CN106344006A - J wave detection method based on pole symmetrical mode decomposition and support vector machine - Google Patents
J wave detection method based on pole symmetrical mode decomposition and support vector machine Download PDFInfo
- Publication number
- CN106344006A CN106344006A CN201610952984.2A CN201610952984A CN106344006A CN 106344006 A CN106344006 A CN 106344006A CN 201610952984 A CN201610952984 A CN 201610952984A CN 106344006 A CN106344006 A CN 106344006A
- Authority
- CN
- China
- Prior art keywords
- vector machine
- wave
- support vector
- electrocardiosignal
- mode decomposition
- 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.)
- Granted
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
- A61B5/346—Analysis of electrocardiograms
- A61B5/349—Detecting specific parameters of the electrocardiograph cycle
Abstract
The invention relates to an algorithm for detecting J wave in an electrocardiosignal based on pole symmetrical mode decomposition and a support vector machine, which is widely suitable for assisting a doctor in diagnosing whether the electrocardiosignal of a patient has J wave and is an important basis for J wave syndrome diagnosis. The invention specifically discloses a J wave detection method based on pole symmetrical mode decomposition and a support vector machine. J wave is ectrosis behind a joint point J between the end of electrocardiosignal QRS wave and the beginning of an ST section. The J wave syndrome initiated by abnormal J wave may result in high risk diseases such as malignant ventricular arrhythmia and sudden cardiac death, and takes more and more attention in clinical medicine. By using the pole symmetrical mode decomposition method, the problems of mode aliasing and the like in empirical mode decomposition are solved, ECG signals are decomposed into a series of eigen mode functions and a trend term, an instantaneous frequency feature and an energy transformation feature are extracted, the dimensionality is reduced by principal component analysis and the signals are input to the support vector machine for J wave detection.
Description
Technical field
The present invention relates to a kind of calculation based on j ripple in pole symmetric mode decomposition and supporting vector machine testing electrocardiosignal
Method, be widely used in assist diagnosis patient electrocardiosignal in whether there is j ripple, be j ripple syndrome diagnosis important according to
According to specially a kind of j wave detecting method based on pole symmetric mode decomposition and support vector machine.
Background technology
J point is that in electrocardiosignal, qrs ripple terminates the binding site initiateing with st section, indicates end and the multipole of sequences of ventricular depolarization
Beginning.Pause and transition in rhythm or melody after j point is j ripple, is a kind of variation of common normal ECG.The j ripple being caused by abnormal j ripple is comprehensive
Close disease and can lead to the symptoms such as malignant ventricular arrhythmia, sudden cardiac death, increasingly receive attention and the concern of medical circle.Accurately
Detection j ripple, can effectively reduce sickness rate and fatality rate, have great importance to clinic.
Because electrocardiosignal is nonlinear and non local boundary value problem, in the research process of conventional j ripple diagnosis, mainly have in Fu a.
Leaf transformation, is changed by the frequency content that study of disease causes, achieves certain success, but frequency domain analysises can only obtain entirely
Office's frequency information, and electrocardiosignal is nonlinear and non local boundary value problem, the diagnosis of j ripple needs to study its local frequency information.B. little
Wave conversion, has good effect in electrocardiosignal identification classification, but the defect of wavelet transformation is non-adaptive, its performance good
Bad depend critically upon wavelet function selection.C. Hilbert-Huang transform, is to detect the more popular method of j Bob at present.Its
In, empirical mode decomposition is the important part of Hilbert-Huang transform, is Huang E (n.e.huang) in NASA
A kind of NEW ADAPTIVE signal time frequency processing method creatively proposing in 1998 with other people, it is according to signal itself
Fluctuation pattern signal decomposition is different time scales component from low to high, more suitable than time and frequency domain analysis method
Close the research of nonlinear and non local boundary value problem.Just because of such feature, Hilbert-Huang transform method is successfully applied to electrocardio
The research of signal medical diagnosis on disease aspect.But detect j ripple using Hilbert-Huang transform there is problems in that screening number of times is difficult
To determine, the trend function decompositing is too rough, and Hilbert spectral analysis are limited by mathematical theory.The present invention is based on above-mentioned asking
Topic, is studied to electrocardiosignal using pole symmetric mode decomposition method, and then extracts validity feature, by support vector machine
Detection j ripple, thus improve the Detection accuracy of j ripple.
