CN104778342B - A kind of heart sound feature extracting method based on wavelet singular entropy - Google Patents
A kind of heart sound feature extracting method based on wavelet singular entropy Download PDFInfo
- Publication number
- CN104778342B CN104778342B CN201410015681.9A CN201410015681A CN104778342B CN 104778342 B CN104778342 B CN 104778342B CN 201410015681 A CN201410015681 A CN 201410015681A CN 104778342 B CN104778342 B CN 104778342B
- Authority
- CN
- China
- Prior art keywords
- heart sound
- cardiechema signals
- wavelet
- singular
- entropy
- 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.)
- Expired - Fee Related
Links
Landscapes
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
Abstract
The invention discloses a kind of heart sound feature extracting method based on wavelet singular entropy, and it is related to a kind of processing method of cardiechema signals.Cardiechema signals sample frequency of the present invention is 8000Hz；Utilize Hilbert-Huang（HHT）Heart sound envelope is extracted, is then segmented again based on heart sound envelope so as to obtain one section of complete cardiechema signals, both including s1, s2 and s3, s4；6 rank wavelet decompositions are carried out to segmentation cardiechema signals using DB6 as morther wavelet；Singular value decomposition is all carried out to the contour signal and detail signal obtained after wavelet transformation, respectively obtains respective singular value matrix；Calculate respective singular entropy；Then weighting obtains the characteristic value of cardiechema signals.Wavelet singular entropy is introduced into cardiechema signals feature extraction by the present invention, is not only able to highlight the feature of cardiechema signals HFS complexity, and can determine the respective frequencies of low frequency signal, and cardiechema signals day part is quantitatively described；Reduce the data operation quantity of cardiechema signals processing, improve the arithmetic speed of heart sound characteristics extraction.
Description
Technical field
The invention belongs to field of signal processing, and in particular to a kind of heart sound feature extracting method based on wavelet singular entropy.
Background technology
Cardiechema signals are the sound from heart collected using signal collecting device, wherein including important angiocarpy
Biological information.Cardiechema signals have entirely different feature on the different person and have high stability.
Lot of domestic and international scientific research personnel has attempted a variety of methods to analyze cardiechema signals at present.Gauthier is using quick
Fourier transformation（FFT）Analysis cardiechema signals (Gauthier D, Akay Y M, Paden R G,et al. Spectral
Analysis of Heart Sounds Associated with Coronary Occlusions [C].6th
International Special Topic Conference on Information Technology Applications
in Biomedicine,2007:49-52).Although Fourier transformation can realize time domain to frequency domain phase with good property
Mutually conversion.But can be seen that it from Fourier transform formula is using sine wave and its higher hamonic wave as standard base, therefore it is
A kind of analysis generally to signal, has a single local positioning ability, that is, in the good location of time domain is with frequency
Whole signal analysis in domain for cost, the good location to frequency domain be whole signal analysis using time domain as cost, Fourier
Either frequency is accurate and the time is fuzzy in the frequency spectrum of leaf transformation, or the time is accurate and frequency is fuzzy
, it can not possibly all have the ability of good positioning in time domain and frequency domain simultaneously.And cardiechema signals are concentrated mainly on s1 and s2
On, relatively whole heart sound cycle s1 and s2 are the periods of two relative brevities, therefore use Fourier transform pairs cardiechema signals
It is not a selection well to carry out analysis.
In addition with other power spectral densities, Sample Entropy method analysis cardiechema signals.But these methods all exist respective
Shortcoming：Power spectrum is premised on signal Gaussian distributed hypothesis, while lost the phase information of signal；Approximate entropy calculates
During the comparison of its data section be present, cause result deviation to be present.
The content of the invention
Present invention aim to address the problem present in cardiechema signals feature extraction at present, there is provided one kind is based on small echo
The heart sound feature extracting method of singular entropy.
Realize the technical scheme is that：It is characterized in that this method includes following 6 processes.
A) heart sound gathers.Cardiechema signals sample frequency is 4000Hz.
b）Heart sound envelope extraction and segmentation.The heart sound obtained in actual acquisition is long therefore needs to carry out envelope to heart sound
Extraction, the envelope extraction method used herein is to utilize Hilbert-Huang（HHT）Heart sound envelope is extracted, is then based on again
Heart sound envelope is segmented so as to obtain one section of complete cardiechema signals, both including s1, s2, and s3, s4.
c）Wavelet transformation.5 rank wavelet decompositions are carried out to segmentation cardiechema signals as morther wavelet using DB6.
d）Singular value decomposition.Because the frequency content of cardiechema signals is concentrated mainly on 300 below Hz, need herein
To the contour signal CA obtained after wavelet transformation（Low-frequency information）With detail signal CD（High-frequency information）All carry out singular value decomposition.
Respectively obtain respective singular value matrix Sa and Sd.Singular value decomposition form is as follows.
Wherein, andFor the whole of matrix A
Non-zero singular value.
e）Calculate singular entropy.Sa and Sd are substituted into following formula respectively and obtain contour signal CA singular entropy Ha and detail signal CD
Singular entropy Hd.
Wherein,。
f）Calculate characteristic value.Contour signal CA singular entropy Ha and detail signal CD singular entropy Hd are closed by weighting
And obtain final characteristic value.
Described heart sound characteristic value is using the singular entropy after heart sound wavelet decomposition, and the form by both according to weighting
With reference to.
The present invention has energetically effect：A kind of heart sound feature extracting method based on wavelet singular entropy of the present invention, no
But the feature of cardiechema signals HFS complexity can be highlighted, and the respective frequencies of low frequency signal can be determined, and
Cardiechema signals day part is quantitatively described.Wavelet singular entropy is introduced into cardiechema signals feature extraction by the present invention, is reduced
The data operation quantity of cardiechema signals processing, improve the arithmetic speed of heart sound characteristics extraction.
Brief description of the drawings
Fig. 1 is the cardiechema signals of sampling；
Fig. 2 is the cardiechema signals of segmentation；
Fig. 3 is heart sound profiles signal after wavelet transformation；
Fig. 4 is heart sound detail signal after wavelet transformation；
Fig. 5 is schematic flow sheet of the present invention；
Embodiment
（Embodiment 1）
A kind of heart sound feature extracting method based on wavelet singular entropy of present embodiment, detailed process such as Fig. 1 institutes
Show, be described as follows：
First, heart sound gathers.Cardiechema signals sample frequency is 8000Hz.Fig. 2 is the normal cardiac sound signal collected.
2nd, heart sound is segmented.The heart sound collected it is long therefore need to heart sound carry out envelope extraction, by this process from
And one section of complete cardiechema signals is obtained, both including s1, s2 and s3, s4.The envelope extraction method used herein is to utilize Xi Er
Bert-Huang（HHT）Heart sound envelope is extracted, is then segmented again based on heart sound envelope.Fig. 3 carries to be converted by HHT
The normal cardiac sound signal for the independent completion got.
3rd, wavelet transformation.This research is decomposed using DB6 as morther wavelet to cardiechema signals.Wavelet decomposition scales with
Sample frequency is relevant with wavelet basis.It is discrete small for the original sampled signal that a length is M according to the sampling thheorem of small echo
Wave Decomposition at most can be signal decomposition into log2M frequency level, therefore the yardstick of wavelet decomposition is limited and decomposition
Series is more, then the amount of calculation needed for it is bigger so according to practical application and should need to select the decomposition scale of small echo.
Cardiechema signals sample frequency used herein is 4000Hz, and according to wavelet sub-band implication, different wavelet coefficients represents difference
Band information, for example, 1 rank decompose detail signal cd1 represent frequency be：2000 Hz ~4000 Hz；The details that 2 ranks are decomposed
Signal cd2 represent frequency be：1000 Hz ~2000 Hz…….And cardiechema signals are analyzed by Short Time Fourier Transform, hair
Existing first heart sound s1 frequency content is concentrated mainly in the range of the Hz of 50 Hz ~ 150, and second heart sound s2 frequency content master
Concentrate in the range of the Hz of 50 Hz ~ 200, second small leak occur so selecting herein in the range of the Hz of 250 Hz ~ 300
Carry out 6 layers of conversion.So obtaining the frequency that the contour signal ca6 that 6 ranks are decomposed is represented is：The Hz of 0 Hz ~ 125, what 6 ranks were decomposed
Detail signal cd6 represent frequency be：125 Hz ~250Hz.
