CN109745002A - A kind of portable sleep monitoring equipment - Google Patents
A kind of portable sleep monitoring equipment Download PDFInfo
- Publication number
- CN109745002A CN109745002A CN201811614919.4A CN201811614919A CN109745002A CN 109745002 A CN109745002 A CN 109745002A CN 201811614919 A CN201811614919 A CN 201811614919A CN 109745002 A CN109745002 A CN 109745002A
- Authority
- CN
- China
- Prior art keywords
- sign
- monitoring equipment
- portable sleep
- sleep monitoring
- sleep
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
For sleep monitor, hospital generally will use the multiple parameters such as brain electricity, eye electricity, lower jaw myoelectricity, mouth and nose air-flow, Breathing movement, electrocardio, blood oxygen, the sound of snoring, main drive and the position of Polysomnography system detection patient, and then analyze the sleep depth period and quality of patient.The present invention proposes a kind of portable sleep monitoring equipment, including Power Management Unit, communication unit and acceleration transducer, it further include sign monitoring modular, sign monitoring modular is equipped with red-light LED, infrared light LED and optical sensor, the shell of the monitoring device is opaque material, and double LED and the corresponding region of optical sensor use light-transmitting materials on shell.The communication unit further includes wireless communication unit, information relay unit and server, passes through network communication between information relay unit and server.Wherein the sign of sign monitoring module monitors includes pulse wave, heart rate value, blood oxygen levels and blood vessel microcirculation parameter.
Description
Technical field
The invention belongs to monitoring device field, especially a kind of portable sleep monitoring equipment.
Background technique
For sleep monitor, hospital generally will use Polysomnography system, and Polysomnography system is generally by leading
Machine, display, amplifier, collecting cassette, EEG/ECG/EOG/EMG sensor, chest and abdomen motion sensor, temperature-sensitive gas flow transducer,
Blood oxygen transducer, sound of snoring sensor, body position sensor, signal cable, isolated power supply composition.Use Polysomnography system
When, need through brain electricity, eye electricity, the myoelectricity etc. when monitoring a whole night sleep, and then evaluation patient sleeps' mass, when sleeping
Between, Sleep efficiency and monitoring by stages, exclude sleep cognition cacodoxy, patient made correctly to recognize the sleeping problems of oneself, it is right
The sleep quality of oneself has an objective appraisal and understanding.
But for the patient, need physically to install various sensors, operation is very complicated and complicated, needs simultaneously
Continuous monitoring for a long time is carried out under hospital's special scenes, this can generate further shadow to the sleeping problems of patient itself
It rings.Therefore some simple and convenient and energy visual record sleep quality information equipment is needed, these equipment pass through wearable side
Formula allows user directly to use in the family.
Summary of the invention
A kind of portable sleep monitoring equipment is proposed based on this present invention, the technical solution adopted is as follows:
A kind of portable sleep monitoring equipment, including Power Management Unit, communication unit and acceleration transducer further include
Sign monitoring modular, sign monitoring modular are equipped with the double LED of feux rouges infrared light and optical sensor, and the shell of the monitoring device is not
Light-transmitting materials, double LED and the corresponding region of optical sensor use light-transmitting materials on shell.
Further, the model MPU9250 of acceleration transducer.
Further, the model JFH111 of sign monitoring modular.
Further, the size of sign monitoring modular is 11.8mm*5mm.
Further, the Power Management Unit includes battery, Charge Management unit and linear voltage regulator, Charge Management list
The component of the VIN pin connection for the model SGM4056, SGM4056 that member uses includes concatenated resistance R1 and capacitor C1, string
Resistance R2, the Schottky diode D1 of connection and the cathode of Schottky diode D2, D1 and D2 are opposite, are also connected between R2 and D1
Resistance R3.
Further, the model SGM2036 of linear voltage regulator.
Further, the enable end of linear voltage regulator is connected with the leading point between D1 and D2.
Further, the communication unit includes bluetooth communication and user terminal.
Further, the model DA14580 of bluetooth communication.
Further, the communication unit further includes wireless communication unit, information relay unit and server, information relaying
Pass through network communication between unit and server.
Further, the pin of the sign monitoring modular includes VCC_HO, is used for digital power system, VCC_3V, is used for LED
Power supply, reset pin, power ground, hanging pin and serial communication pin.
Further, the sign of sign monitoring module monitors includes pulse wave, heart rate value, blood oxygen levels and blood vessel microcirculation
Parameter.
Further, sign monitoring modular is every returns one acquisition data through 64 sampled points.
Compared with prior art, the beneficial effects of the present invention are: using monitoring device proposed by the present invention it is small in size, can
To be applied in different wearable devices, pulse wave technology and sleep monitor can be combined using this equipment, favorably
In the acceleration promotion and application of pulse wave technology.
Detailed description of the invention
Fig. 1 is monitoring device external structure schematic diagram；
Fig. 2 is linear voltage regulator chip pin figure；
Fig. 3 is JFH111 pin schematic diagram；
Fig. 4 is Power Management Unit structural schematic diagram；
Fig. 5 is bluetooth module pin schematic diagram；
Fig. 6 is the deep learning network structure for calculating heart rate and blood oxygen；
Fig. 7 is the deep learning model structure for calculating pulse wave and microcirculation parameter；
Fig. 8 is respiratory characteristic curve synoptic diagram；
Fig. 9 is heart rate indicatrix schematic diagram；
Figure 10 is blood oxygen change curve schematic diagram.
Description of symbols:
Shell -1, optical sensor transparent area -2, LED transparent area -3.
Specific embodiment
In the present embodiment, a wearable information collecting device for sleep monitor is devised, the equipment volume is small and exquisite,
It is easy to operate, information collection end need to be only pasted on to the forehead of user, pass through data by connection mobile phone terminal app software either bed
Interactive terminal can record the sleep info of user, and the heart rate of real-time display active user, blood oxygen, microcirculation, body
The information such as position.After user wake up, sleep monitor equipment can automatically analyze out the sleep info of user.
As shown in Figures 1 to 5, in the present embodiment, sleep monitor equipment includes fulgurite reason cell S GM4056, and acceleration passes
The MPU9250 of sensor, linear voltage regulator SGM2036, bluetooth communication DA14580 and sign monitoring modular JFH111 accelerate
Degree sensor, sign detection module are connected with bluetooth communication, and Power Management Unit is acceleration transducer, linear voltage stabilization list
Member, bluetooth-communication unit and the power supply of sign monitoring modular, linear voltage regulator monitor mould for burning voltage administrative unit and sign
The voltage of block.The shell 1 of monitoring device uses the opaque material of black, optical sensor transparent area 2 and LED light transmission on shell 1
Area 3 uses and feux rouges and the good material of infrared light light transmission is made.
In the present embodiment, the signal acquisition part point of sign monitoring modular includes the double LED of feux rouges infrared light and optical sensor, is adopted
The signal of collection directly exports pulse wave, heart rate value, blood oxygen levels and blood vessel microcirculation parameter after the processing of sign monitoring modular.Sign
Monitoring modular is every through 64 sampled points (time-consuming 1.28s) transmission primaries data packet, and data packet draws sign song for user terminal
Line.
In the present embodiment, sign monitoring modular obtains heart rate value and blood oxygen levels by deep learning network, is directed to
Deep learning network model uses multi-task learning mode, i.e., the two predict task sharings convolutional layer, and use different
Full articulamentum carries out last prediction, as shown in fig. 6, model structure is five layers of convolutional neural networks layer being sequentially connected, convolution
Containing gate linear unit between neural net layer, gate linear unit formula is as follows:
H=A*sigmoid (B)
Wherein H is output, and A and B are the half port number of input.The form of multiplication is point-by-point is multiplied.
In the present embodiment, heart rate loss function that when training pattern uses are as follows:
Wherein oiFor i-th of the actual output of two classifiers, yiFor the desired output of two classifiers.
The oxygen content of blood loss function used when training pattern are as follows:
L2=(Y '-Y)2
Wherein Y ' is prediction blood oxygen saturation, and Y is practical blood oxygen saturation.
The regularization loss function used when training pattern are as follows:
L3=(θ)2
Wherein θ is model parameter.
The loss function of entire depth learning network model when training pattern are as follows:
L=α 1*L1+ α 2*L2+ α 3*L3
Wherein 1 α, α 2, α 3 are the weight of three loss functions.
When primary data pass through convolutional neural networks layer after, according to the high dimensional feature that convolutional neural networks extract come
To heart rate and blood oxygen saturation.Task is predicted for blood oxygen saturation, increases by two full articulamentums after shared convolutional layer, finally
One full articulamentum output numerical value is blood oxygen saturation.And the prediction of heart rate will use ordinal regression (ordinal
Regression method), after shared convolutional layer, we can add one layer of full articulamentum, then add after full articulamentum
Upper 256 two classifiers, each classifier output 0 or 1, if HR values are greater than classifier serial number, classifier output is
1, it is otherwise 0.
When measurement, by optical sensor, collected signal is input in network model sign monitoring modular in real time, and network is defeated
Blood oxygen saturation out is to predict blood oxygen saturation, and the output of 256 classifiers of heart rate predicted portions is added together i.e.
For the heart rate of prediction.
In the present embodiment, microcirculation value is sought using deep learning model, when calculating microcirculation value, it is necessary first to be fitted
Pulse wave signal is fitted pulse wave signal using one-dimensional gaussian profile in the present embodiment, it is assumed that stochastic variable X obeys position
Parameter is μ, and scale parameter is the distribution of σ, then probability density function are as follows:
The top of Gaussian Profile is at mean value, and both sides are symmetrical about mean value, in addition maximum value at mean valueFor
Fitting pulse wave signal, needs two Gaussian Profiles, if reconstruction of function is G.Then
G (x)=max (a1*f1(x), a2*f2(x))
So needing to estimate a altogether1, μ1, σ1, a2, μ2, σ2This six parameters.
As shown in fig. 7, the depth network architecture that reconfigurable measurement signal uses includes 4 layers of convolution depth being sequentially connected
Network layer.In order to train depth network, period divisions are carried out to raw measured signal and obtain the measuring signal of signal period, it will be single
The input when measuring signal in a period is as training pattern.
The error loss function used when training depth network model are as follows:
Wherein N is the length of a cycle measuring signal P, and x is the data point of a cycle measuring signal, and θ (P) is that will count
The parameter in double gauss function exported after strong point input depth network model, including a1, a2, μ1, μ2, σ1, σ2。
The regularization loss function used when training pattern are as follows:
L2=(θ)2
Wherein θ is depth network model parameter.
The loss function of entire depth network model are as follows:
L=α 1*L1+ α 2*L2
Wherein 1 α, α 2 are the weight of two loss functions.Two loss function weights are respectively set as 1,0.00005.
Network parameter is optimized using Adam algorithm when training, the deconditioning when loss function convergence.Due to every
The measuring signal indefinite length in a period, so the sample size (batch) of each input network is set as 1.
Measuring signal inputs network model when use, then can be obtained by parameter a required for Gaussian reconstruction1, μ1, σ1,
a2, μ2, σ2, utilize two peak value a of measuring signal1, a2And the distance between two peak values | μ1-μ2| available microcirculation
Value.
It is as follows wherein to calculate the formula that microcirculation value uses:
K=a (2) * k2+a(1)*k+a(0)
Wherein: K is microcirculation, and vector a passes through the multinomial coefficient that mass data is fitted, independent variable k=ampPaddy*
10/AC, AC are a monocyclic exchange value: AC=ABS (ampPeak-ampPaddy), ampPeakFor monocycle measuring signal peak-peak
The amplitude at place, ampPaddyFor the amplitude at minimum trough.
In the present embodiment, posture and fine motion value of the user in sleep procedure are judged by acceleration transducer, thus
Assess the respiratory rhythm and respiratory intensity when user's sleep.
In the present embodiment, the method for user's sleep quality is judged using data are as follows:
1. when not causing the exception and lowered blood oxygen of heart rate, being indicated when respiratory rhythm and respiratory intensity change
It is low that adnormal respiration degree occurs.
2. when respiratory rhythm and respiratory intensity change, while causing increased heart rate, but do not cause the wave of blood oxygen
It is dynamic, illustrate that sleep apnea occurs for sleep, but intensity is little.
3. when respiratory rhythm and respiratory intensity change, when increased heart rate, lowered blood oxygen, explanation, which is had occurred, once shadow
Loud sleep apnea.
4. according to apnea duration, the intensity and blood oxygen of the speed for causing changes in heart rate and variation occurs
Reduced degree come judge occur apnea seriousness.
5. being exhaled according to the intensity of apnea occurs in user's sleep procedure number and single to react the sleep of user
Inhale quality.
In the present embodiment, sleep monitor equipment can send the data information acquired in sleep procedure on mobile phone, or
It after being sent to information transit terminal, then uploads onto the server, and establishes sleep info for each individual using server
Library, to help user to make early warning to certain sleeping disorders.
The foregoing is merely the preferred embodiments of the invention, are not intended to limit the invention creation, all at this
Within the spirit and principle of innovation and creation, any modification, equivalent replacement, improvement and so on should be included in the invention
Protection scope within.
Claims (8)
1. a kind of portable sleep monitoring equipment, including Power Management Unit, communication unit and acceleration transducer, feature exist
In further including sign monitoring modular, sign monitoring modular is equipped with red-light LED, infrared light LED and optical sensor, and the monitoring is set
Standby housing body portion is opaque, double LED and the corresponding region light transmission of optical sensor on shell.
2. a kind of portable sleep monitoring equipment as described in claim 1, which is characterized in that the Power Management Unit includes electricity
Pond, Charge Management unit and linear voltage regulator, the VIN pin of model SGM4056, SGM4056 that Charge Management unit uses
The component of connection includes concatenated resistance R1 and capacitor C1, concatenated resistance R2, Schottky diode D1 and two pole of Schottky
The cathode of pipe D2, D1 and D2 are opposite, and resistance R3 is also connected between R2 and D1.
3. a kind of portable sleep monitoring equipment as claimed in claim 2, which is characterized in that the enable end of linear voltage regulator and D1
Leading point between D2 is connected.
4. a kind of portable sleep monitoring equipment as claimed in claim 2, which is characterized in that the communication unit includes that bluetooth is logical
Believe module and user terminal.
5. a kind of portable sleep monitoring equipment as described in claim 1, which is characterized in that the communication unit includes channel radio
Believe unit, information relay unit and server, passes through network communication between information relay unit and server.
6. a kind of portable sleep monitoring equipment as described in claim 1, which is characterized in that the pin of the sign monitoring modular
Including VCC_HO, it is used for digital power system, VCC_3V, for LED power supply, reset pin, power ground, hanging pin and serial communication
Pin.
7. a kind of portable sleep monitoring equipment as described in claim 1, which is characterized in that the sign of sign monitoring module monitors
Including pulse wave, heart rate value, blood oxygen levels and blood vessel microcirculation parameter.
8. a kind of portable sleep monitoring equipment as claimed in claim 7, which is characterized in that sign monitoring modular is every to be adopted through 64
Sampling point returns one acquisition data packet.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811614919.4A CN109745002A (en) | 2018-12-27 | 2018-12-27 | A kind of portable sleep monitoring equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811614919.4A CN109745002A (en) | 2018-12-27 | 2018-12-27 | A kind of portable sleep monitoring equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109745002A true CN109745002A (en) | 2019-05-14 |
Family
ID=66403242
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811614919.4A Pending CN109745002A (en) | 2018-12-27 | 2018-12-27 | A kind of portable sleep monitoring equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109745002A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110236503A (en) * | 2019-06-19 | 2019-09-17 | 杭州电子科技大学 | A kind of flexible wearable sleep physiology parameter detection method and device |
| CN110432870A (en) * | 2019-08-13 | 2019-11-12 | 重庆邮电大学 | A kind of sleep signal based on 1D CNN-LSTM method by stages automatically |
| CN112890828A (en) * | 2021-01-14 | 2021-06-04 | 重庆兆琨智医科技有限公司 | Electroencephalogram signal identification method and system for densely connecting gating network |
| CN112914506A (en) * | 2021-01-19 | 2021-06-08 | 青岛歌尔智能传感器有限公司 | Sleep quality detection method, device and computer readable storage medium |
| TWI754458B (en) * | 2019-11-18 | 2022-02-01 | 新加坡商先進分析科技私人有限公司 | Device, sysyem, and method for user apnea events detection |
| CN115363544A (en) * | 2022-08-31 | 2022-11-22 | 北京雪扬科技有限公司 | Method for monitoring respiratory frequency and sleep disordered breathing based on wearable device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN205913352U (en) * | 2016-05-23 | 2017-02-01 | 深圳眠虫科技有限公司 | Portable sleep monitor and physiological parameter monitor device |
| CN206251989U (en) * | 2016-09-30 | 2017-06-16 | 深圳大学 | A kind of Wearable multifunctional helmet device |