Content of the invention
The present invention is in order to solve the above problems, there is provided the j ripple inspection based on pole symmetric mode decomposition and support vector machine
Survey method.
The present invention adopts the following technical scheme that realization: the inspection based on pole symmetric mode decomposition and support vector machine
The method of j ripple in the thought-read signal of telecommunication, comprises the following steps:
The first step: obtain original electro-cardiologic signals x (t), original electro-cardiologic signals x (t) include normal electrocardiosignal and ripple containing j
Electrocardiosignal, find out all maximum points of each original electro-cardiologic signals x (t) and minimum point, be denoted as pi, 1≤i≤n；
Second step: with line segment by all of extreme point piIt is sequentially connected, the midpoint of every section of line segment is labeled as ti, 1≤i≤n-
1, and border midpoint t is replenished at two ends about the total segment that line segment connection is formed0And tn；
3rd step: build s article of interpolat curve l using n+1 midpoint1,…,ls, s >=1, calculate its meansigma methods l*=(l1
+…ls)/s；
4th step: to x (t)-l*Sequence repeats above three step, until | l*|≤ε or screening number of times reach default
Spacing value k, ε is allowable error, obtains first intrinsic mode function c1(t)；
5th step: to residue sequence x (t)-c1Repeat aforementioned four step, until residual term r is single signal or no longer
More than previously given extreme point, intrinsic mode function c just can be respectively obtained2(t),c3(t)…,ci(t),cmT (), calculates sequence
The variances sigma of row x (t)-r2；
6th step: limiting interval [kmin,kmax] spacing value k of interior change, repeat above-mentioned five steps, and to σ/σ0And k
Drawn, σ0It is the standard deviation of x (t), find out σ/σ in figure0The corresponding k of minima0, with k0Weigh again as restrictive condition
Multiple above-mentioned five steps, last residual term r is exactly the self adaptation overall situation all lines of primary signal x (t), that is, trend term, through undue
Solution, original electro-cardiologic signals x (t) are represented byUtilize pole symmetric mode decomposition by electrocardiosignal
X (t) resolves into a series of intrinsic mode functions and a residual term；
7th step: (1) finds intrinsic mode function ciT the extreme point of (), calculates two adjacent maximum points and two phases
Time difference between adjacent minimum point；
(2) by these time differences be considered as local period value be assigned to two adjacent maximum points and two adjacent minimum points it
Between midpoint on time shafts, drawing the time cycle corresponds to point diagram；
(3) obtain its local frequencies by inverted for these local period values, then do cubic spline interpolation obtaining instantaneous frequency
Curve；
8th step: the corresponding analytical expression of i-th intrinsic mode function is:
9th step: choose instantaneous frequency f1, f2, f3, f4, f5, f6, f7 and the energy conversion of the first seven intrinsic mode function
Feature e (t) composition characteristic space [f1, f2, f3, f4, f5, f6, f7, e (t)], is entered to feature space from PCA
Row dimensionality reduction, chooses front 8 main constituents of each feature, and that is, the feature after dimensionality reduction amounts to 64；
Tenth step: features described above is input to support vector machine, support vector machine are trained, extract the heart to be detected
Feature after the dimensionality reduction of the signal of telecommunication is input to support vector machine, you can tell the type of electrocardiosignal to be detected.
The present invention is non-linear non-stationary property using electrocardiosignal, and electrocardiosignal is carried out pole symmetric mode decomposition,
Solve end condition in empirical mode decomposition not knowing, the problems such as modal overlap, will the instantaneous frequency that try to achieve and energy variation special
Levy and carry out principal component analysiss dimensionality reduction, reduce operand, and then detect the electrocardiosignal containing j ripple using support vector machine, improve
Detection accuracy, assists doctor's Accurate Diagnosis j ripple syndrome.
