CN108921090A - A kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis - Google Patents
A kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis Download PDFInfo
- Publication number
- CN108921090A CN108921090A CN201810703162.XA CN201810703162A CN108921090A CN 108921090 A CN108921090 A CN 108921090A CN 201810703162 A CN201810703162 A CN 201810703162A CN 108921090 A CN108921090 A CN 108921090A
- Authority
- CN
- China
- Prior art keywords
- heart rate
- signal
- fetal heart
- wavelet
- time
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
- G06F2218/06—Denoising by applying a scale-space analysis, e.g. using wavelet analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
- G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
Abstract
The present invention discloses a kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis.The present invention is for passing through the collected Doppler's fetal heart rate signal of ultrasonic probe, noise suppression preprocessing is carried out to fetal heart rate signal first with the method for empirical mode decomposition and wavelet energy distribution, time frequency analysis is carried out using wavelet transformation for pretreated fetal heart rate signal and obtains Wavelet time-frequency figure, it recycles Paasche Wa Er theorem to convert time energy diagram for Wavelet time-frequency figure, the instantaneous heart rate value of ultrasonic Doppler fetal heart rate signal is obtained by the time interval between time energy diagram upward peak.The present invention calculates the Fetal Heart Rate of collected ultrasonic Doppler fetal heart rate signal, and method is simple and effective and stablizes, and flexibility is good, and accuracy is high.
Description
Technical field
The present invention relates to supersonic sounding fields, in particular to a kind of to be based on empirical modal (Empirical Mode
Decomposition, EMD) decompose and wavelet time-frequency analysis Fetal Heart Rate extracting method.
Background technique
With popularizing for prenatal and postnatal care, the strategy of the happy China of Health China is promoted, and people increasingly pay attention to medical technology
Development, and people-oriented theory also gradually penetrates into current medicine and hygiene fields.And fetal heart monitoring be exactly ensure it is pregnant
There are also fetal well-beings by woman, realize a kind of important channel of prenatal and postnatal care.The heart of fetus not only provides oxygen to fetus, also provides
Nutrition needed for growth and development, heartbeat affect the constancy of whole body.Fetal rhythm directly or indirectly receives blood flow
In dynamic change there are also except the adjusting of the body fluid such as hormone, also suffer from the domination of the central nervous system in brain.This
Cutting can be embodied by the variation of fetal heart frequency.Monitoring fetal heart frequency be exactly big event in current fetal heart monitoring it
One, so the extraction and monitoring to Fetal Heart Rate are significantly.
Data shows that the signal component returned by ultrasonic probe is complicated, and includes the noise that much will cause interference, is
A kind of nonlinear random signal of the non-stationary of narrowband, it is also difficult to provide a specific digital signal model.At present on the market
The signal processing method that uses mostly of doppler baby's heart instrument be：A bandpass filter is first passed around, removal partial noise is dry
After disturbing, the envelope of signal is extracted, auto-correlation then is carried out to envelope.It is thicker to the noise suppression preprocessing of signal in this method
Rough, measurement accuracy is less reliable, especially in the lower situation of Signal-to-Noise.So on this basis, many scholars exist
The Fetal Heart Rate of fetal heart rate signal can carry out a noise suppression preprocessing to signal first in extracting, and current more scholar attempts to be become with small echo
Change or blind separation in independent component analysis method to fetal heart rate signal carry out noise suppression preprocessing, also some scholars are to Wavelet Denoising Method
It improves and applies in the pretreatment of fetal heart rate signal.But based on the signal denoising of wavelet transformation firstly the need of to wavelet basis into
The wavelet basis of row selection, difference selection has bigger difference to denoising effect, so such noise suppression preprocessing method is artificially selected
It interferes larger；It is pre-processed using independent component analysis method, effective fetal heart rate signal can be extracted when noise is relatively high,
But noise it is relatively low when denoising effect it is not satisfactory.
Based on this, the present invention provides a kind of Fetal Heart Rate extraction side based on empirical mode decomposition and wavelet time-frequency analysis
Method.Empirical mode decomposition is to utilize warp in a kind of widely used method of nonlinear and non local boundary value problem process field at present
Test during mode decomposition decomposes signal, the frequency of each intrinsic mode function component decomposed it is orderly from big
To small.Empirical mode decomposition method based on this multiresolution has the advantage of wavelet transformation, while overcoming in Wavelet Denoising Method
Choose probabilistic problem of basic function.And decomposed from signal itself, local adaptation is better than other decomposition methods,
Based entirely on signal feature itself, there are very powerful flexibility and validity.In addition, after fetal heart rate signal pretreatment, this hair
It is bright that subsequent processing is carried out to fetal heart rate signal from time frequency analysis, the time-frequency distributions of fetal heart rate signal are obtained using wavelet time-frequency analysis,
It can intuitively observe that signal frequency changes over time trend, in order to further extract the instantaneous heart rate value of fetal heart rate signal, this
Wavelet time-frequency figure is converted to time energy diagram using Paasche Wa Er theorem by invention, passes through each peak value on time energy diagram later
Time interval obtains the instantaneous heart rate value of fetal heart rate signal.
Summary of the invention
According to some drawbacks present in current fetal heart rate signal treatment process, the present invention provides one kind to be based on empirical modal
It decomposes and the Fetal Heart Rate extracting method of wavelet time-frequency analysis, the method remains to more effectively extract tire when noise is relatively low
Heart rate value, and method is simple and stablizes, and flexibility is good, and accuracy is high, and this method can be applied in fetus-voice meter.
The purpose of the present invention is realized at least through one of following technical solution.
Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis, including：For passing through ultrasonic probe
Doppler's fetal heart rate signal back, removes fetal heart rate signal first with the method that empirical mode decomposition is distributed with wavelet energy
It makes an uproar pretreatment, time frequency analysis is carried out to pretreated signal using wavelet analysis and obtains Wavelet time-frequency figure, recycles Paasche watt
Wavelet time-frequency figure is converted time energy diagram by that theorem, is then obtained by the time interval between time energy diagram upward peak
The instantaneous heart rate value of ultrasonic Doppler fetal heart rate signal.Fetal Heart Rate meter of the present invention for collected ultrasonic Doppler fetal heart rate signal
It calculates, method is simple and effective and stablizes, and flexibility is good, and accuracy is high.
Based on the above technical solution, the method being distributed using empirical mode decomposition and wavelet energy is to fetal rhythm
Signal carries out noise suppression preprocessing process, is that a kind of non-stationary is nonlinear random by the Doppler's fetal heart rate signal of probe back
Signal, noise is relatively low, and denoising process is to obtain a set of frequencies successively from height after signal is carried out empirical mode decomposition first
To low intrinsic modal components (Intrinsic Mode Function, IMF), since noise is usually more distributed in signal
Radio-frequency component, and corresponding low-frequency component is smaller by noise jamming, so there are an intrinsic modal components IMFk, there is IMFk
And the component before it is all that noise accounts for leading ingredient, and IMFk+1It and in the component after it is all that useful signal accounts for master
Ingredient is led, this critical point is obtained using the wavelet energy distribution curve of fetal heart rate signal, class is carried out to the component before critical point
Wavelet threshold processing, the signal after critical point retain, then important come reconstruction signal using institute, obtain pretreated tire
Heart signal.
Based on the above technical solution, the pretreated fetal heart rate signal carries out time frequency analysis, is to utilize small echo
Analysis obtains Wavelet time-frequency figure, judge that fetal heart rate signal frequency changes with time trend by Wavelet time-frequency figure, from T/F connection
It closes and observes fetal heart rate signal feature in distribution, in order to further extract heart rate value, using Paasche Wa Er theorem by Wavelet time-frequency figure
It is converted to time energy curve, obtains Doppler's fetal heart rate signal using the time interval between each peak value on time energy diagram later
Instantaneous heart rate value.
Based on the above technical solution, a kind of Fetal Heart Rate based on empirical mode decomposition and wavelet time-frequency analysis
Extracting method specific implementation step is as follows：
(1) empirical mode decomposition (EMD) is carried out to fetal heart rate signal signal, obtains N number of IMF modal components；
(2) centre frequency is sought to obtained each IMF component；
(3) the wavelet energy distribution map of fetal heart rate signal is drawn, and finds a corresponding maximum point on this figure, really
This maximum point respective frequencies is made to this range of the centre frequency of signal most high-order；Maximum point if it does not exist is then won the confidence
Number lowest center frequency to this frequency range of highest centre frequency；
(4) IMF is foundkComponent, the centre frequency corresponding to it is just in (3) in affiliated frequency range, and IMFk+1
For the centre frequency of component not in (3) in affiliated frequency range, this k value is our critical point being looked for；
(5) IMF is takenk+1And its IMF that component, that is, noise before plays a leading role1~IMFk+1, carry out class wavelet threshold
Processing obtains IMF1'~IMFk+1'；
(6) reconstruction signal：This signal is the pretreated tire needed for us
Heart signal.
Wherein in the class wavelet threshold processing in (5), selection of threshold function soft-threshold：
Wherein, sgn is sign function, x>When 0, sgn (x)=1；When x=0, sgn (x)=0；x<When 0, sgn (x)=- 1.
For jth layer IMF component, Wo MenquWherein L is signal length, σjFor the noise mark in the signal of jth layer
It is quasi- poor, herein using σj=media/0.6745 is estimated that media is the absolute intermediate value of the IMF component of signal jth layer,
Above formula is applied on IMF component, is modified slightly to obtain：
T in above formulajIt is the threshold value of jth layer component, calculation expression is as follows：
Wherein media () is the function for taking intermediate value.
(7) time frequency analysis that small echo is carried out to pretreated signal, obtains Wavelet time-frequency figure；
(8) time energy distribution curve is obtained by Paasche Wa Er theorem, is obtained according to peak curve time interval instantaneous
Heart rate value.
Compared with prior art, the invention has the advantages that and technical effect:
The present invention carries out noise suppression preprocessing to fetal heart rate signal using empirical mode decomposition method, wherein utilizing empirical modal point
Solution carries out signal to decompose the advantage with wavelet transformation, while overcoming and choosing the probabilistic of basic function in Wavelet Denoising Method
Problem.And decomposed from signal itself, local adaptation is better than other decomposition methods, based entirely on signal feature itself,
There are very powerful flexibility and validity.It remains to reach preferable denoising effect when noise is relatively low.
Time frequency distribution map is acquired to fetal heart rate signal using wavelet time-frequency analysis, can be visually observed that fetal heart rate signal frequency with
The Variation Features of time, and be converted to the instantaneous heart rate value that time energy curve acquires and compare current auto-correlation extracting method
It is more stable and effective.
Detailed description of the invention
Fig. 1 is original Doppler's fetal heart rate signal figure of the invention.
Fig. 2 is specific implementation flow chart of the invention.
Fig. 3 is that the present invention is based on the fetal heart rate signal noise suppression preprocessing flow charts of empirical mode decomposition.
Fig. 4 is each component that fetal heart rate signal of the present invention obtains after empirical mode decomposition.
Fig. 5 is the wavelet energy distribution curve of fetal heart rate signal of the present invention.
Fig. 6 is the fetal heart rate signal after noise suppression preprocessing of the present invention.
Fig. 7 is the Wavelet time-frequency figure of signal after present invention pretreatment.
Fig. 8 is the time energy diagram that the present invention is converted according to Wavelet time-frequency figure.
Fig. 9 is the instantaneous heart rate figure of fetal heart rate signal of the present invention.
Specific embodiment
For a better understanding of the invention, below in conjunction with attached drawing to the present invention is based on empirical mode decompositions and Wavelet time-frequency point
The Fetal Heart Rate extracting method of analysis is further described.
It is as shown in Figure 1 original Doppler's fetal heart rate signal figure, thus figure can find that original fetal heart rate signal noise is relatively low,
Noise jamming is serious, and the periodicity of fetal heart rate signal is unobvious, can be carried out subsequent heart rate so noise suppression preprocessing must be carried out
It extracts.
It is illustrated in figure 2 the flow chart that the present invention is implemented.For passing through the collected Doppler's fetal heart rate signal of ultrasonic probe,
Noise suppression preprocessing is carried out to fetal heart rate signal first with the method for empirical mode decomposition and wavelet energy distribution, utilizes wavelet transformation
Guangdong Province's plan of science and technology public good research and capacity building special project project carry out time frequency analysis to pretreated signal and obtain small echo
Time-frequency figure recycles Paasche Wa Er theorem by Wavelet time-frequency figure to convert time energy diagram, by time energy diagram upward peak it
Between time interval obtain the instantaneous heart rate value of ultrasonic Doppler fetal heart rate signal.
It is illustrated in figure 3 the flow chart of fetal heart rate signal noise suppression preprocessing, pretreated specific implementation step is as follows：
Step 1：Empirical mode decomposition (EMD) is carried out to fetal heart rate signal, obtains N number of IMF modal components, and is obtained each
The centre frequency of component, obtained each component map such as Fig. 4；
Step 2：Draw the wavelet energy distribution map of fetal heart rate signal, wavelet energy distribution map such as Fig. 5, and on this figure
A corresponding maximum point is found, determines this maximum point respective frequencies to this model of the centre frequency of signal most high-order
It encloses；Maximum point if it does not exist, then the lowest center frequency for the number of winning the confidence to this frequency range of highest centre frequency；
Step 3：Find an IMFkComponent, the centre frequency corresponding to it just in step 2 in affiliated frequency range, and
And IMFk+1For the centre frequency of component not in step 2 in affiliated frequency range, this k value is the critical point to be looked for；
Step 4：Take IMFk+1And its IMF that component, that is, noise before plays a leading role1~IMFk+1, carry out class small echo
Threshold process obtains IMF1'~IMFk+1'；
Step 5：Reconstruction signal：This signal is pretreated needed for us
Fetal heart rate signal, fetal heart rate signal such as Fig. 6 after denoising.
The threshold function table wherein used in step 4 is：
Wherein, sgn is sign function, x>When 0, sgn (x)=1；When x=0, sgn (x)=0；x<When 0, sgn (x)=- 1.
For jth layer IMF component, Wo MenquWherein L is signal length, σjFor the noise mark in the signal of jth layer
It is quasi- poor, herein using σj=media/0.6745 is estimated that media is the absolute intermediate value of the IMF component of signal jth layer,
Above formula is applied on IMF component, is modified slightly to obtain：
T in above formulajIt is the threshold value of jth layer component, calculation expression is as follows：
Wherein media is the function for taking intermediate value.
It is illustrated in figure 7 the Wavelet time-frequency figure of fetal heart rate signal after noise suppression preprocessing, thus figure can be observed on each time point
Frequency component.
It is illustrated in figure 8 to further extract instantaneous heart rate, Wavelet time-frequency figure is transformed time energy diagram, by
Observable goes out significantly periodically this figure.
It is as shown in Figure 9 the instantaneous heart rate figure of Doppler's fetal heart rate signal according to obtained by time energy diagram.
The above embodiment is a preferred embodiment of the present invention, but embodiments of the present invention are not by above-described embodiment
Limitation, other any changes, modifications, substitutions, combinations, simplifications made without departing from the spirit and principles of the present invention,
It should be equivalent substitute mode, be included in protection scope of the present invention.
Claims (4)
1. a kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis, it is characterised in that specifically include：
For the Doppler's fetal heart rate signal returned by ultrasonic probe, first with the method for empirical mode decomposition and wavelet energy distribution
Noise suppression preprocessing is carried out to fetal heart rate signal, time frequency analysis is carried out to pretreated signal with wavelet transformation and obtains Wavelet time-frequency
Figure recycles Paasche Wa Er theorem to convert time energy diagram for Wavelet time-frequency figure, by between time energy diagram upward peak
Time interval obtains the instantaneous heart rate value of ultrasonic Doppler fetal heart rate signal.
2. a kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis according to claim 1,
It is characterized in that being a kind of nonlinear random signal of non-stationary by the Doppler's fetal heart rate signal of probe back, noise compares
It is low, it is to obtain a set of frequencies successively from high to low after signal is carried out empirical mode decomposition first during noise suppression preprocessing
Intrinsic modal components IMF, there are an intrinsic modal components IMFk, there is IMFkAnd the component before it is all that noise accounts for master
Lead ingredient, and IMFk+1And in the component after it be all that useful signal accounts for leading ingredient, utilize the wavelet energy of fetal heart rate signal
Distribution curve obtains this critical point, carries out class wavelet threshold processing to the component before critical point, the signal after critical point
Retain, it is then important come reconstruction signal using institute, obtain pretreated fetal heart rate signal.
3. a kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis according to claim 1,
It is characterized in that pretreated fetal heart rate signal progress wavelet transformation is obtained Wavelet time-frequency figure, fetal rhythm is judged by Wavelet time-frequency figure
Signal frequency changes with time trend, and the fetal heart rate signal feature from T/F Joint Distribution utilizes Paasche Wa Erding
Wavelet time-frequency figure is converted to time energy curve by reason, obtains Doppler using the time interval between each peak value on time energy diagram
The instantaneous heart rate value of fetal heart rate signal.
4. a kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis according to claim 1,
It is characterized in that specifically comprising the following steps:
(1) empirical mode decomposition (EMD) is carried out to fetal heart rate signal signal, obtains N number of IMF modal components；
(2) centre frequency is sought to obtained each IMF component；
(3) the wavelet energy distribution map of fetal heart rate signal is drawn, and finds a corresponding maximum point on this figure, is determined
This maximum point respective frequencies is to this range of the centre frequency of signal most high-order；Maximum point if it does not exist, the then number of winning the confidence
Lowest center frequency is to this frequency range of highest centre frequency；
(4) IMF is foundkComponent, the centre frequency corresponding to it is just in (3) in affiliated frequency range, and IMFk+1Component
Centre frequency not in (3) in affiliated frequency range, this k value is our critical point being looked for；
(5) IMF is takenk+1And its IMF that component, that is, noise before plays a leading role1~IMFk+1, carry out class wavelet threshold and handle
To IMF1'~IMFk+1'；
(6) reconstruction signal：This signal is the pretreated fetal rhythm letter needed for us
Number；
Wherein in the class wavelet threshold processing in (5), selection of threshold function soft-threshold：
Wherein, sgn is sign function, x>When 0, sgn (x)=1；When x=0, sgn (x)=0；x<When 0, sgn (x)=- 1；For
Jth layer IMF component, takesWherein L is signal length, σjIt is poor for the noise criteria in the signal of jth layer, herein
Utilize σj=media/0.6745 is estimated that media is the absolute intermediate value of the IMF component of signal jth layer, and above formula is applied
On IMF component, it is modified slightly to obtain：
T in above formulajIt is the threshold value of jth layer component, calculation expression is as follows：
Wherein media is the function for taking intermediate value；
(7) time frequency analysis that small echo is carried out to pretreated signal, obtains Wavelet time-frequency figure；
(8) time energy distribution curve is obtained by Paasche Wa Er theorem, instantaneous heart rate is obtained according to peak curve time interval
Value.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810703162.XA CN108921090A (en) | 2018-06-30 | 2018-06-30 | A kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis |
| CN201910132143.0A CN109657660B (en) | 2018-06-30 | 2019-02-22 | Fetal heart rate extraction method based on empirical mode decomposition and wavelet time-frequency analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810703162.XA CN108921090A (en) | 2018-06-30 | 2018-06-30 | A kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108921090A true CN108921090A (en) | 2018-11-30 |
Family
ID=64422280
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810703162.XA Pending CN108921090A (en) | 2018-06-30 | 2018-06-30 | A kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis |
| CN201910132143.0A Active CN109657660B (en) | 2018-06-30 | 2019-02-22 | Fetal heart rate extraction method based on empirical mode decomposition and wavelet time-frequency analysis |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910132143.0A Active CN109657660B (en) | 2018-06-30 | 2019-02-22 | Fetal heart rate extraction method based on empirical mode decomposition and wavelet time-frequency analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN108921090A (en) |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4569356A (en) * | 1984-11-05 | 1986-02-11 | Nihon Kohden Corporation | Method and apparatus for detecting fetal heart rate by autocorrelation |
| WO2001082099A1 (en) * | 1999-05-01 | 2001-11-01 | The Court Of Napier University | Method of analysis of medical signals |
| US7747316B2 (en) * | 2005-01-27 | 2010-06-29 | The Board Of Trustees Of The University Of Illinois | Blind adaptive filter extraction of fetal electrocardiogram signal estimate |
| WO2012017364A1 (en) * | 2010-08-02 | 2012-02-09 | Koninklijke Philips Electronics N.V. | Method of reducing ultrasound signal ambiguity during fetal monitoring |
| CN103263262B (en) * | 2013-05-28 | 2015-02-11 | 捷普科技（上海）有限公司 | System and method for measuring heart rate of fetus |
| CN103424600A (en) * | 2013-08-20 | 2013-12-04 | 昆明理工大学 | Voltage sag source identification method based on Hilbert-Huang transformation and wavelet packet energy spectra |
| CN103462642B (en) * | 2013-08-20 | 2015-05-06 | 广东工业大学 | Instant heart rate detection method and device for Doppler fetal heart sound based on time-frequency analysis |
| CN104182625A (en) * | 2014-08-15 | 2014-12-03 | 重庆邮电大学 | Electrocardiosignal denoising method based on morphology and EMD (empirical mode decomposition) wavelet threshold value |
| CN106798554A (en) * | 2017-01-12 | 2017-06-06 | 安徽大学 | A kind of denoising method of noisy IMF components and electrocardiosignal |
| CN106889981B (en) * | 2017-01-26 | 2018-04-20 | 浙江铭众科技有限公司 | A kind of intelligent terminal for being used to extract fetal heart frequency |
| CN107693004A (en) * | 2017-09-05 | 2018-02-16 | 广东工业大学 | Fetal ECG extraction and fetal heart frequency recognition methods based on hilbert conversion |
-
2018
- 2018-06-30 CN CN201810703162.XA patent/CN108921090A/en active Pending
-
2019
- 2019-02-22 CN CN201910132143.0A patent/CN109657660B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109657660B (en) | 2023-04-25 |
| CN109657660A (en) | 2019-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110338786B (en) | Epileptic discharge identification and classification method, system, device and medium | |
| CN108416367B (en) | Sleep staging method based on multi-sensor data decision-level fusion | |
| US20080262371A1 (en) | Method for Adaptive Complex Wavelet Based Filtering of Eeg Signals | |
| CN111657889A (en) | Non-contact driver fatigue detection method based on millimeter wave radar | |
| CN102488516A (en) | Nonlinear electroencephalogram signal analysis method and device | |
| CN104873173A (en) | Non-contact type sleep stage classification and sleep breathing disorder detection method | |
| CN108056770A (en) | A kind of heart rate detection method based on artificial intelligence | |
| CN107693004A (en) | Fetal ECG extraction and fetal heart frequency recognition methods based on hilbert conversion | |
| CN104688220A (en) | Method for removing ocular artifacts in EEG signals | |
| CN104305992B (en) | A kind of interactive fast automatic extracting method of Fetal ECG | |
| CN101972145A (en) | Fetus electrocardio blind separation method based on relative sparsity of time domain of source signal | |
| CN108992053B (en) | Method for real-time non-binding detection of heart rate and heartbeat interval | |
| CN110974196A (en) | Non-contact respiration and heart rate detection method in motion state | |
| CN108717535A (en) | A kind of depth of anesthesia method of estimation based on composite character and long memory network in short-term | |
| CN109657660A (en) | A kind of Fetal Heart Rate extracting method based on empirical mode decomposition and wavelet time-frequency analysis | |
| CN106691437A (en) | Fetal heart rate extraction method based on maternal electrocardiosignals | |
| Lee et al. | Vital sign quality assessment based on IR-UWB radar sensor | |
| CN114983359A (en) | Vital sign detection method based on IR-UWB radar | |
| Anishchenko et al. | Comparative analysis of methods for classifying the cardiovascular system's states under stress | |
| CN115251845A (en) | Sleep monitoring method for processing brain wave signals based on TB-TF-BiGRU model | |
| Lu et al. | Pulse waveform analysis for pregnancy diagnosis based on machine learning | |
| CN109146006B (en) | Template library combined nonlinear template matching fetal heart rate extraction method | |
| CN106682630A (en) | Feature sampling method based on electrocardiographic signal | |
| Zhang et al. | Human pulse recognition based on wavelet transform and BP network | |
| CN106419912A (en) | Multi-lead electroencephalogram signal ocular artifact removing method |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181130 |