Fig. 4 is heart sound profiles signal after wavelet transformation；Fig. 5 is heart sound detail signal after wavelet transformation.
4th, singular value decomposition.Because the frequency content of cardiechema signals is concentrated mainly on 300 below Hz, need herein
Singular value decomposition is carried out to cd6, ca6, substitutes into following formula.Respectively obtain respective singular value matrix Sd and Sa.
5th, singular entropy is calculated.Sd and Sa are substituted into following formula respectively and obtain the singular entropy Ha for the contour signal ca6 that 6 ranks are decomposed
The detail signal cd6 decomposed with 6 ranks singular entropy Hd.
Wherein,。
6th, characteristic value is extracted.The singular entropy Ha of contour signal and the singular entropy Hd of detail signal are not closed by weighting
And obtain final characteristic value.
Claims (2)
1. a kind of heart sound feature extracting method based on wavelet singular entropy, it is characterised in that this method includes following 6 processes：
A) heart sound gathers, and cardiechema signals sample frequency is 4000Hz；
b）Heart sound envelope extraction and segmentation, the heart sound obtained in actual acquisition is long therefore needs to carry heart sound progress envelope
Take, the envelope extraction method used herein is to utilize Hilbert-Huang（HHT）Heart sound envelope is extracted, is then based on the heart again
Sound envelope is segmented so as to obtain one section of complete cardiechema signals, both including s1, s2 and s3, s4；
c）Wavelet transformation, 5 rank wavelet decompositions are carried out to segmentation cardiechema signals as morther wavelet using DB6；
d）Singular value decomposition, because the frequency content of cardiechema signals is concentrated mainly on 300 below Hz, need herein to small
The contour signal CA obtained after wave conversion（Low-frequency information）With detail signal CD（High-frequency information）Singular value decomposition is all carried out, respectively
Respective singular value matrix Sa and Sd is obtained, singular value decomposition form is shown in formula 1；
（1）
Wherein, andFor matrix A
Whole non-zero singular values；
e）Singular entropy is calculated, Sd and Sa is substituted into formula 2 respectively obtains contour signal CA singular entropy Ha and detail signal CD
Singular entropy Hd；
（2）
Wherein,；
f）Characteristic value is calculated, contour signal CA singular entropy Ha and detail signal CD singular entropy Hd are merged by weighting,
Obtain final characteristic value.
A kind of 2. described heart sound feature extracting method based on wavelet singular entropy before being required according to right 1, it is characterised in that：The heart
Sound characteristic value is using heart sound wavelet decomposition rear profile signal and the singular entropy of detail signal, and the shape by both according to weighting
Formula combines.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410015681.9A CN104778342B (en) | 2014-01-14 | 2014-01-14 | A kind of heart sound feature extracting method based on wavelet singular entropy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410015681.9A CN104778342B (en) | 2014-01-14 | 2014-01-14 | A kind of heart sound feature extracting method based on wavelet singular entropy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104778342A CN104778342A (en) | 2015-07-15 |
| CN104778342B true CN104778342B (en) | 2017-12-15 |
Family
ID=53619803
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410015681.9A Expired - Fee Related CN104778342B (en) | 2014-01-14 | 2014-01-14 | A kind of heart sound feature extracting method based on wavelet singular entropy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104778342B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105877706A (en) * | 2016-03-31 | 2016-08-24 | 济南大学 | Heart-sound enhancement method based on improved spectral subtraction |
| CN107480637B (en) * | 2017-08-15 | 2019-08-30 | 重庆大学 | Heart failure based on heart sound feature method by stages |
| CN110346157A (en) * | 2018-04-04 | 2019-10-18 | 国网安徽省电力有限公司电力科学研究院 | A kind of application method that wavelet singular entropy is detected in piler cyclic breakdown |
| CN110101407B (en) * | 2019-04-16 | 2021-09-07 | 华南师范大学 | Fetal heart sound denoising method, system and device and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202005762U (en) * | 2011-03-07 | 2011-10-12 | 济南大学 | Program-controlled amplifying and filtering device for heart sound signal acquisition |
-
2014
- 2014-01-14 CN CN201410015681.9A patent/CN104778342B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202005762U (en) * | 2011-03-07 | 2011-10-12 | 济南大学 | Program-controlled amplifying and filtering device for heart sound signal acquisition |
Non-Patent Citations (5)
| Title |
|---|
| Feature Extraction for Systolic Heart Murmur Classification;Christer Ahlistrom，et al.;《Annals of Biomedical Engineering》;20061130;第34卷(第11期);第1666-1677页 * |
| Matrix decomposition based feature extraction for murmur classification;Yuerong chen，et al.;《Medical Engineering & Physics》;20121231;第756-761页 * |
| 基于奇异谱分析的心音信号小波包去噪算法研究;卢德林，郭星明.;《振动与冲击》;20131231;第32卷(第18期);第63-69页 * |
| 基于心音的病理特性研究;高攀.;《中国优秀硕士学位论文全文数据库 医药卫生科技辑》;20120815(第08期);第E080-2页 * |
| 改进HHT算法及在心音信号分析中的应用;李彬彬，等.;《四川大学学报（工程科学版）》;20070731;第39卷(第4期);第160-163页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104778342A (en) | 2015-07-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107811649B (en) | Heart sound multi-classification method based on deep convolutional neural network | |
| Üstündağ et al. | Denoising of weak ECG signals by using wavelet analysis and fuzzy thresholding | |
| CN104778342B (en) | A kind of heart sound feature extracting method based on wavelet singular entropy | |
| CN102799892B (en) | Mel frequency cepstrum coefficient (MFCC) underwater target feature extraction and recognition method | |
| CN102697493A (en) | Method for rapidly and automatically identifying and removing ocular artifacts in electroencephalogram signal | |
| Wang et al. | Application of the dual-tree complex wavelet transform in biomedical signal denoising | |
| CN102646415B (en) | Characteristic parameter extraction method in speech recognition | |
| CN110169764A (en) | A kind of LMS adaptive-filtering PPG signal heart rate extracting method | |
| CN101972143A (en) | Blind source extraction-based atrial fibrillation monitoring method | |
| CN105686801A (en) | An atrial fibrillation spontaneous termination predication method based on inherent time scale decomposing | |
| CN104092447A (en) | Method for constructing biorthogonal wavelet filter bank | |
| CN105266800A (en) | Fetal electrocardiogram blind separation method based on low signal-to-noise ratio | |
| CN117688422A (en) | Under-modeling state parameter identification method based on improved sparse component analysis, computer equipment and storage medium | |
| CN103889335A (en) | A device and method for processing heart sounds for auscultation | |
| CN109009088B (en) | TWA multichannel fusion estimation method based on tensor decomposition | |
| CN104382589A (en) | Fetal electrocardiogram separation extraction method based on partial resampling by segments | |
| Ma et al. | A multichannel nonlinear adaptive noise canceller based on generalized FLANN for fetal ECG extraction | |
| CN101940475A (en) | Method for improving detection accuracy of blood oxygen saturation | |
| CN108836305A (en) | A kind of ECG feature extracting method of fusion Butterworth filtering and wavelet transformation | |
| CN112336369A (en) | Coronary heart disease risk index evaluation system of multichannel heart sound signals | |
| CN111143927A (en) | Constraint modal decomposition and frequency identification method based on structural response linear combination | |
| Sun et al. | A ECG waveform detection algorithm based on differential threshold and wavelet transform | |
| Sunjing et al. | Heart sound signals based on CNN classification research | |
| Xu et al. | Research on Heart Sound Denoising Method Based on CEEMDAN and Optimal Wavelet | |
| CN105740761A (en) | Weak-signal target detection optimizing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171215 Termination date: 20200114 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |