CN105520726B - Heartbeat detection mould group and its detection, denoising method - Google Patents
Heartbeat detection mould group and its detection, denoising method Download PDFInfo
- Publication number
- CN105520726B CN105520726B CN201410522143.9A CN201410522143A CN105520726B CN 105520726 B CN105520726 B CN 105520726B CN 201410522143 A CN201410522143 A CN 201410522143A CN 105520726 B CN105520726 B CN 105520726B
- Authority
- CN
- China
- Prior art keywords
- value
- heartbeat
- frequency domain
- denoising
- group
- 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
Abstract
The present invention relates to heartbeat detection mould group and its detections, denoising method.Heartbeat detection mould group includes light volume measurement device, motion sensing apparatus and processing unit.The smooth volume measurement device is to detect skin surface during the test with output light volume change signal.The motion sensing apparatus is to export acceleration signal during the relatively described detection.The processing unit determines denoising parameter to first frequency domain information denoising and according to the maximum spectrum peak computational heartbeat of the first frequency domain information after denoising to convert the body of light product variable signal and the acceleration signal respectively into the first frequency domain information and the second frequency domain information, according to the maximum spectrum peak value of second frequency domain information.
Description
Technical field
The present invention about a kind of heartbeat detection mould group, especially with regard to a kind of heartbeat detection mould group with denoising function and
Its detection, denoising method.
Background technique
It is dense that known blood oxygen saturation instrument (pulse oximeter) detects the blood oxygen of user in the way of non-intrusion type
Degree and Pulse Rate, can produce red light beam (about 660 nanometers of wavelength) and infrared light (about 910 nanometers of wavelength) penetrate it is to be measured
Position, and using band oxygen ferroheme (oxyhemoglobin) and deoxygenated blood red pigment (deoxyheamoglobin) to special spectrum
Characteristic with different absorptivities penetrates the intensity variation of light to detect, referring for example to U.S. Patent No. 7,072,701, mark
Monitoring mode (the Method for spectrophotometric blood oxygenation of entitled blood oxygen concentration
monitoring).After the intensity variation for penetrating light for detecting two kinds of wavelength, such as light volume change
(Photoplethysmography) signal or make PPG signal (PPG signal), then blood oxygen concentration calculated with following equation,
Blood oxygen concentration=100% × [HbO2]/([HbO2]+[Hb])；Wherein, [HbO2] indicates band oxygen hemoglobin concentration；[Hb] is indicated
Deoxygenated blood red pigment concentration.
Power can be presented with heartbeat in the luminous intensity for penetrating light of two kinds of wavelength detected by general blood oxygen saturation instrument
Variation, this is because blood vessel can with heartbeat and constantly expansion and contraction and blood volume that light beam is passed through changes, into
And change the absorbed ratio of light energy.Whereby, the heartbeat of user can be calculated according to continually changing intensity information.
However, can then detect chaotic waveform when blood oxygen saturation instrument and detected part detected relatively move
And it is difficult to detect by correct body of light product variable signal, thus under the testing conditions of nonstatic state (such as apply and can take
When formula electronic device or wearable electronic device), correct heartbeat may be unable to get.
Summary of the invention
In view of this, the present invention proposes a kind of heartbeat detection mould group and its detection, denoising method with denoising function.
The present invention provides a kind of heartbeat detection mould group.The heartbeat detection mould group includes light volume measurement device, kinesthesia
Survey device and processing unit.The smooth volume measurement device is to detect skin surface during the test with output light volume change
Signal.The motion sensing apparatus is to export acceleration signal during the relatively described detection.The processing unit is to respectively
Convert body of light product variable signal and the acceleration signal into the first frequency domain information and the second frequency domain information, according to described the
The maximum spectrum peak values of two frequency domain informations determines denoising parameter to denoise to first frequency domain information and according to after denoising the
The maximum spectrum peak computational heartbeat of one frequency domain information.
The present invention also provides a kind of heartbeat detecting method suitable for heartbeat detection mould group, the heartbeat detection mould group includes
Light volume measurement device, motion sensing apparatus and processing unit.The heartbeat detecting method is comprised the steps of with the body of light
Product measuring device detects skin surface during the test with output light volume change signal；With the motion sensing apparatus with respect to institute
Acceleration signal is exported during stating detection；The body of light product variable signal and acceleration letter are received with the processing unit
Number；Convert the body of light product variable signal and the acceleration signal respectively as the first frequency domain information and the second frequency domain information；Root
Denoising parameter is determined according to the maximum spectrum peak value of second frequency domain information to denoise to first frequency domain information；And according to
The maximum spectrum peak computational heartbeat of first frequency domain information after denoising.
The present invention also provides a kind of denoising methods of heartbeat detection mould group.The denoising method, which comprises the steps of, to be examined
Body of light product variable signal and acceleration signal are received during survey；The body of light product variable signal is converted as frequency domain light volume change letter
Number and generate have first group of frequency indices value and relevant first group of spectrum value the first frequency domain information；Convert the acceleration
Signal is frequency domain acceleration signal and generates second frequency domain with second group of frequency indices value and relevant second group of spectrum value
Information；Identify three frequency indices values and second frequency domain corresponding to first three big spectrum peak in first frequency domain information
Reference key value corresponding to maximum spectrum peak value in information；And according to three frequency indices values and the reference key
Value denoises first group of spectrum value.
In order to which above and other objects, features and advantages of the invention can be become apparent from, will hereafter be illustrated appended by cooperation, in detail
Carefully it is described as follows.In addition, identical component is indicated in explanation of the invention with identical symbol, first state clearly herein.
Detailed description of the invention
Fig. 1 is the block diagram of the heartbeat detection mould group of the embodiment of the present invention；
Fig. 2A accumulates the schematic diagram of variable signal for body of light before the filtering of the embodiment of the present invention；
Fig. 2 B accumulates the schematic diagram of variable signal for body of light after the filtering of the embodiment of the present invention；
Fig. 3 is the flow chart of the heartbeat detecting method of the embodiment of the present invention；
Fig. 4 A is the spectrogram of the frequency domain body of light product variable signal of the embodiment of the present invention；
Fig. 4 B is the schematic diagram of the first frequency domain information of the spectrogram of corresponding diagram 4A；
Fig. 5 A is the spectrogram of the frequency domain acceleration signal of the embodiment of the present invention；
Fig. 5 B is the schematic diagram of the second frequency domain information of the spectrogram of corresponding diagram 5A；
Fig. 6 is the first frequency domain information of the embodiment of the present invention and the schematic diagram of the second frequency domain information；
Fig. 7 is the flow chart of the denoising method of the embodiment of the present invention；
Fig. 8 is the frequency indices value, reference key value and the schematic diagram for denoising range of the embodiment of the present invention.