| CN206441329U (en) * | 2016-12-25 | 2017-08-25 | 广东交通职业技术学院 | Human body physical sign data remote transmission system based on ZIGBEE networks |
| CN109044323A (en) * | 2018-09-29 | 2018-12-21 | 天津惊帆科技有限公司 | Heart rate and oxygen saturation measurement equipment based on deep learning |
-
2018
- 2018-12-27 CN CN201811614919.4A patent/CN109745002A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN205913352U (en) * | 2016-05-23 | 2017-02-01 | 深圳眠虫科技有限公司 | Portable sleep monitor and physiological parameter monitor device |
| CN206251989U (en) * | 2016-09-30 | 2017-06-16 | 深圳大学 | A kind of Wearable multifunctional helmet device |
| CN206441329U (en) * | 2016-12-25 | 2017-08-25 | 广东交通职业技术学院 | Human body physical sign data remote transmission system based on ZIGBEE networks |
| CN109044323A (en) * | 2018-09-29 | 2018-12-21 | 天津惊帆科技有限公司 | Heart rate and oxygen saturation measurement equipment based on deep learning |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110236503A (en) * | 2019-06-19 | 2019-09-17 | 杭州电子科技大学 | A kind of flexible wearable sleep physiology parameter detection method and device |
| CN110236503B (en) * | 2019-06-19 | 2022-05-27 | 杭州电子科技大学 | Flexible wearable sleep physiological parameter detection method and device |
| CN110432870A (en) * | 2019-08-13 | 2019-11-12 | 重庆邮电大学 | A kind of sleep signal based on 1D CNN-LSTM method by stages automatically |
| TWI754458B (en) * | 2019-11-18 | 2022-02-01 | 新加坡商先進分析科技私人有限公司 | Device, sysyem, and method for user apnea events detection |
| US11723588B2 (en) | 2019-11-18 | 2023-08-15 | Advanced Analyzer Technologies Pte Ltd | Device for apnea detection, system and method for expediting detection of apnea events of a user |
| CN112890828A (en) * | 2021-01-14 | 2021-06-04 | 重庆兆琨智医科技有限公司 | Electroencephalogram signal identification method and system for densely connecting gating network |
| CN112914506A (en) * | 2021-01-19 | 2021-06-08 | 青岛歌尔智能传感器有限公司 | Sleep quality detection method, device and computer readable storage medium |
| CN115363544A (en) * | 2022-08-31 | 2022-11-22 | 北京雪扬科技有限公司 | Method for monitoring respiratory frequency and sleep disordered breathing based on wearable device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109745002A (en) | A kind of portable sleep monitoring equipment | |
| WO2016168980A1 (en) | Physiological sign information acquisition method and system | |
| CN203953662U (en) | A kind of Wearable sleep detection monitoring system | |
| CN204192596U (en) | Size intelligent healthcare instrument | |
| CN106073736A (en) | A kind of extendible human parameters monitoring and control recording system | |
| CN104257353A (en) | Sleep apnea syndrome detecting system | |
| CN110706816B (en) | Method and equipment for sleep environment regulation and control based on artificial intelligence | |
| CN107334460A (en) | Based on electrocardio, breathing, body position signal sleep quality assessment system and method | |
| WO2014210588A1 (en) | Systems and methods for determining sleep patterns and circadian rhythms | |
| US10736515B2 (en) | Portable monitoring device for breath detection | |
| CN205432239U (en) | If respiratory sensor's gauze mask and respiratory frequency monitored control system | |
| CN113633260B (en) | Polysomnography, computer equipment and readable storage medium | |
| CN102178529A (en) | Body area network-based respiratory disease remote diagnosis monitoring system | |
| CN204637245U (en) | A kind of sleep Clock system based on physiologic information monitoring | |
| CN201949007U (en) | Sleep breathing monitoring system | |
| CN107456218A (en) | A kind of mood sensing system and wearable device | |
| CN211862809U (en) | Novel wearable respiratory signal gathers waist binder | |
| CN103908241A (en) | Method and device for sleep and breathing detection | |
| CN114403808A (en) | Human sleep analysis method and device based on environmental monitoring and storage medium | |
| CN110811638A (en) | SVM classifier construction method, system and method for monitoring sleep | |
| CN208769759U (en) | Based on electrocardio, breathing, body position signal sleep quality assessment system | |
| CN113633277A (en) | Intelligent mattress with health monitoring function | |
| CN107349510A (en) | Fall asleep navigation and the depth of a kind of biofeedback loosen guidance system and method | |
| CN204468059U (en) | A kind of portable button health monitoring device | |
| CN107348945A (en) | A kind of monitoring system and method for snoring of talking in one's dream of non-recording |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |