution of the present invention: in described step 003, each breath signal section first for this sleep person to be measured carries out pretreatment respectively, removes the part of wherein non-respiratory signal.
As a preferred technical solution of the present invention: in described step 001, obtain the continuous breath signal of single sleep person to be measured according to default sample frequency real-time continuous admission;
In described step 003, after each breath signal section for this sleep person to be measured carries out pretreatment respectively, in real time the n-th moment rise, each moment of adjacent n/2 duration successively, process in real time for using the breath signal section corresponding to this each moment as finish time, namely continue to perform follow-up in steps, realize the real-time process for each breath signal section.
As a preferred technical solution of the present invention: in described step 003, after executing and carrying out pretreated operation respectively for each breath signal section of this sleep person to be measured, respectively for the subsequent operation of each breath signal section, perform in accordance with the following steps respectively, operate for each breath signal section:
Step 003-1. obtains the average energy of this breath signal section as reference energy, dynamically arranges the energy threshold of this breath signal section; The highest zero-crossing rate simultaneously obtaining this breath signal section, as benchmark zero-crossing rate, dynamically arranges the zero-crossing rate thresholding of this breath signal section;
Step 003-2. utilizes a smoothing windows for this breath signal section, and the short-time average energy in calculating acquisition smoothing windows is as current actual energy; Utilize a smoothing windows for this breath signal, the short-time average zero-crossing rate in calculating acquisition smoothing windows is as current actual zero-crossing rate simultaneously;
Step 003-3. carries out framing for this breath signal section, for each the breath signal frame on this breath signal, judge current actual energy and current actual zero-crossing rate whether respectively corresponding energy threshold and the zero-crossing rate thresholding exceeding this breath signal section of breath signal frame respectively, judge that this breath signal frame is as sound of snoring signal, and according to using the initial time of moment corresponding for this sound of snoring signal first frame as this sound of snoring signal, moment corresponding to last frame is as the finish time of this sound of snoring signal, and then obtain the position of this sound of snoring signal in corresponding breath signal section, and the duration of this sound of snoring signal, otherwise judge that this breath signal frame is not sound of snoring signal.
As a preferred technical solution of the present invention: in described step 005, obtain all exist primary occlusion breathe adjacent sound of snoring signal after, by this all exist primary occlusion breathe adjacent sound of snoring signal carry out data storage.
As a preferred technical solution of the present invention: in described step 006, judge whether there is obstructive respiration in this each adjacent sound of snoring signal respectively, if do not exist, then judge that this sleep person to be measured does not exist obstructive respiration in continuous breath signal admission process in step 001; If exist, continue number of times and single duration that statistics obtains obstructive respiration in this adjacent sound of snoring signal, thus and then according to this sleep person to be measured of the statistics number of times of obstructive respiration and single duration in continuous breath signal admission process in step 001, and with this according to presetting sleep quality decision rule, judge to obtain the sleep quality of this sleep person to be measured in step 001 in continuous breath signal admission process.
As a preferred technical solution of the present invention: described sleep breath monitoring method realizes based on client and server, wherein, described step 001 performs in the client to step 005, step 006 performs in the server, is intercomed mutually between client with server by cordless communication network; Wherein, in described step 006, server is after obtaining the sleep quality of this sleep person to be measured in step 001 in continuous breath signal admission process, this sleep quality is fed back to client by cordless communication network by server, the sleep quality that client will receive, the adjacent sound of snoring signal feedback that the primary occlusion in conjunction with corresponding stored is breathed is to user.
As a preferred technical solution of the present invention: described monitoring of respiration method realizes based on client and server, wherein, described step 001 performs in the client to step 005, and step 006 performs in the server, is intercomed mutually between client with server by cordless communication network.
A kind of sleep breath monitoring method based on sound of snoring signal of the present invention adopts above technical scheme compared with prior art, there is following technique effect: the sleep breath monitoring method based on sound of snoring signal of the present invention's design, clear logic, based on sound of snoring signal determining, whether sleep person exists obstructive respiration, Monitor in time can be realized for sleep person, for sleep person provides science, accurately sleep breath monitoring, pay close attention to health when sleep person sleeps in real time; And based on client and server as hardware carrier, realize method for designing of the present invention, effectively control the cost in the method practical application, wherein, not only achieve the layered shaping framework of data analysis, and ensure that the quality of data hierarchy process data communication each other simultaneously, be easy to commercially promote.
Accompanying drawing explanation
Fig. 1 for the present invention design in application messages push agency plant implementation block flow diagram.
Detailed description of the invention
Be described in further detail for the specific embodiment of the present invention below in conjunction with Figure of description.
As shown in Figure 1, a kind of sleep breath monitoring method based on sound of snoring signal of the present invention's design in actual applications, specifically can perform see following steps:
Step 001. is because the audible frequency of people is at 300 ~ 3400Hz, therefore client (smart mobile phone) is adopted to obtain the continuous breath signal of single sleep person to be measured according to default sample frequency 8000Hz real-time continuous admission, simultaneously, in order to save the memory space of data, also in order to save the data traffic of user's upload server, quantize data and be decided to be 8bit, enter step 002.
For the operation in above-mentioned steps 001, there are two kinds of situations in practical application, one is process multiple sleep person to be measured and sleep person to be measured cooperates with breath signal of pre-recording on one's own initiative simultaneously, two is enrolling the institute of the continuous breath signal of sleep person to be measured in the environment, there are other non-sleep persons or sleep person to be measured does not cooperate with breath signal of pre-recording on one's own initiative, for both of these case, step 001 can specifically be done as follows:
When to process the breath signal of multiple sleep person to be measured simultaneously, enroll the breath signal of each sleep person to be measured first respectively, extract acoustic features wherein respectively, and training obtains the Breathiness model of each sleep person to be measured corresponding respectively; Then admission obtains the continuous breath signal of multiple sleep person to be measured continuously, and pass through the Breathiness model of each sleep person to be measured corresponding respectively, obtain the continuous Breathiness of each sleep person to be measured corresponding respectively, finally respectively for the continuous Breathiness of each sleep person to be measured, perform follow-up Overall Steps respectively.
When enrolling the institute of the continuous breath signal of sleep person to be measured in the environment, there is other non-sleep persons or sleep person to be measured does not cooperate with on one's own initiative when pre-recording breath signal, according to the number presetting sleep person to be measured, for the continuous breath signal of admission acquisition continuously, realize each sleep person's breath signal automatic cluster to be measured, obtain the continuous Breathiness of each sleep person to be measured corresponding respectively, finally respectively for the continuous Breathiness of each sleep person to be measured, perform follow-up Overall Steps respectively.
Step 002. is for the continuous breath signal of this sleep person to be measured, using from the 0th moment, each moment of adjacent 2 minutes durations is successively as each initial time, and by each initial time respectively corresponding duration be that each breath signal of 4 minutes forms each breath signal section respectively, enter step 003.
When step 003. is considered and is gathered continuous breath signal, also likely collect sound of the wind, automobile is through sound, hum etc., these acoustical signals and breath signal have obvious difference, can pass through Independent Component Analysis (ICA), each breath signal section first for this sleep person to be measured carries out pretreatment respectively, removes the part of wherein non-respiratory signal; Then, in real time the 4th minute moment rise, each moment of adjacent 2 minutes durations successively, process in real time for using the breath signal section corresponding to this each moment as finish time, namely continue to perform follow-up in steps, realize the real-time process for each breath signal section, wherein, whether judge respectively in each breath signal section containing sound of snoring signal, if each breath signal Duan Zhongjun is not containing sound of snoring signal, then judge to there is not obstructive respiration in the admission process of this sleep person to be measured continuous breath signal in step 001, monitoring method terminates; If in each breath signal section, exist at least one breath signal section and contain sound of snoring signal, then obtain the position of sound of snoring signal in this corresponding breath signal sections all, and the duration of sound of snoring signal, enter step 004; Wherein, operation is carried out for each breath signal section as follows:
Step 003-1. obtains the average energy of this breath signal section as reference energy, dynamically arranges the energy threshold of this breath signal section; The highest zero-crossing rate simultaneously obtaining this breath signal section, as benchmark zero-crossing rate, dynamically arranges the zero-crossing rate thresholding of this breath signal section.
Step 003-2. utilizes a smoothing windows for this breath signal section, and the short-time average energy in calculating acquisition smoothing windows is as current actual energy; Utilize a smoothing windows for this breath signal, the short-time average zero-crossing rate in calculating acquisition smoothing windows is as current actual zero-crossing rate simultaneously.
Step 003-3. carries out framing for this breath signal section, for each the breath signal frame on this breath signal, judge current actual energy and current actual zero-crossing rate whether respectively corresponding energy threshold and the zero-crossing rate thresholding exceeding this breath signal section of breath signal frame respectively, judge that this breath signal frame is as sound of snoring signal, and according to using the initial time of moment corresponding for this sound of snoring signal first frame as this sound of snoring signal, moment corresponding to last frame is as the finish time of this sound of snoring signal, and then obtain the position of this sound of snoring signal in corresponding breath signal section, and the duration of this sound of snoring signal, otherwise judge that this breath signal frame is not sound of snoring signal.
Step 004. is for all sound of snoring signals obtained, according to the position of sound of snoring signal and the duration of sound of snoring signal, the interval judging whether to exist adjacent two sections of sound of snoring signals exceedes to preset to breathe tentative interval lower limit 10 seconds but be less than to preset breathes the tentative interval upper limit 100 seconds, and have continuous sound of snoring signal in the default 30 seconds judgement time before these two sections of sound of snoring signals, preset in 30 seconds judgement time after these two sections of sound of snoring signals simultaneously and have continuous sound of snoring signal, if do not exist, then judge to there is not obstructive respiration in the admission process of this sleep person to be measured continuous breath signal in step 001, monitoring method terminates, if exist, then enter step 005,
Step 005. is for each group that meets condition described in step 004 adjacent sound of snoring signal, according to judging the method whether containing sound of snoring signal in breath signal section in step 003, judge whether there is breathing signal in each group of adjacent sound of snoring signal respectively, if all containing breathing signal in the adjacent sound of snoring signal of each group, then judge to there is not obstructive respiration in the admission process of this sleep person to be measured continuous breath signal in step 001, monitoring method terminates; If in the adjacent sound of snoring signal of each group, exist at least one group of adjacent sound of snoring signal not containing breathing signal, then judge that there is primary occlusion in this corresponding adjacent sound of snoring signal breathes, and carry out data merging for the repeating part in this corresponding adjacent sound of snoring signal, obtain all adjacent sound of snoring signals that there is primary occlusion and breathe, this all adjacent sound of snoring signal that there is primary occlusion breathing is carried out data storage by client, and be transferred to server by cordless communication network, enter step 006.
Step 006. server receives the adjacent sound of snoring signal that there is primary occlusion and breathe, for all adjacent sound of snoring signals that there is primary occlusion and breathe, extract the acoustic features in each adjacent sound of snoring signal respectively, and according to the obstructive respiration model established in advance, judge whether there is obstructive respiration in this each adjacent sound of snoring signal respectively, if do not exist, then judge that this sleep person to be measured does not exist obstructive respiration in continuous breath signal admission process in step 001; If exist, continue number of times and single duration that statistics obtains obstructive respiration in this adjacent sound of snoring signal, thus and then according to this sleep person to be measured of the statistics number of times of obstructive respiration and single duration in continuous breath signal admission process in step 001, and with this according to presetting sleep quality decision rule, judge to obtain the sleep quality of this sleep person to be measured in step 001 in continuous breath signal admission process.
Wherein, server needs the method setting up obstructive respiration model to be: organize blood oxygen and sound of snoring signal based on leading hypnotic instrument collection more more, and the index of reference blood oxygen, extract sound of snoring signal when blood oxygen reduces, utilize the feature of this sound of snoring signal, set up obstructive respiration model, idiographic flow is: sound of snoring data when first utilizing blood oxygen index acquisition blood oxygen to reduce, then the acoustic features of these sound of snoring data is extracted, comprise: energy, frequency, zero-crossing rate, mfcc, lpcc etc., recycling model, comprise: gmm, svdd etc. carry out modeling to the acoustic features extracted, to realize whether there is obstructive respiration in intelligent decision sound of snoring signal.The initial stage of obstructive respiration model mainly leans on the oximetry data and sound of snoring data that gather from hospital, set up primary occlusion respiratory model model, after system cloud gray model gets up, the data uploaded by utilizing user are constantly revised and sophisticated model, improve the accuracy of obstructive respiration Model Identification, concrete grammar is: by the artificial method labelling obstructive respiration event monitored, compare with the obstructive respiration event of system identification and associate, continuous amendment sound of snoring model parameter, improves obstructive respiration Model Identification accuracy.
The sleep-respiratory state of sleep person, comprises quiet sleep time, snoring time, the moment of obstructive respiration event, total degree, single duration statistics; Preset sleep quality decision rule as follows:
Sleep quality is excellent: do not have obstructive respiration;
Sleep quality is good: obstructive respiration number of times is less than or equal to 15 times, and each duration is less than or equal to 20 seconds;
In sleep quality: obstructive respiration number of times is greater than 15 times and is less than or equal to 30 times, and each duration is less than or equal to 40 seconds;
Poor sleeping quality: obstructive respiration number of times is greater than 30 times, or single duration is greater than 40 seconds.
As a preferred technical solution of the present invention: in described step 006, judge whether there is obstructive respiration in this each adjacent sound of snoring signal respectively, if do not exist, then judge that this sleep person to be measured does not exist obstructive respiration in continuous breath signal admission process in step 001; If exist, continue number of times and single duration that statistics obtains obstructive respiration in this adjacent sound of snoring signal, thus and then according to this sleep person to be measured of the statistics number of times of obstructive respiration and single duration in continuous breath signal admission process in step 001, and with this according to presetting sleep quality decision rule, judge to obtain the sleep quality of this sleep person to be measured in step 001 in continuous breath signal admission process.
As a preferred technical solution of the present invention: described sleep breath monitoring method realizes based on client and server, wherein, described step 001 performs in the client to step 005, step 006 performs in the server, is intercomed mutually between client with server by cordless communication network; Wherein, in described step 006, server is after obtaining the sleep quality of this sleep person to be measured in step 001 in continuous breath signal admission process, this sleep quality is fed back to client by cordless communication network by server, the sleep quality that client will receive, the adjacent sound of snoring signal feedback that the primary occlusion in conjunction with corresponding stored is breathed is to user.
Wherein, the content of sleep quality comprises: comprise sound of snoring number of times, the sound of snoring and the length of one's sleep ratio, block number of times, single blocks duration, accumulate and block duration, block and breathe period of right time, Sleep quality scores, suggestion etc., and in practical application, for client, design is adopted as the smart mobile phone installing corresponding software.
In practical application of the present invention, if the continuous breath signal that step 001 realizes realizing the whole night for sleep person to be measured is enrolled, then finally both can judge to obtain this sleep person to be measured sleep quality the whole night.
Can also for Client Design hypnosis function and arousal function in practical application of the present invention, hypnosis function is the soft music of fixed time airplay, voice etc. mainly, help user to fall asleep, and automatically stop sound playing after falling asleep; Arousal function is mainly waken up by the patient of the tinkle of bells by sleep state or obstructive respiration state as required, changes its current state.
After detecting that sleep person there occurs obstructive respiration event several times, when obstructive respiration event again being detected, the mobile phone of cell phone automatic dialing guardian can be set, guardian can the sound of snoring of real-time listening sleep person, when confirming that sleep person breathes for Severe blockage (number of times more or single duration more than 40 seconds), select to send instruction to sleep person's mobile phone, wake sleep person up; Wherein, guardian can be the relatives of sleep person, also can be the back-stage management personnel of design system of the present invention, care provider, or other smart machine (this smart machine has the ability of ONLINE RECOGNITION obstructive respiration state).
The sleep breath monitoring method based on sound of snoring signal of the present invention's design, clear logic, based on sound of snoring signal determining, whether sleep person exists obstructive respiration, Monitor in time can be realized for sleep person, for sleep person provides science, accurately sleep breath monitoring, pay close attention to health when sleep person sleeps in real time; And based on client and server as hardware carrier, realize method for designing of the present invention, effectively control the cost in the method practical application, wherein, not only achieve the layered shaping framework of data analysis, and ensure that the quality of data hierarchy process data communication each other simultaneously, be easy to commercially promote.
By reference to the accompanying drawings embodiments of the present invention are explained in detail above, but the present invention is not limited to above-mentioned embodiment, in the ken that those of ordinary skill in the art possess, can also makes a variety of changes under the prerequisite not departing from present inventive concept.
Claims (10)
1., based on a sleep breath monitoring method for sound of snoring signal, it is characterized in that, comprise the steps:
Step 001. enrolls the continuous breath signal obtaining single sleep person to be measured continuously according to default sample frequency, enter step 002;
Step 002. is for the continuous breath signal of this sleep person to be measured, using from the 0th moment, each moment of adjacent n/2 duration is successively as each initial time, and by each initial time respectively corresponding duration be that each breath signal of n forms each breath signal section respectively, n > 0, enters step 003;
First step 003. carries out pretreatment respectively for each breath signal section of this sleep person to be measured; Then, whether judge respectively in each breath signal section containing sound of snoring signal, if each breath signal Duan Zhongjun is not containing sound of snoring signal, then judge to there is not obstructive respiration in the admission process of this sleep person to be measured continuous breath signal in step 001, monitoring method terminates; If in each breath signal section, exist at least one breath signal section and contain sound of snoring signal, then obtain the position of sound of snoring signal in this corresponding breath signal sections all, and the duration of sound of snoring signal, enter step 004;
Step 004. is for all sound of snoring signals obtained, according to the position of sound of snoring signal and the duration of sound of snoring signal, the interval judging whether to exist adjacent two sections of sound of snoring signals exceedes to preset to breathe tentative interval lower limit but be less than to preset breathes the tentative interval upper limit, and have continuous sound of snoring signal in the default judgement time before these two sections of sound of snoring signals, preset in the judgement time after these two sections of sound of snoring signals simultaneously and have continuous sound of snoring signal, if do not exist, then judge to there is not obstructive respiration in the admission process of this sleep person to be measured continuous breath signal in step 001, monitoring method terminates; If exist, then enter step 005;
Step 005. is for each group that meets condition described in step 004 adjacent sound of snoring signal, according to judging the method whether containing sound of snoring signal in breath signal section in step 003, judge whether there is breathing signal in each group of adjacent sound of snoring signal respectively, if all containing breathing signal in the adjacent sound of snoring signal of each group, then judge to there is not obstructive respiration in the admission process of this sleep person to be measured continuous breath signal in step 001, monitoring method terminates; If in the adjacent sound of snoring signal of each group, exist at least one group of adjacent sound of snoring signal not containing breathing signal, then judge that there is primary occlusion in this corresponding adjacent sound of snoring signal breathes, and carry out data merging for the repeating part in this corresponding adjacent sound of snoring signal, obtain all adjacent sound of snoring signals that there is primary occlusion and breathe, enter step 006;
Step 006. is for all adjacent sound of snoring signals that there is primary occlusion and breathe, extract the acoustic features in each adjacent sound of snoring signal respectively, and according to the obstructive respiration model established in advance, judge whether there is obstructive respiration in this each adjacent sound of snoring signal respectively, and then whether there is obstructive respiration in the admission process judging this sleep person to be measured continuous breath signal in step 001.
2. a kind of sleep breath monitoring method based on sound of snoring signal according to claim 1, it is characterized in that: in described step 001, when to process the breath signal of multiple sleep person to be measured simultaneously and sleep person to be measured cooperate with on one's own initiative pre-record breath signal time, first the breath signal of each sleep person to be measured is enrolled respectively, extract acoustic features wherein respectively, and training obtains the Breathiness model of each sleep person to be measured corresponding respectively; Then admission obtains the continuous breath signal of multiple sleep person to be measured continuously, and pass through the Breathiness model of each sleep person to be measured corresponding respectively, obtain the continuous Breathiness of each sleep person to be measured corresponding respectively, finally respectively for the continuous Breathiness of each sleep person to be measured, perform follow-up Overall Steps respectively.
3. a kind of sleep breath monitoring method based on sound of snoring signal according to claim 2, it is characterized in that: in described step 001, when enrolling the institute of the continuous breath signal of sleep person to be measured in the environment, there is other non-sleep persons or sleep person to be measured does not cooperate with on one's own initiative when pre-recording breath signal, according to the number presetting sleep person to be measured, for the continuous breath signal of admission acquisition continuously, realize each sleep person's breath signal automatic cluster to be measured, obtain the continuous Breathiness of each sleep person to be measured corresponding respectively, finally respectively for the continuous Breathiness of each sleep person to be measured, perform follow-up Overall Steps respectively.
4. a kind of sleep breath monitoring method based on sound of snoring signal according to claim 1, it is characterized in that: in described step 003, each breath signal section first for this sleep person to be measured carries out pretreatment respectively, removes the part of wherein non-respiratory signal.
5. a kind of sleep breath monitoring method based on sound of snoring signal according to claim 1, is characterized in that: in described step 001, obtains the continuous breath signal of single sleep person to be measured according to default sample frequency real-time continuous admission;
In described step 003, after each breath signal section for this sleep person to be measured carries out pretreatment respectively, in real time the n-th moment rise, each moment of adjacent n/2 duration successively, process in real time for using the breath signal section corresponding to this each moment as finish time, namely continue to perform follow-up in steps, realize the real-time process for each breath signal section.
6. a kind of sleep breath monitoring method based on sound of snoring signal according to claim 1, it is characterized in that: in described step 003, after executing and carrying out pretreated operation respectively for each breath signal section of this sleep person to be measured, respectively for the subsequent operation of each breath signal section, perform in accordance with the following steps respectively, operate for each breath signal section: step 003-1. obtains the average energy of this breath signal section as reference energy, dynamically arranges the energy threshold of this breath signal section; The highest zero-crossing rate simultaneously obtaining this breath signal section, as benchmark zero-crossing rate, dynamically arranges the zero-crossing rate thresholding of this breath signal section;
Step 003-2. utilizes a smoothing windows for this breath signal section, and the short-time average energy in calculating acquisition smoothing windows is as current actual energy; Utilize a smoothing windows for this breath signal, the short-time average zero-crossing rate in calculating acquisition smoothing windows is as current actual zero-crossing rate simultaneously;
Step 003-3. carries out framing for this breath signal section, for each the breath signal frame on this breath signal, judge current actual energy and current actual zero-crossing rate whether respectively corresponding energy threshold and the zero-crossing rate thresholding exceeding this breath signal section of breath signal frame respectively, judge that this breath signal frame is as sound of snoring signal, and according to using the initial time of moment corresponding for this sound of snoring signal first frame as this sound of snoring signal, moment corresponding to last frame is as the finish time of this sound of snoring signal, and then obtain the position of this sound of snoring signal in corresponding breath signal section, and the duration of this sound of snoring signal, otherwise judge that this breath signal frame is not sound of snoring signal.
7. a kind of sleep breath monitoring method based on sound of snoring signal according to claim 1, it is characterized in that: in described step 005, obtain all exist primary occlusion breathe adjacent sound of snoring signal after, by this all exist primary occlusion breathe adjacent sound of snoring signal carry out data storage.
8. a kind of sleep breath monitoring method based on sound of snoring signal according to claim 7, it is characterized in that: in described step 006, judge whether there is obstructive respiration in this each adjacent sound of snoring signal respectively, if do not exist, then judge that this sleep person to be measured does not exist obstructive respiration in continuous breath signal admission process in step 001; If exist, continue number of times and single duration that statistics obtains obstructive respiration in this adjacent sound of snoring signal, thus and then according to this sleep person to be measured of the statistics number of times of obstructive respiration and single duration in continuous breath signal admission process in step 001, and with this according to presetting sleep quality decision rule, judge to obtain the sleep quality of this sleep person to be measured in step 001 in continuous breath signal admission process.
9. a kind of sleep breath monitoring method based on sound of snoring signal according to claim 8, it is characterized in that: described sleep breath monitoring method realizes based on client and server, wherein, described step 001 performs in the client to step 005, step 006 performs in the server, is intercomed mutually between client with server by cordless communication network; Wherein, in described step 006, server is after obtaining the sleep quality of this sleep person to be measured in step 001 in continuous breath signal admission process, this sleep quality is fed back to client by cordless communication network by server, the sleep quality that client will receive, the adjacent sound of snoring signal feedback that the primary occlusion in conjunction with corresponding stored is breathed is to user.
10. a kind of sleep breath monitoring method based on sound of snoring signal according to claim 1, it is characterized in that: described monitoring of respiration method realizes based on client and server, wherein, described step 001 performs in the client to step 005, step 006 performs in the server, is intercomed mutually between client with server by cordless communication network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510043540.2A CN104688229A (en) | 2015-01-28 | 2015-01-28 | Method for monitoring sleep respiration based on snore signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510043540.2A CN104688229A (en) | 2015-01-28 | 2015-01-28 | Method for monitoring sleep respiration based on snore signals |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104688229A true CN104688229A (en) | 2015-06-10 |
Family
ID=53336098
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510043540.2A Pending CN104688229A (en) | 2015-01-28 | 2015-01-28 | Method for monitoring sleep respiration based on snore signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104688229A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105286867A (en) * | 2014-06-24 | 2016-02-03 | 中兴通讯股份有限公司 | Method and device for collecting voice information |
| CN105963065A (en) * | 2016-07-15 | 2016-09-28 | 北京博智卓康科技有限公司 | Bed and method for restraining snores |
| CN107411704A (en) * | 2017-03-13 | 2017-12-01 | 嘉兴礼海电气科技有限公司 | A kind of sleep quality automatic judging method judged based on Pittsburgh |
| CN107811610A (en) * | 2017-09-27 | 2018-03-20 | 深圳和而泰智能控制股份有限公司 | A kind of respiratory rate detection method, device, electronic equipment and storage medium |
| CN107898432A (en) * | 2017-10-17 | 2018-04-13 | 深圳和而泰智能控制股份有限公司 | Sound of snoring detection method, device, equipment and medium |
| CN107981844A (en) * | 2017-12-08 | 2018-05-04 | 绵眠（上海）智能科技有限公司 | A kind of sound of snoring recognition methods and system based on piezoelectric membrane |
| CN108369813A (en) * | 2017-07-31 | 2018-08-03 | 深圳和而泰智能家居科技有限公司 | Specific sound recognition methods, equipment and storage medium |
| CN108814796A (en) * | 2018-05-02 | 2018-11-16 | 王美金 | A kind of air-cushion type artificial intelligence snore disease monitoring down feather pillow and application method |
| CN109044279A (en) * | 2018-08-20 | 2018-12-21 | 深圳和而泰数据资源与云技术有限公司 | A kind of sound of snoring detection method and relevant device |
| WO2019037456A1 (en) * | 2017-08-24 | 2019-02-28 | 深圳创达云睿智能科技有限公司 | Snoring detection method and system, and terminal device |
| CN110151151A (en) * | 2019-05-10 | 2019-08-23 | 深圳市奋达科技股份有限公司 | It is a kind of to monitor method, system and the bracelet snored |
| CN110338797A (en) * | 2019-08-12 | 2019-10-18 | 苏州小蓝医疗科技有限公司 | A kind of intermediate frequency snore stopper data processing method based on the sound of snoring and blood oxygen |
| CN110349587A (en) * | 2019-07-12 | 2019-10-18 | 南京航空航天大学 | Target individual is snored differentiating method under a kind of two people's scenes |
| CN111489824A (en) * | 2020-04-09 | 2020-08-04 | 杭州电子科技大学 | OSAHS prediction system based on Internet of things |
| CN111904424A (en) * | 2020-08-06 | 2020-11-10 | 苏州国科医工科技发展(集团)有限公司 | Sleep monitoring and regulating system based on phased array microphone |
| CN112107293A (en) * | 2020-08-11 | 2020-12-22 | 山东大学 | Snore monitoring system based on breathing parameters during sleep |
| TWI728839B (en) * | 2020-06-12 | 2021-05-21 | 臺北醫學大學 | Household sleeping monitor, household sleeping monitoring system and household sleep monitoring and method for sleep breathing event classification |
| CN113314143A (en) * | 2021-06-07 | 2021-08-27 | 南京优博一创智能科技有限公司 | Apnea judgment method and device and electronic equipment |
-
2015
- 2015-01-28 CN CN201510043540.2A patent/CN104688229A/en active Pending
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105286867A (en) * | 2014-06-24 | 2016-02-03 | 中兴通讯股份有限公司 | Method and device for collecting voice information |
| CN105963065A (en) * | 2016-07-15 | 2016-09-28 | 北京博智卓康科技有限公司 | Bed and method for restraining snores |
| CN105963065B (en) * | 2016-07-15 | 2018-08-24 | 北京博智卓康科技有限公司 | A kind of bed of snore inhibiting |
| CN107411704A (en) * | 2017-03-13 | 2017-12-01 | 嘉兴礼海电气科技有限公司 | A kind of sleep quality automatic judging method judged based on Pittsburgh |
| CN108369813A (en) * | 2017-07-31 | 2018-08-03 | 深圳和而泰智能家居科技有限公司 | Specific sound recognition methods, equipment and storage medium |
| WO2019037456A1 (en) * | 2017-08-24 | 2019-02-28 | 深圳创达云睿智能科技有限公司 | Snoring detection method and system, and terminal device |
| CN107811610A (en) * | 2017-09-27 | 2018-03-20 | 深圳和而泰智能控制股份有限公司 | A kind of respiratory rate detection method, device, electronic equipment and storage medium |
| CN107811610B (en) * | 2017-09-27 | 2020-12-15 | 深圳和而泰智能控制股份有限公司 | Respiration rate detection method and device, electronic equipment and storage medium |
| CN107898432B (en) * | 2017-10-17 | 2021-02-26 | 深圳和而泰智能控制股份有限公司 | Snore detection method, device, equipment and medium |
| CN107898432A (en) * | 2017-10-17 | 2018-04-13 | 深圳和而泰智能控制股份有限公司 | Sound of snoring detection method, device, equipment and medium |
| CN107981844A (en) * | 2017-12-08 | 2018-05-04 | 绵眠（上海）智能科技有限公司 | A kind of sound of snoring recognition methods and system based on piezoelectric membrane |
| CN108814796A (en) * | 2018-05-02 | 2018-11-16 | 王美金 | A kind of air-cushion type artificial intelligence snore disease monitoring down feather pillow and application method |
| CN109044279A (en) * | 2018-08-20 | 2018-12-21 | 深圳和而泰数据资源与云技术有限公司 | A kind of sound of snoring detection method and relevant device |
| CN110151151A (en) * | 2019-05-10 | 2019-08-23 | 深圳市奋达科技股份有限公司 | It is a kind of to monitor method, system and the bracelet snored |
| CN110349587B (en) * | 2019-07-12 | 2022-03-25 | 南京航空航天大学 | Target individual calling distinguishing method under two-person scene |
| CN110349587A (en) * | 2019-07-12 | 2019-10-18 | 南京航空航天大学 | Target individual is snored differentiating method under a kind of two people's scenes |
| CN110338797A (en) * | 2019-08-12 | 2019-10-18 | 苏州小蓝医疗科技有限公司 | A kind of intermediate frequency snore stopper data processing method based on the sound of snoring and blood oxygen |
| CN111489824A (en) * | 2020-04-09 | 2020-08-04 | 杭州电子科技大学 | OSAHS prediction system based on Internet of things |
| TWI728839B (en) * | 2020-06-12 | 2021-05-21 | 臺北醫學大學 | Household sleeping monitor, household sleeping monitoring system and household sleep monitoring and method for sleep breathing event classification |
| CN111904424A (en) * | 2020-08-06 | 2020-11-10 | 苏州国科医工科技发展(集团)有限公司 | Sleep monitoring and regulating system based on phased array microphone |
| CN111904424B (en) * | 2020-08-06 | 2021-08-24 | 苏州国科医工科技发展(集团)有限公司 | Sleep monitoring and regulating system based on phased array microphone |
| CN112107293A (en) * | 2020-08-11 | 2020-12-22 | 山东大学 | Snore monitoring system based on breathing parameters during sleep |
| CN112107293B (en) * | 2020-08-11 | 2021-09-21 | 山东大学 | Snore monitoring system based on breathing parameters during sleep |
| CN113314143A (en) * | 2021-06-07 | 2021-08-27 | 南京优博一创智能科技有限公司 | Apnea judgment method and device and electronic equipment |
| CN113314143B (en) * | 2021-06-07 | 2024-01-30 | 南京优博一创智能科技有限公司 | Method and device for judging apnea and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104688229A (en) | Method for monitoring sleep respiration based on snore signals | |
| CN109431470B (en) | Sleep respiration monitoring method and device | |
| CN105748043B (en) | Sleep quality monitoring method and device, wearable device | |
| CN109920544B (en) | Real-time self-adaptive intelligent building system based on somatosensory information | |
| Gu et al. | Sleep hunter: Towards fine grained sleep stage tracking with smartphones | |
| CN100478998C (en) | Wake-up alarm signal generating apparatus and method | |
| CN107591162B (en) | Cry recognition method based on pattern matching and intelligent nursing system | |
| CN104545844B (en) | Multi-parameter sleep monitoring and intelligent diagnosis system based on 4G mobile communication technology and application method of multi-parameter sleep monitoring and intelligent diagnosis system | |
| WO2017193497A1 (en) | Fusion model-based intellectualized health management server and system, and control method therefor | |
| US11298101B2 (en) | Device embedded in, or attached to, a pillow configured for in-bed monitoring of respiration | |
| CN111166297B (en) | Method and device for evaluating sleep quality based on user sleep audio | |
| CN105899129A (en) | Fatigue monitoring and management system | |
| CN102138796A (en) | Sleep monitoring obstructive locator based on snore analysis | |
| CN110367934B (en) | Health monitoring method and system based on non-voice body sounds | |
| CN202904817U (en) | Soundless sleep apnea prompting system | |
| CN202069591U (en) | Snore analysis-based positioning instrument for monitoring blocking in sleep | |
| CN109350014A (en) | A kind of sound of snoring recognition methods and system | |
| CN103610445A (en) | Physiological signal detection processing method and system | |
| CN108510075A (en) | The monitoring inference system of sleep quality and environmental variance correlation | |
| CN110575139A (en) | Sleep monitoring method and equipment | |
| Romero et al. | Deep learning features for robust detection of acoustic events in sleep-disordered breathing | |
| US20180085050A1 (en) | Systems and methods for individualized sleep optimization | |
| WO2021208656A1 (en) | Sleep risk prediction method and apparatus, and terminal device | |
| CN105962895A (en) | User state reminding method and system | |
| CN109730658A (en) | A kind of artificial intelligence mattress system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into 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: 20150610 |