Description of symbols
1 heartbeat detection mould group
10 smooth volume measurement devices
12 motion sensing apparatus
14 processing units
140 modulus of conversion groups
142 peak values extract mould group
144 calculate mould group
Mould group is tracked in 146 heartbeats
16 bandpass filters
18 indicate unit
I1First frequency domain information
I2Second frequency domain information
N1、N2、N3Frequency indices value
NHRHeartbeat index value
PMAXMaximum spectrum peak value
PMAXMaximum spectrum peak value after ' denoising
R reference key value
R1/2Half times reference key value
R2Twice of reference key value
S10-S25Step
Sa acceleration signal
Sp body of light accumulates variable signal
Specific embodiment
The present invention provides a kind of heartbeat detection mould group with denoising function, can be combined in such as, but not limited to, intelligent
Wrist-watch, bracelet, glasses, wearable device or running gear.In some embodiments, the wearable device or running gear can be wrapped
Contain or do not include display function.In some embodiments, the heartbeat detection mould group can be independent detection device and using suitable
When mode is incorporated into described device, just it is configured when need to use, to increase practicability.
It please refers to shown in Fig. 1, is the block diagram of the heartbeat detection mould group 1 of the embodiment of the present invention, include light cubing
Device 10, motion sensing apparatus 12 and processing unit 14, wherein the processing unit 14 includes modulus of conversion group 140, peak value extraction
Mould group 142 and calculating mould group 144.In some embodiments, two bandpass filters 16 are separately positioned on the smooth cubing dress
It sets between 10 and the processing unit 14 and between the motion sensing apparatus 12 and the processing unit 14.Some embodiments
In, the processing unit 14 is also comprising heartbeat tracking mould group 146 to record the calculating mould group 144 heartbeat calculated.It can
To be appreciated that, power supply mould group (not being painted) is electrically connected the heartbeat detection mould group 1 and to provide the heartbeat detection mould
Required electric power when 1 operation of group.
The smooth volume measurement device 10 is to detect skin surface during the test with output light volume change signal Sp.
In general, the smooth volume measurement device 10 has illuminating module and sensing area.The smooth volume measurement device 10 can be anti-
Formula or penetration light volume measurement device are penetrated, specific limitation is had no.The smooth volume measurement device 10 is produced according to detection optical signal
The mode of third contact of a total solar or lunar eclipse volume change signal has been it is known that therefore details are not described herein.The detected skin of the smooth volume measurement device 10
The position on surface has no specific limitation, according to be applicable in electronic device.
The motion sensing apparatus 12 may be, for example, gyroscope (gyroscope), accelerometer (accelerometer),
Gravity sensor (G sensor) or other devices to sense human motion.In the present embodiment, the motion sensing apparatus
12 are illustrated by taking accelerometer as an example, export and accelerate during the detection to the relatively described smooth volume measurement device 10
Signal Sa is spent, so that the acceleration signal Sa and body of light product variable signal Sp have corresponding relationship.In one embodiment,
The motion sensing apparatus 12 can utilize the skill of MEMS (micro-electro-mechanical systems, MEMS)
Art is made.
In the present embodiment, there are two bandpass filters 16 to be separately positioned on the body of light product for the tool of heartbeat detection mould group 1
Between measuring device 10 and the processing unit 14 between the motion sensing apparatus 12 and the processing unit 14, and to
The body of light product variable signal Sp and acceleration signal Sa is filtered.For example, described in Fig. 2A and Fig. 2 B shows respectively
Body of light product variable signal Sp is by the schematic diagram of 16 front and back of bandpass filter, and wherein x-axis indicates the time and y-axis indicates vibration
Width.In general, mankind's heartbeat is between 30 beats/min to 240 beats/min, according to 60 beats/min of heartbeat corresponding 1Hz in the case where,
The signal frequency range of mankind's heartbeat is 0.5Hz to 4Hz.Therefore, the passband (passband) of the bandpass filter 16 is for example
It can increase the body of light product variable signal Sp's and acceleration signal Sa from 0.5Hz to 4Hz or from 0.45Hz to 4.5Hz
Signal quality (namely filtering out and the incoherent frequency of mankind's heartbeat signal), but not limited to this.To simplify explanation, present invention explanation
In, the filtered body of light product variable signal of the bandpass filter 16 and the acceleration signal still respectively with symbol Sp and
Sa is indicated.
It should be noted that although Fig. 1 shows that the bandpass filter 16 is not included among the processing unit 14,
The present invention is not limited thereto.In some embodiments, the bandpass filter 16 can be separately positioned on the smooth volume measurement device
10 and the motion sensing apparatus 12 among.In some embodiments, the bandpass filter 16 may be provided at the processing unit
Among 14.
The processing unit 14 be, for example, digital signal processor (digital signal processor, DSP) or its
He is used for the processing unit of signal processing, and realizes its calculation function using the mode of software, hardware or firmware.Institute
Stating processing unit 14 can be used to be eliminated according to the acceleration signal Sa in the body of light product variable signal Sp, due to the body of light
Noise caused by the sensing area of product measuring device 10 and skin surface relative motion.For example, in some embodiments, the processing
Unit 14 converts the body of light product variable signal Sp and acceleration signal Sa respectively as the first frequency domain information I1With the second frequency
Domain information I2, according to the second frequency domain information I2Maximum spectrum peak value determine denoising parameter to first frequency domain information
I1Denoising and according to the maximum spectrum peak computational heartbeat of the first frequency domain information after denoising.
The modulus of conversion group 140 of the processing unit 14 is to convert the body of light product variable signal Sp as frequency domain body of light
Product variable signal simultaneously generates first group of frequency indices value and relevant first group of spectrum value using as the first frequency domain information I1,
And the acceleration signal Sa is converted as frequency domain acceleration signal and generates second group of frequency indices value and relevant second group of frequency
Spectrum is using as the second frequency domain information I2。
The peak value extraction mould group 142 of the processing unit 14 is to judge the first frequency domain information I1With described
Two frequency domain information I2In multiple spectrum peaks, and export the frequency indices value of the corresponding spectrum peak to the calculating mould group
144。
The calculating mould group 144 of the processing unit 14 according to the frequency indices value of the correspondence spectrum peak to arrange
Except the first frequency domain information I1In noise to calculate heartbeat (be specified in after).
Heartbeat tracking mould group 146 to recorded during the multiple detections of correspondence the heartbeat variation tendency so that
In when the calculating mould group 144 can not according to the first frequency domain information directly calculates heartbeat after denoising when, can also be according to the variation
Trend estimates the heartbeat (after being specified in).
It will be appreciated that the embodiment of the present invention modulus of conversion group 140, the peak value extract mould group 142, the calculating mould
Group 144 and heartbeat tracking mould group 146 indicate function block (function block) inside the processing unit 14 or
Programming (program instruction).It is the modulus of conversion group 140, described it will be appreciated that in other embodiments
Peak value extracts mould group 142, the processing unit realization for calculating mould group 144 and heartbeat tracking mould group 146 can be different.It must
It should be noted that although showing two modulus of conversion groups 140 and two peak values extraction mould group 142, however, the present invention is not limited thereto, institute in Fig. 1
Stating processing unit 14 can be only comprising modulus of conversion group 140 and peak value extraction mould group 142.
In some embodiments, the heartbeat detection mould group 1 may include indicating unit 18 to the side by sound or image
Formula indicates the heartbeat, such as the expression unit 18 includes loudspeaker or display etc..The power supply mould group also provides at this time
Electric power needed for the expression unit 18.
In some embodiments, the expression unit 18 is not included among the heartbeat detection mould group 1, such as when the heart
It is described to indicate that unit 18 be the screen of Smartphone when jump detection mould group 1 is incorporated into intelligent bracelet.At this point, the heart
Jump detection mould group 1 wirelessly (such as bluetooth, Wi-Fi, ZigBee or other wireless communication protocols) from the wisdom
The transmission of type bracelet shows instant heartbeat and its variation tendency to the Smartphone comprising the signal of heartbeat message.
It is described to indicate the setting of unit 18 in the computer system of connection cloud system (cloud system) in some embodiments
System, at this time the heartbeat detection mould group 1 wirelessly transmit include heartbeat message signal to the cloud system with
The heartbeat is recorded for the cloud system.In medical applications, medical staff can be by making described in computer system monitoring
The heartbeat of user.
It will be appreciated that the heartbeat detection mould group 1 heartbeat detected can be used as various different applications, the present invention exists
In the signal noise for eliminating body of light product variable signal using acceleration signal, to promote the calculating accuracy rate of heartbeat.
Fig. 3 is the flow chart of the heartbeat detecting method of the embodiment of the present invention, is comprised the steps of with light volume measurement device
Detect skin surface during the test with output light volume change signal (step S10)；With the relatively described detection of motion sensing apparatus
Period exports acceleration signal (step S11)；The body of light product variable signal and the acceleration signal are received with processing unit
(step S12)；The body of light product variable signal is converted respectively and the acceleration signal is believed into the first frequency domain information and the second frequency domain
Cease (step S13)；Determine denoising parameter to first frequency domain information according to the maximum spectrum peak value of second frequency domain information
Denoise (step S14)；And maximum spectrum peak computational heartbeat (the step S according to the first frequency domain information after denoising15)。
Referring to Fig. 1,3,4A, 4B, 5A, 5B and 6, then illustrate the embodiment of the present embodiment；Wherein, Fig. 4 A and
4B is respectively the spectrogram of the frequency domain body of light product variable signal of one embodiment of the invention and schematic diagram, Fig. 5 A of the first frequency domain information
It is respectively the spectrogram of the frequency domain acceleration signal of one embodiment of the invention and the schematic diagram and Fig. 6 of the second frequency domain information with 5B
For the first frequency domain information of the embodiment of the present invention and the schematic diagram of the second frequency domain information.It will be appreciated that Fig. 4 A, 4B, 5A, 5B
It is only to illustrate, is not intended to limit the invention with 6.
Step S10-S11: firstly, the light volume measurement device 10 of heartbeat detection mould group 1 detects skin surface during the test
With output light volume change signal Sp；Meanwhile acceleration signal Sa is exported during the relatively described detection of motion sensing apparatus 12.For
Simplify explanation, the body of light product variable signal Sp and acceleration signal Sa in explanation can refer to after bandpass filter 16 filters below
Signal, and not otherwise stated.It should be noted that since the acceleration signal Sa mainly becomes to eliminate the body of light product
Change in signal Sp, it is described because of noise caused by the sensing area of the smooth volume measurement device 10 and skin surface relative motion
Body of light is accumulated during the variable signal Sp and acceleration signal Sa is preferably relevant to roughly the same detection so that in the heartbeat
Detecting mould group 1 can be according to the information of the acceleration signal Sa to body of light product variable signal Sp's while calculating heartbeat
Information is denoised.
Step S12: then, the processing unit 14 receives the body of light product variable signal Sp and acceleration letter simultaneously
Number Sa is to be post-processed.As shown in Figure 1, the body of light product variable signal Sp and acceleration signal Sa is separately input into institute
State the modulus of conversion group 140 of processing unit 14.
Step S13: the modulus of conversion group 140 of the processing unit 14 converts the body of light product variable signal Sp as frequency domain
Body of light product variable signal simultaneously generates first group of frequency indices value and relevant first group of spectrum value, wherein each frequency indices value pair
Answer spectrum value.It should be noted that the embodiment of the present invention modulus of conversion group 140 can utilize fast Fourier transform (Fast
Fourier Transform, FFT) body of light product variable signal Sp is converted from time domain to frequency domain (frequency
Domain) to generate the frequency domain body of light product variable signal, however, the present invention is not limited thereto.In other embodiments, the modulus of conversion
Discrete Fourier conversion (Discrete Fourier Transform, DFT) can also be used in group 140 or other time-frequency domains turn
It changes mode (namely spectrum analysis) and converts the body of light product variable signal Sp.
It will be appreciated that frequency domain body of light product variable signal be discrete signal so that the processing unit 14 can accordingly into
Row Digital Signal Processing.In some embodiments, when the body of light product variable signal that the smooth volume measurement device 10 is exported
Sp is continuous time-domain signal, and the modulus of conversion group 140 converts the body of light product variable signal Sp first as discrete time-domain signal
(such as sampling the body of light product variable signal Sp with sampling frequency), is then converted to discrete frequency-region signal, but not accordingly again
It is limited to this.In other embodiments, the modulus of conversion group 140 is first converted the body of light product variable signal Sp and is believed into continuous frequency domain
Number, be then converted to discrete frequency-region signal accordingly again.
As previously mentioned, the signal frequency range of mankind's heartbeat is between 0.5Hz between 4Hz.Assuming that the signal of mankind's heartbeat
The maximum value of frequency is 4Hz (240 beats/min corresponding), and the sampling frequency, which has to be larger than 8Hz (such as 10Hz or 20Hz), to expire
Sufficient Nyquist's theorem (Nyquist Theorem).In a kind of embodiment using fast Fourier transform, the sampling frequency
Rate is 20Hz, but not limited to this, depending on the operational capability of the processing unit 14.
The modulus of conversion group 140 is frequency domain body of light product using the product variable signal Sp of body of light described in fast Fourier transform
After variable signal, the spectrogram of the corresponding frequency domain body of light product variable signal can produce, as shown in Figure 4 A, wherein the frequency
The x-axis of spectrogram is the frequency indices value (frequency index of FFT) of fast Fourier transform, and y-axis is spectrum intensity.
In the present embodiment, the frequency indices value of Fig. 4 A and corresponding spectrum intensity are as first group of frequency indices value and relevant
One group of spectrum value namely the first frequency domain information I1, as shown in Figure 4 B.
It should be noted that the quantity of the frequency indices value of fast Fourier transform such as, but not limited to, be 1024 points
(bins), wherein each frequency indices value corresponds to a frequency.For example, frequency corresponding to frequency indices value 256 is (20Hz/
1024) × 256=5Hz.It will be appreciated that working as, the sampling frequency is 20Hz and the quantity of the frequency indices value is 1024
At a, the first frequency domain information I1Frequency resolution (frequency resolution) be 20Hz/1024=
0.0195Hz.When the sampling frequency is fixed value, between the more more then two side frequency index values of the quantity of frequency indices value
Difference on the frequency is smaller, so that susceptibility with higher when the heartbeat detection mould group 1 is according to the calculating heartbeat of frequency indices value.
It should be noted that first frequency domain is believed since mankind's heartbeat is generally between 30 beats/min to 240 beats/min
Cease I1The frequency indices value range of middle corresponding mankind's heartbeat is about between 25 to 205.Therefore, in some embodiments, the place
Reason unit 14 gives up (or release) frequency indices value less than 25 and/or greater than 205 and relevant spectrum value is to save system money
Source, but not limited to this.
In the same manner, another modulus of conversion group 140 utilizes the conversion body of light product variable signal Sp's in the processing unit 14
Same way converts the acceleration signal Sa to generate the spectrogram of the corresponding acceleration signal Sa, as shown in Figure 5A, and
Second group of frequency indices value and relevant second group of spectrum value are generated using as the second frequency domain information I2, as shown in Figure 5 B.It is certain
In embodiment, the second frequency domain information I2The frequency indices of the frequency indices value range (such as 25 to 205) can only be retained
Value and relevant spectrum value.
Step S14: obtaining the second frequency domain information I2Afterwards, peak value extracts mould group 142 according to second frequency domain information
I2Middle maximum spectrum peak value PMAXCorresponding frequency indices value determines reference key value R.For example, Fig. 6 is please referred to, described
In two groups of spectrum values, maximum spectrum value 460, the peak value extraction mould group 142 can recognize the maximum spectrum peak value P at this timeMAX
460 and to export the maximum spectrum peak value PMAXCorresponding frequency indices value 60 is to the calculating mould group 144 using as described
Reference key value R.Then, the reference key value R and twice of the ginseng for calculating mould group 144 and calculating half times
Examine index value R.For example, when the reference key value R is 60, half times reference key value R1/2For 30 and twice refers to
Index value R2It is 120.It will be appreciated that since the frequency indices value all represents frequency, twice of reference key value R2Institute
Corresponding frequency is the frequency multiplication of frequency corresponding to the reference key value R, and the half times reference key value R1/2Institute
Corresponding frequency is the half frequency multiplication of frequency corresponding to the reference key value R.
At this point, the calculating mould group 144 can be according to the reference key value R and the half times reference key value
R1/2With twice of reference key value R2Determine denoising parameter to denoise to first group of spectrum value one of at least within；
Such as denoising parameter may include index value R and R1/2, index value R and R2Or index value R and R1/2And R2.To first group of frequency spectrum
Value denoising, which refers to, excludes the first frequency domain information I according to the obtained denoising parameter of the reference key value R1Middle corresponding institute
State the spectrum value near reference key value.For example, working as the reference key value R1/2, R and R2Respectively 30,60 and 120, it is described
Processing unit 14 can add and subtract respectively preset range according to the reference key value and determine that denoising range is 20-40,50-70 and 110-
130 (namely 30 ± 10,60 ± 10 and 120 ± 10), and exclude the frequency of the related denoising range in first group of spectrum value
Spectrum is as to the first frequency domain information I1The mode of denoising.In some embodiments, the preset range is in the heartbeat
It is set before detection mould group 1 is dispatched from the factory or in the heartbeat detection mould group 1 initialization.
Further, since the second frequency domain information I2The denoising parameter is determined with for the processing unit 14, it is certain
In embodiment, the maximum spectrum peak value P is obtained from peak value extraction mould group 142 in the calculating mould group 144MAXOr it determines
After the denoising parameter, the processing unit 14 gives up (or release) described second frequency domain information I2To save system resource,
But not limited to this.
Step S15: finally, the mould group 144 that calculates is according to the maximum spectrum peak computational heart of the first frequency domain information after denoising
It jumps.In more detail, in the calculating mould group 144 from the first frequency domain information I1Institute is excluded when middle identification maximum spectrum peak value
State denoising range corresponding to spectrum value (namely in first group of frequency indices value frequency index value be 20-40,50-70 and 110-
Spectrum value corresponding to 130), such as (hatched example areas after the spectrum value according to corresponding to exclusion denoising range in the embodiment of Fig. 6
Indicate exclude spectrum value range), can determine denoising after the first frequency domain information maximum spectrum peak value be 930 (namely denoising after
Maximum spectrum peak value PMAX′).The calculating mould group 144 is according to maximum spectrum peak value P after the denoisingMAXThe frequency of ' corresponding frequency
Rate index value (namely 100) calculates the heartbeat.As previously mentioned, in the case where according to 60 beats/min of heartbeat corresponding 1Hz, the heart
Jump is (20/1024) × 100 × 60=117.19 beats/min.Accordingly, even if the smooth volume measurement device 10 is in nonstatic state
Lower output has the body of light product variable signal of chaotic waveform, and the heartbeat detection mould group 1 still can calculate standard according to the step
True heartbeat.
It should be noted that in the present embodiment, the calculating mould group 144 is from the first frequency domain information I1Middle identification is maximum
When spectrum peak when heartbeat (such as calculate) only exclude spectrum value corresponding to (or ignoring) described denoising parameter and it is indirect from
The spectrum value, however, the present invention is not limited thereto are deleted in memory body.In some embodiments, in step S15It goes before or described in decision
After parameter of making an uproar, the processing unit 14 can first remove the first frequency domain information I in memory body1The middle correlation denoising ginseng
Several frequency indices values and spectrum value are to save system resource.
It on the other hand, is to increase the accuracy for calculating heartbeat, in some embodiments, the processing unit 14 will be the after denoising
The maximum spectrum peak value (such as the P of one frequency domain informationMAX') corresponding frequency indices value is as heartbeat index value NHR(such as
100).Then, further according to the heartbeat index value NHRWith the heartbeat index value NHRSide frequency index value calculate the heart
It jumps.For example, please continue to refer to Fig. 6, as the heartbeat index value NHRIt is 100, the heartbeat detection mould group 1 is according to the heartbeat rope
Draw value NHRWith the heartbeat index value NHRAdjacent two frequency indices value 99 and 101 and its corresponding spectrum value of institute
930,890 and 920 energy barycenter is calculated as (99 × 890+100 × 930+101 × 920)/(890+930+920)=100.011.
Then, it is (20/1024) × 100.011 × 60=that the calculating mould group 144, which calculates the heartbeat according to the energy barycenter,
117.20 beat/min, but not limited to this.The heartbeat calculates 144 mould groups and can be indexed according to the heartbeat index value and the heartbeat
Multiple adjacent frequency indices values (such as 4 or 6) of value calculate the heartbeat.
Since the heartbeat detection mould group 1 can calculate a heartbeat during each detection, the heartbeat detection mould group 1 can root
Heartbeat variation tendency during calculating the detection according to the heartbeat numerical value during multiple detections simultaneously estimates heartbeat accordingly.Certain implementations
In example, the processing unit 14 is also comprising heartbeat tracking mould group 146 to record the heartbeat during the multiple detections of correspondence
Variation tendency.For example, in the embodiment in fig 6, (wherein the time is for example big after one time of user strenuous exercise
During the detection more than at least twice), it is assumed that the denoising range has not been changed and the heartbeat index value NHRBecome from 100
At 110, since the calculating mould group 144 is from the first frequency domain information I1Described go can be ignored when middle identification maximum spectrum peak value
Spectrum value corresponding to range of making an uproar (namely frequency index value is 110-130 corresponding to frequency spectrum in first group of frequency indices value
Value), the heartbeat index value N at this timeHRIt can be ignored, the calculating mould group 144 then can also track mould group 146 according to the heartbeat
The variation tendency (such as the heartbeat index value N in the time recordedHRBecome 110 trend from 100) estimation mesh
Preceding heartbeat.
According in the aspect of Fig. 6 embodiment, as the heartbeat index value NHRGradually become 110 from 100, the calculating mould
Spectrum value corresponding to the denoising range is ignored in 144 parts of group, such as ignores frequency index value in first group of frequency indices value
It is spectrum value corresponding to 20-40 and 50-70 without ignoring spectrum value corresponding to 110-130；That is, the calculating mould
Group 144 can be according to heartbeat index value NHRVariation by denoise range 110-130 be considered as invalid denoising range.At this point, the meter
Mould group 144 is calculated then according to the heartbeat index value NHROr after denoising the first frequency domain information maximum spectrum peak value (such as according to the
It is 120) to calculate the heartbeat that spectrum value, which is frequency indices value corresponding to 1350, in one group of spectrum value.
Fig. 7 is the flow chart of the denoising method of the embodiment of the present invention, comprises the steps of and receives body of light product during the test
Variable signal and acceleration signal (step S21)；The body of light product variable signal is converted as frequency domain body of light product variable signal and is generated
The first frequency domain information (step S with first group of frequency indices value and relevant first group of spectrum value22)；Convert the acceleration
Signal is frequency domain acceleration signal and generates second frequency domain with second group of frequency indices value and relevant second group of spectrum value
Information (step S23)；Identify in first frequency domain information three frequency indices values corresponding to first three big spectrum peak and described
Reference key value (step S corresponding to maximum spectrum peak value in second frequency domain information24)；And according to three frequency indices
Value and the reference key value are denoised (step S to first group of spectrum value25)。
Referring to shown in Fig. 1,6,7 and 8, then illustrate the embodiment of the present embodiment；Wherein Fig. 8 is the present invention one
Frequency indices value, reference key value and the schematic diagram for denoising range of embodiment.
Step S21: firstly, receiving body of light product variable signal Sp and acceleration signal Sa during the test.It is scrutable
It is that the body of light product variable signal Sp and acceleration signal Sa is for example respectively by light volume measurement device 10 and motion sensing
Device 12 is issued, as shown in Figure 1.
Step S22: then, the body of light product is converted using the conversion regime of fast Fourier transform or other time-frequency domains
Variable signal Sp is frequency domain body of light product variable signal and generates with first group of frequency indices value and relevant first group of spectrum value
The first frequency domain information I1, such as shown in Fig. 6.
Step S23: it is using the acceleration signal Sa is converted with the same way for converting the body of light product variable signal Sp
Frequency domain acceleration signal simultaneously generates the second frequency domain information I with second group of frequency indices value and relevant second group of spectrum value2。
In the present embodiment, since there are two independent modulus of conversion group 140, step S for the heartbeat detection mould group 1 tool23With step S22It can be same
Shi Jinhang, but not limited to this.
It will be appreciated that processing unit 14 can retain the first frequency domain information I1With the second frequency domain information I2In
Required frequency indices value and frequency spectrum value information are simultaneously stored in memory unit, for example, only reserve frequency index value 0 to 225 and
Relevant spectrum value, but not limited to this.
Step S24: obtain the first frequency domain information I1With the second frequency domain information I2Later, the processing unit 14
Identify the first frequency domain information I1In three frequency indices value N corresponding to first three big spectrum peak1、N2、N3With described second
Frequency domain information I2Reference key value R corresponding to middle maximum spectrum peak value.For example, the first frequency domain information I1In first three big frequency
Three frequency indices values N corresponding to spectrum peak1、N2And N3Respectively 58,73 and 117, and the second frequency domain information I2
The reference key value R corresponding to middle maximum spectrum peak value is 120, as shown in Figure 8.
Step S25: finally, the processing unit 14 calculates half times reference key value according to the reference key value R
R1/2And/or twice of reference key value R260 and 240 and to determine denoising range, wherein the denoising range is for example according to
Reference key value R1/2, R and R2Plus-minus 5 is respectively to determine and be 55-65,115-125 and 235-245, as shown in Figure 8.Accordingly,
The processing unit 14 is according to three frequency indices values N1-N3The denoising range determined with the reference key value R
Frequency domain body of light product variable signal is denoised.
As previously mentioned, the smooth volume measurement device 10 may export incorrect body of light product under nonstatic state
Variable signal, so that the processing unit 14 can not directly calculate accurate heartbeat according to body of light product variable signal.Therefore,
The step S through this embodiment21-S25After determining the denoising range, the first frequency domain information I1Described in middle correspondence
The relevant spectrum value of frequency indices value institute for denoising range may be noise, and the processing unit 14 can exclude first frequency
Domain information I1The frequency indices value of the middle correspondence denoising range or relevant spectrum value are to the first frequency domain information I1It carries out
Denoising.
According to the denoising method, a kind of application can for example calculate heartbeat.Please continue to refer to Fig. 8, when first frequency
Domain information I1Frequency indices value N1And N3Fall into the denoising range (namely 58 and 117 respectively 55-65 and 115-125 it
Between) and frequency indices value N2Not within the scope of the denoising, the processing unit 14 can be according to the denoising range from described
Three frequency indices value N1-N3Among determine heartbeat index value NHRFor 73 (namely frequency indices value N2).Then, the processing is single
Member 14 can be according to the heartbeat index value NHRCalculate the heartbeat.For example, the heartbeat is (20/1024) × 73 × 60=
85.55 beats/min.In some embodiments, the processing unit 14 is according to the heartbeat index value NHRWith the heartbeat index value NHR
Side frequency index value calculate the heartbeat, calculation method is as previously mentioned, therefore details are not described herein.
It should be noted that the denoising range is with the frequency indices value N1-N3On the basis of, then add and subtract preset range (example
Such as add and subtract 5) caused by, wherein the preset range is related or the sampling frequency of the uncorrelated modulus of conversion group 140 and
The quantity of frequency indices value.As previously mentioned, the quantity of the sampling frequency and frequency indices value determines frequency resolution.Certain realities
It applies in example, the preset range and the frequency resolution are in inversely related, but not limited to this.
In some embodiments, the processing unit 14 is also according to the denoising range from three index value N1-N3In
Determine two remaining index values (namely the index value N that is 58 and 1171And N3).Assuming that the denoising range and the remaining index value
N1、N3Do not change, after one time of user strenuous exercise, since the heartbeat of the user rises, the related heartbeat
Index value N2It can move closer to index value N3So that index value N2It falls within denoising range (namely index value 115-125).
At this point, the processing unit 14 can not then be indexed according to the middle decision heartbeat of denoising range from three index values
Value NHR.Therefore, as the heartbeat index value NHR(such as index value N2) one of wherein (such as indexed with the remaining index value
Value N1Or N3) difference be less than threshold value when, the processing unit 14 equally can be according to the heartbeat index value NHRCorresponding multiple institutes
Variation tendency during stating detection estimates the heartbeat.
For example, it is assumed that the threshold value is the 10 and heartbeat index value NHRBecome 110 from 73 by the time, at this time institute
State heartbeat index value NHRWith (namely the frequency indices value N of the remaining index value 1173) difference be 7 and be less than the threshold value, institute
Processing unit 14 is stated then according to the heartbeat index value NHRDescribed in variation tendency estimation during corresponding multiple detections
Heartbeat, wherein the method for heartbeat is calculated according to the variation tendency and frequency indices value as previously mentioned, therefore details are not described herein.
In the various embodiments described above, the body of light product variable signal Sp of the smooth volume measurement device 10 and the kinesthesia
The acceleration signal Sa of device 12 is surveyed not only to calculate heartbeat, the processing unit 14 can also be according to body of light product
Variable signal Sp and the acceleration signal Sa calculate the physiological status of user and exercise data (such as step counting, calculates running
Or cycle speed and record run duration), depending on practical application.
In summary, it is known that the blood oxygen saturation instrument of heartbeat detection mould group calculates heartbeat under the testing conditions of nonstatic state
When can generate incorrect body of light product variable signal, and reduce the accuracy for calculating heartbeat.Therefore, the present invention also proposes a kind of tool
There are the heartbeat detection mould group (Fig. 1) and its detection (Fig. 2), denoising method (Fig. 7) of denoising function, is determined by acceleration signal
Surely denoising parameter uses the accuracy for improving and calculating heartbeat to filter out the noise in body of light product measuring signal.
Although the present invention is disclosed by previous examples, it is not intended to limit the invention, belonging to any present invention
In technical field with usual knowledge technical staff, without departing from the spirit and scope of the present invention, when can make it is various more
Dynamic and modification.Therefore protection scope of the present invention is subject to the range defined depending on appended claims.
Claims (17)
1. a kind of heartbeat detection mould group, which includes:
Light volume measurement device, to detect skin surface during the test with output light volume change signal；
Motion sensing apparatus, to export acceleration signal during the relatively described detection；And
Processing unit, to
Convert body of light product variable signal and the acceleration signal respectively as the first frequency domain information and the second frequency domain information,
In, first frequency domain information includes first group of frequency indices value and relevant first group of spectrum value, and second frequency domain is believed
Breath includes second group of frequency indices value and relevant second group of spectrum value,
Denoising parameter is determined according to the maximum spectrum peak value of second frequency domain information, and according to denoising parameter decision
Denoising range in first group of frequency indices value, the denoising range include multiple frequency indices values,
The relevant spectrum value of the multiple frequency indices value of the denoising range in first group of spectrum value is excluded with right
First frequency domain information denoises and generates the first frequency domain information after denoising,
According to the maximum spectrum peak computational heartbeat of the first frequency domain information after the denoising, and
Record the variation tendency of the heartbeat during corresponding multiple detections, and the first frequency domain when can not be according to the denoising after
When information directly calculates heartbeat, current heartbeat is estimated also according to the variation tendency.
2. heartbeat detection mould group according to claim 1, wherein the processing unit includes:
Modulus of conversion group, to convert the body of light product variable signal as frequency domain body of light product variable signal and generate first group of frequency
Rate index value and relevant first group of spectrum value, and convert the acceleration signal as frequency domain acceleration signal and generate institute
State second group of frequency indices value and relevant second group of spectrum value；And
Peak value extracts mould group, to judge in first frequency domain information and second frequency domain information multiple spectrum peaks and defeated
The frequency indices value of the spectrum peak is corresponded to out.
3. heartbeat detection mould group according to claim 2, wherein first group of frequency indices value and second group of frequency
Rate index value has identical quantity.
4. heartbeat detection mould group according to claim 2, wherein the processing unit also includes to calculate mould group to basis
Frequency indices value corresponding to the maximum spectrum peak value of second frequency domain information determines reference key value using as the denoising
Parameter, and the reference key value and twice of the reference key value of half times are calculated, wherein the denoising parameter
Also the reference key value comprising the half times and described twice of the reference key value be one of at least within.
5. heartbeat detection mould group according to claim 4, wherein the processing unit also includes:
Mould group is tracked in heartbeat, to record the variation tendency of the heartbeat during the multiple detections of correspondence,
Wherein, the calculating mould group estimates the current heartbeat also according to the variation tendency.
6. heartbeat detection mould group according to claim 1, which also includes:
Bandpass filter, to be filtered to body of light product variable signal and the acceleration signal.
7. heartbeat detection mould group according to claim 1, wherein the heartbeat detection mould group be incorporated into wearable device or
Running gear.
8. a kind of heartbeat detecting method for being useful in heartbeat detection mould group, the heartbeat detection mould group is filled comprising light cubing
It sets, motion sensing apparatus and processing unit, the heartbeat detecting method includes:
Detect skin surface during the test with the smooth volume measurement device with output light volume change signal；
To export acceleration signal during the relatively described detection of the motion sensing apparatus；
The body of light product variable signal and the acceleration signal are received with the processing unit；
Convert body of light product variable signal and the acceleration signal respectively as the first frequency domain information and the second frequency domain information,
In, first frequency domain information includes first group of frequency indices value and relevant first group of spectrum value, and second frequency domain is believed
Breath includes second group of frequency indices value and relevant second group of spectrum value；
Denoising parameter is determined according to the maximum spectrum peak value of second frequency domain information, and according to denoising parameter decision
Denoising range in first group of frequency indices value, the denoising range include multiple frequency indices values；
The relevant spectrum value of the multiple frequency indices value of the denoising range in first group of spectrum value is excluded with right
First frequency domain information denoises and generates the first frequency domain information after denoising；
According to the maximum spectrum peak computational heartbeat of the first frequency domain information after the denoising；And
Record the variation tendency of the heartbeat during corresponding multiple detections, and the first frequency domain when can not be according to the denoising after
When information directly calculates heartbeat, current heartbeat is estimated also according to the variation tendency.
9. heartbeat detecting method according to claim 8, wherein convert respectively the body of light product variable signal and it is described plus
In the step of speed signal is first frequency domain information and second frequency domain information, also include:
The body of light product variable signal is converted as frequency domain body of light product variable signal and generates first group of frequency indices value and institute
State relevant first group of spectrum value；And
The acceleration signal is converted as frequency domain acceleration signal and generates second group of frequency indices value and described relevant
Second group of spectrum value.
10. heartbeat detecting method according to claim 9, which also includes:
Reference key value is determined according to frequency indices value corresponding to the maximum spectrum peak value in second frequency domain information；
And
Using the multiple value of the reference key value one of at least with the reference key value as the denoising parameter.
11. heartbeat detecting method according to claim 10, wherein according to the first frequency domain information after the denoising
In the step of heartbeat described in maximum spectrum peak computational, also include:
Heartbeat index value is determined according to the maximum spectrum peak value of the first frequency domain information after the denoising；And
The heartbeat is calculated according to the heartbeat index value.
12. heartbeat detecting method according to claim 10, wherein according to the first frequency domain information after the denoising
In the step of heartbeat described in maximum spectrum peak computational, also include:
Heartbeat index value is determined according to the maximum spectrum peak value of the first frequency domain information after the denoising；And
The heartbeat is calculated according to the side frequency index value of the heartbeat index value and the heartbeat index value.
13. heartbeat detecting method according to claim 8, which also includes:
Body of light product variable signal and the acceleration signal are filtered with bandpass filter.
14. a kind of denoising method of heartbeat detection mould group, which includes:
Body of light product variable signal and acceleration signal are received during the test；
The body of light product variable signal is converted as frequency domain body of light product variable signal and is generated with first group of frequency indices value and phase
First frequency domain information of the first group of spectrum value closed；
The acceleration signal is converted as frequency domain acceleration signal and is generated with second group of frequency indices value and relevant second
Second frequency domain information of group spectrum value；
Identify three frequency indices values and second frequency domain corresponding to first three big spectrum peak in first frequency domain information
Reference key value corresponding to maximum spectrum peak value in information；
Determine that denoising range, the denoising range include multiple frequency indices values according to the reference key value；
According in three frequency indices values not the reference key value it is corresponding it is described denoising range in frequency indices
Value determines heartbeat index value, to denoise to first group of spectrum value；
Heartbeat is calculated according to the heartbeat index value；And
The variation tendency of the heartbeat is recorded during corresponding multiple detections, and when the heartbeat index value is changed to described go
When range of making an uproar, current heartbeat is estimated according to the variation tendency of record.
15. denoising method according to claim 14, which also includes:
Calculate the reference key value and twice of the reference key value of half times；And
According to the reference key value, the reference key value of the half times with described with reference to rope with described twice
Draw value and determines the denoising range.
16. denoising method according to claim 15, which also includes:
The heartbeat index value and two remaining index values are determined among three frequency indices values according to the denoising range.
17. denoising method according to claim 16, which also includes:
When the difference of the heartbeat index value and one of the remaining index value is less than threshold value, according to multiple detections
The variation tendency of period estimates the current heartbeat.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410522143.9A CN105520726B (en) | 2014-09-30 | 2014-09-30 | Heartbeat detection mould group and its detection, denoising method |
| CN201910586579.7A CN110292372B (en) | 2014-09-30 | 2014-09-30 | Detection device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410522143.9A CN105520726B (en) | 2014-09-30 | 2014-09-30 | Heartbeat detection mould group and its detection, denoising method |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910586579.7A Division CN110292372B (en) | 2014-09-30 | 2014-09-30 | Detection device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105520726A CN105520726A (en) | 2016-04-27 |
| CN105520726B true CN105520726B (en) | 2019-07-30 |
Family
ID=55763491
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410522143.9A Active CN105520726B (en) | 2014-09-30 | 2014-09-30 | Heartbeat detection mould group and its detection, denoising method |
| CN201910586579.7A Active CN110292372B (en) | 2014-09-30 | 2014-09-30 | Detection device |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910586579.7A Active CN110292372B (en) | 2014-09-30 | 2014-09-30 | Detection device |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN105520726B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10478128B2 (en) | 2014-09-26 | 2019-11-19 | Pixart Imaging Inc. | Heart rate detection architecture |
| CN108852326A (en) * | 2017-05-08 | 2018-11-23 | 原相科技股份有限公司 | The rhythm of the heart detects framework |
| CN114224309B (en) * | 2017-11-16 | 2023-11-17 | 原相科技股份有限公司 | Physiological detection system and physiological detection device |
| CN110141203A (en) * | 2018-02-12 | 2019-08-20 | 光宝新加坡有限公司 | Heart rate detecting system and the wearable device for using it |
| CN112494001B (en) * | 2020-11-25 | 2023-05-26 | 青岛歌尔智能传感器有限公司 | PPG signal quality evaluation method and device, wearable device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6129676A (en) * | 1993-12-20 | 2000-10-10 | Seiko Instruments Inc. | Movement rate monitor |
| CN1762300A (en) * | 2004-10-18 | 2006-04-26 | 香港中文大学 | Human body physiological parameter monitor device based on high frequency light capacity trace signal |
| CN101039617A (en) * | 2004-10-15 | 2007-09-19 | 普尔塞特拉瑟技术有限公司 | Motion cancellation of optical input signals for physiological pulse measurement |
| CN102625667A (en) * | 2009-09-11 | 2012-08-01 | 高通股份有限公司 | Method and apparatus for artifacts mitigation with multiple wireless sensors |
| WO2013109390A1 (en) * | 2012-01-16 | 2013-07-25 | Valencell, Inc. | Reduction of physiological metric error due to inertial cadence |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7914420B2 (en) * | 2007-07-18 | 2011-03-29 | Brunswick Corporation | Sensing applications for exercise machines |
| CN103781414B (en) * | 2011-09-16 | 2016-08-24 | 皇家飞利浦有限公司 | For estimating equipment and the method for the heart rate during motion |
| GB201302451D0 (en) * | 2013-02-12 | 2013-03-27 | Isis Innovation | Method and system for signal analysis |
-
2014
- 2014-09-30 CN CN201410522143.9A patent/CN105520726B/en active Active
- 2014-09-30 CN CN201910586579.7A patent/CN110292372B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6129676A (en) * | 1993-12-20 | 2000-10-10 | Seiko Instruments Inc. | Movement rate monitor |
| CN101039617A (en) * | 2004-10-15 | 2007-09-19 | 普尔塞特拉瑟技术有限公司 | Motion cancellation of optical input signals for physiological pulse measurement |
| CN1762300A (en) * | 2004-10-18 | 2006-04-26 | 香港中文大学 | Human body physiological parameter monitor device based on high frequency light capacity trace signal |
| CN102625667A (en) * | 2009-09-11 | 2012-08-01 | 高通股份有限公司 | Method and apparatus for artifacts mitigation with multiple wireless sensors |
| WO2013109390A1 (en) * | 2012-01-16 | 2013-07-25 | Valencell, Inc. | Reduction of physiological metric error due to inertial cadence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110292372A (en) | 2019-10-01 |
| CN110292372B (en) | 2022-03-01 |
| CN105520726A (en) | 2016-04-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI538660B (en) | Heart rate detection module, and detection and denoising method thereof | |
| US10987007B2 (en) | Method of processing electrophysiological signals and corresponding system, vehicle, and computer program product | |
| CN105520726B (en) | Heartbeat detection mould group and its detection, denoising method | |
| EP3229676B1 (en) | Method and apparatus for physiological monitoring | |
| US9936886B2 (en) | Method for the estimation of the heart-rate and corresponding system | |
| EP3334337B1 (en) | Monitoring of sleep phenomena | |
| US11771375B2 (en) | Respiration rate detection device and breath detection device adopting motion denoising | |
| CN109222949A (en) | Heart rate detection method and heartbeat detection device | |
| US11412972B2 (en) | Detection of atrial fibrillation | |
| CN108937957B (en) | Detection method, device and detection equipment | |
| KR102463076B1 (en) | Oxygen saturation measuring apparatus and oxygen saturation measuring method thereof | |
| US20210153756A1 (en) | Reliable acquisition of photoplethysmographic data | |
| EP3773185A1 (en) | Photoplethysmogram data analysis and presentation | |
| US10405802B2 (en) | Method for detecting a heart rate | |
| WO2018106345A1 (en) | Cardiovascular parameter estimation in the presence of motion | |
| EP3906854A1 (en) | Method and device to determine sleep apnea of a user | |
| De Giovanni et al. | Ultra-low power estimation of heart rate under physical activity using a wearable photoplethysmographic system | |
| JP2013106837A (en) | Heart rate detection method, heart rate detector, and mental stress measuring apparatus | |
| CN104688199B (en) | A kind of contactless pulses measure method based on skin pigment concentration difference | |
| CN107405086A (en) | Determine device, assay method and program | |
| US20220369943A1 (en) | Method and system for evaluating the quality of ratio of ratios values | |
| TW201511735A (en) | A PPG-based physiological sensing system with a spatio-temporal sampling approach towards identifying and removing motion artifacts from optical signals | |
| Yuan et al. | Effects of autonomic nervous system on the quality of non-invasive blood diagnosis by PPG-based sensor system | |
| JC | Spot and continuous monitoring of heart rate by combining time and frequency domain analysis of photoplethysmographic signals at rest conditions. | |
| CN108852326A (en) | The rhythm of the heart detects framework |
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 |