Specific embodiment
Based on the j wave detecting method of pole symmetric mode decomposition and support vector machine, comprise the following steps:
The first step: obtain original electro-cardiologic signals x (t), original electro-cardiologic signals x (t) include normal electrocardiosignal and ripple containing j
Electrocardiosignal, find out all maximum points of each original electro-cardiologic signals x (t) and minimum point, be denoted as pi, 1≤i≤n；
Second step: with line segment by all of extreme point piIt is sequentially connected, the midpoint of every section of line segment is labeled as ti(1≤i≤n-
1), and connect two ends about the total segment being formed in line segment and replenish border midpoint t0And tn；
3rd step: build s article of interpolat curve l using n+1 midpoint1,…,ls, s >=1, calculate its meansigma methods l*=(l1
+…ls)/s；
4th step: to x (t)-l*Sequence repeats above three step, until | l*|≤ε or screening number of times reach default
Spacing value k, ε is allowable error, obtains first intrinsic mode function c1(t)；
5th step: to residue sequence x (t)-c1Repeat aforementioned four step, until residual term r is single signal or no longer
More than previously given extreme point, intrinsic mode function c just can be respectively obtained2(t),c3(t)…,ci(t),cmT (), calculates sequence
The variances sigma of row x (t)-r2；
6th step: limiting interval [kmin,kmax] spacing value k of interior change, repeat above-mentioned five steps, and to σ/σ0And k
Drawn, σ0It is the standard deviation of x (t), find out σ/σ in figure0The corresponding k of minima0, with k0Weigh again as restrictive condition
Multiple above-mentioned five steps, last residual term r is exactly the self adaptation overall situation all lines of primary signal x (t), that is, trend term, through undue
Solution, original electro-cardiologic signals x (t) can representUsing pole symmetric mode decomposition, electrocardio is believed
Number x (t) resolves into a series of intrinsic mode functions and a residual term；Compared to Hilbert-Huang transform, decompose electrocardiosignal
During, pole symmetric mode decomposition will be used in empirical mode decomposition determining three rank battens of primary signal outer envelope line
Interpolation method is improved to inherent pole Symmetric Interpolation method, and borrow " method of least square " thought come to optimize finally remaining mode so as to
Become " the self adaptation overall situation all lines " in whole signal sequence catabolic process, to determine the optimal screening time in catabolic process with this
Number.Also abandon Hilbert spectral analysis method, using direct differential technique, solved well: the cycle has needed with respect to one section
Time defines and frequency will have instantaneous meaning.
7th step: similar with Hilbert-Huang transform, pole symmetric mode decomposition is also classified into two parts: Part I be through
Test mode decomposition, several intrinsic mode functions and the optimal self-adaptive overall situation all line can be produced；Part II is time frequency analysis,
Relate generally to " direct interpolation " and the energy variation problem of instantaneous frequency.By entering to each intrinsic mode function of electrocardiosignal
Row time frequency analysis, extract instantaneous frequency feature and energy conversion feature.
(1) find intrinsic mode function ciT the extreme point of (), calculates the adjacent maximum point of any two and two consecutive roots
Time difference between little value point；
(2) by these time differences be considered as local period value be assigned to two adjacent maximum points and two adjacent minimum points it
Between midpoint on time shafts, drawing the time cycle corresponds to point diagram；
(3) obtain its local frequencies by inverted for these local period values, then do cubic spline interpolation obtaining instantaneous frequency
Curve；
8th step: because the frequency of each intrinsic mode function and amplitude are time dependent, so this method is put
Abandon using the tradition spectrum analysis means (Fourier spectrum and Hilbert spectrum) premised on constant by energy, energy variation
Feature is as object of study.The corresponding analytical expression of i-th intrinsic mode function is: ci(t)=ai(t)cosθi(t),(1≤
I≤m), then its amplitude function a under pole symmetriciT () has gradual feature,
9th step: choose instantaneous frequency f1, f2, f3, f4, f5, f6, f7 and the energy conversion of the first seven intrinsic mode function
Feature e (t) composition characteristic space [f1, f2, f3, f4, f5, f6, f7, e (t)], is entered to feature space from PCA
Row dimensionality reduction, chooses front 8 main constituents of each feature, and that is, the feature after dimensionality reduction amounts to 64；
Tenth step: features described above is input to support vector machine, support vector machine are trained, extract the heart to be detected
Feature after the dimensionality reduction of the signal of telecommunication is input to support vector machine, you can tell the type of electrocardiosignal to be detected.
PCA is a kind of statistical method of dimensionality reduction, and it is related by its component by means of an orthogonal transformation
Former random vector changes into the incoherent new random vector of its component, and this shows as the covariance of former random vector on algebraically
Battle array is transformed into diagonal form battle array, is geometrically showing as, by the orthogonal coordinate system of former coordinate system transformation Cheng Xin, being allowed to point to sample point
Spread the n orthogonal direction opened most, then dimension-reduction treatment is carried out to multidimensional variable system, make it to turn with a higher precision
Change low-dimensional variable system into, then by constructing suitable cost function, further low-dimensional system is changed into unidimensional system.
Classification and Detection module of the present invention selects support vector machine, Selection of kernel function Radial basis kernel function.Support vector machine side
Method is a kind of new method proposing in recent years.
The main thought of support vector machine may be summarized to be at 2 points:
(1) it is that linear can a point situation be analyzed, and during for linearly inseparable, reflects by using non-linear
Penetrate algorithm, this programme uses Radial basis kernel function, the sample of low-dimensional input space linearly inseparable is converted into high dimensional feature empty
Between make its linear separability, so that high-dimensional feature space carries out linear analysiss using linear algorithm to the nonlinear characteristic of sample
It is possibly realized, the present invention chooses Radial basis kernel function；
It is based on structural risk minimization theory in feature space construction optimum segmentation hyperplane so that study
Device obtains global optimization, and the expected risk in whole sample space meets certain upper bound with certain probability.
Claims (1)
1. the j wave detecting method based on pole symmetric mode decomposition and support vector machine is it is characterised in that comprise the following steps:
The first step: obtain original electro-cardiologic signals x (t), original electro-cardiologic signals x (t) include normal electrocardiosignal and the heart containing j ripple
The signal of telecommunication, finds out all maximum points of each original electro-cardiologic signals x (t) and minimum point, is denoted as pi, 1≤i≤n；
Second step: with line segment by all of extreme point piIt is sequentially connected, the midpoint of every section of line segment is labeled as ti, 1≤i≤n-1, and
Connect two ends about the total segment being formed in line segment and replenish border midpoint t0And tn；
3rd step: build s article of interpolat curve l using n+1 midpoint1,…,ls, s >=1, calculate its meansigma methods l*=(l1+…ls)/
s；
4th step: above three step is repeated to x (t)-l* sequence, until | l*|≤ε or screening number of times reach default spacing
Value k, ε is allowable error, obtains first intrinsic mode function c1(t)；
5th step: to residue sequence x (t)-c1Repeat aforementioned four step, until residual term r is single signal or is no larger than pre-
The extreme point first giving, just can respectively obtain intrinsic mode function c2(t),c3(t)…,ci(t),cm(t), sequence of calculation x (t)-
The variances sigma 2 of r；
6th step: limiting interval [kmin,kmax] spacing value k of interior change, repeat above-mentioned five steps, and to σ/σ0Painted with k
Figure, σ0It is the standard deviation of x (t), find out σ/σ in figure0The corresponding k of minima0, with k0Repeat above-mentioned again as restrictive condition
Five steps, last residual term r is exactly the self adaptation overall situation all lines of primary signal x (t), that is, trend term, through decomposing, original
Electrocardiosignal x (t) is represented byPole symmetric mode decomposition is utilized to divide electrocardiosignal x (t)
Solution becomes a series of intrinsic mode functions and a residual term；
7th step: (1) finds intrinsic mode function ciThe extreme point of (t), calculate two adjacent maximum points and two adjacent minimum
Time difference between value point；
(2) when these time differences being considered as local period value be assigned between two adjacent maximum points and two adjacent minimum points
Midpoint on countershaft, draws time cycle correspondence point diagram；
(3) obtain its local frequencies by inverted for these local period values, then do cubic spline interpolation obtaining instantaneous frequency profile；
8th step: the corresponding analytical expression of i-th intrinsic mode function is: ci(t)=ai(t)cosθi(t), 1≤i≤m,
9th step: choose instantaneous frequency f1, f2, f3, f4, f5, f6, f7 and the energy conversion characteristics of the first seven intrinsic mode function
E (t) composition characteristic space [f1, f2, f3, f4, f5, f6, f7, e (t)], is dropped to feature space from PCA
Dimension, chooses front 8 main constituents of each feature, and that is, the feature after dimensionality reduction amounts to 64；
Tenth step: features described above is input to support vector machine, support vector machine are trained, extract electrocardio letter to be detected
Number dimensionality reduction after feature be input to support vector machine, you can tell the type of electrocardiosignal to be detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610952984.2A CN106344006B (en) | 2016-11-03 | 2016-11-03 | J wave detecting methods based on pole symmetric mode decomposition and support vector machines |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610952984.2A CN106344006B (en) | 2016-11-03 | 2016-11-03 | J wave detecting methods based on pole symmetric mode decomposition and support vector machines |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106344006A true CN106344006A (en) | 2017-01-25 |
| CN106344006B CN106344006B (en) | 2018-06-26 |
Family
ID=57864277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610952984.2A Active CN106344006B (en) | 2016-11-03 | 2016-11-03 | J wave detecting methods based on pole symmetric mode decomposition and support vector machines |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106344006B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107233204A (en) * | 2017-07-12 | 2017-10-10 | 李城钰 | A kind of heart arrest first aid integrated apparatus |
| CN107729861A (en) * | 2017-10-30 | 2018-02-23 | 太原理工大学 | J wave detecting methods based on random forest |
| CN108615026A (en) * | 2018-05-09 | 2018-10-02 | 广东工业大学 | The discriminating gear and equipment of the malignant ventricular rhythm of the heart |
| CN108680782A (en) * | 2018-05-04 | 2018-10-19 | 上海电力学院 | The voltage flicker parameter detection method decomposed based on extreme point symmetric pattern |
| CN109933839A (en) * | 2019-01-11 | 2019-06-25 | 青岛理工大学 | A kind of integrated form pole symmetric mode decomposition method about the analysis of Space Time data |
| CN110558973A (en) * | 2019-09-06 | 2019-12-13 | 江苏华康信息技术有限公司 | Electrocardiogram signal quantitative analysis method based on extreme value energy decomposition method |
| CN110619384A (en) * | 2019-08-13 | 2019-12-27 | 浙江工业大学 | PM2.5 concentration value prediction method based on neural network |
| CN110956136A (en) * | 2019-12-02 | 2020-04-03 | 天津市计量监督检测科学研究院 | Hydrophone signal denoising method based on ESMD and roughness penalty smoothing technology |
| CN111177211A (en) * | 2019-12-11 | 2020-05-19 | 华北电力大学 | Runoff sequence change characteristic analysis method based on pole symmetric modal decomposition |
| CN111904417A (en) * | 2020-07-06 | 2020-11-10 | 天津大学 | Ultra-wideband microwave early breast cancer detection device based on support vector machine |
| WO2021042589A1 (en) * | 2019-09-06 | 2021-03-11 | 江苏华康信息技术有限公司 | Extremum energy decomposition-based electrocardiogram signal analysis method |
| CN113974647A (en) * | 2021-10-26 | 2022-01-28 | 福州大学 | System and method for reconstructing sudden cardiac death risk factor by multi-feature quantization and model parameter optimization of nonlinear support vector machine |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102302361A (en) * | 2011-06-24 | 2012-01-04 | 东北大学 | Device for monitoring heart rate and physical state |
| JP2014171677A (en) * | 2013-03-08 | 2014-09-22 | Fukuda Denshi Co Ltd | Electrocardiogram data processing device and controlling method thereof |
| CN104367315A (en) * | 2014-11-13 | 2015-02-25 | 太原理工大学 | Blind source separation method for extracting J wave signals in electrocardiogram |
| CN105266801A (en) * | 2015-09-10 | 2016-01-27 | 太原理工大学 | J wave detection and classification method based on probabilistic neural network (PNN) |
-
2016
- 2016-11-03 CN CN201610952984.2A patent/CN106344006B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102302361A (en) * | 2011-06-24 | 2012-01-04 | 东北大学 | Device for monitoring heart rate and physical state |
| JP2014171677A (en) * | 2013-03-08 | 2014-09-22 | Fukuda Denshi Co Ltd | Electrocardiogram data processing device and controlling method thereof |
| CN104367315A (en) * | 2014-11-13 | 2015-02-25 | 太原理工大学 | Blind source separation method for extracting J wave signals in electrocardiogram |
| CN105266801A (en) * | 2015-09-10 | 2016-01-27 | 太原理工大学 | J wave detection and classification method based on probabilistic neural network (PNN) |
Non-Patent Citations (1)
| Title |
|---|
| 刘学博: "基于支持向量机的J波诊断技术研究", 《中国优秀硕士学位论文全文数据库医药卫生科技辑》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107233204B (en) * | 2017-07-12 | 2019-08-06 | 李城钰 | A kind of sudden cardiac arrest first aid integrated apparatus |
| CN107233204A (en) * | 2017-07-12 | 2017-10-10 | 李城钰 | A kind of heart arrest first aid integrated apparatus |
| CN107729861A (en) * | 2017-10-30 | 2018-02-23 | 太原理工大学 | J wave detecting methods based on random forest |
| CN108680782B (en) * | 2018-05-04 | 2020-10-09 | 上海电力学院 | Voltage flicker parameter detection method based on extreme point symmetric mode decomposition |
| CN108680782A (en) * | 2018-05-04 | 2018-10-19 | 上海电力学院 | The voltage flicker parameter detection method decomposed based on extreme point symmetric pattern |
| CN108615026A (en) * | 2018-05-09 | 2018-10-02 | 广东工业大学 | The discriminating gear and equipment of the malignant ventricular rhythm of the heart |
| CN108615026B (en) * | 2018-05-09 | 2022-05-10 | 广东工业大学 | Device and equipment for judging malignant ventricular rhythm |
| CN109933839A (en) * | 2019-01-11 | 2019-06-25 | 青岛理工大学 | A kind of integrated form pole symmetric mode decomposition method about the analysis of Space Time data |
| CN109933839B (en) * | 2019-01-11 | 2023-04-14 | 青岛理工大学 | Integrated pole symmetric modal decomposition method for time-space data analysis |
| CN110619384A (en) * | 2019-08-13 | 2019-12-27 | 浙江工业大学 | PM2.5 concentration value prediction method based on neural network |
| WO2021042589A1 (en) * | 2019-09-06 | 2021-03-11 | 江苏华康信息技术有限公司 | Extremum energy decomposition-based electrocardiogram signal analysis method |
| CN110558973B (en) * | 2019-09-06 | 2022-02-18 | 江苏华康信息技术有限公司 | Computer equipment for executing electrocardiogram signal quantitative analysis method based on extreme value energy decomposition method |
| CN110558973A (en) * | 2019-09-06 | 2019-12-13 | 江苏华康信息技术有限公司 | Electrocardiogram signal quantitative analysis method based on extreme value energy decomposition method |
| CN110956136A (en) * | 2019-12-02 | 2020-04-03 | 天津市计量监督检测科学研究院 | Hydrophone signal denoising method based on ESMD and roughness penalty smoothing technology |
| CN111177211A (en) * | 2019-12-11 | 2020-05-19 | 华北电力大学 | Runoff sequence change characteristic analysis method based on pole symmetric modal decomposition |
| CN111904417A (en) * | 2020-07-06 | 2020-11-10 | 天津大学 | Ultra-wideband microwave early breast cancer detection device based on support vector machine |
| CN111904417B (en) * | 2020-07-06 | 2021-12-03 | 天津大学 | Ultra-wideband microwave early breast cancer detection device based on support vector machine |
| CN113974647A (en) * | 2021-10-26 | 2022-01-28 | 福州大学 | System and method for reconstructing sudden cardiac death risk factor by multi-feature quantization and model parameter optimization of nonlinear support vector machine |
| CN113974647B (en) * | 2021-10-26 | 2023-08-04 | 福州大学 | System and method for reconstructing sudden cardiac death risk factor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106344006B (en) | 2018-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106344006A (en) | J wave detection method based on pole symmetrical mode decomposition and support vector machine | |
| CN106108889B (en) | Electrocardiogram classification method based on deep learning algorithm | |
| Übeyli | Combining recurrent neural networks with eigenvector methods for classification of ECG beats | |
| Yang et al. | Spatiotemporal differentiation of myocardial infarctions | |
| Park et al. | Hierarchical support vector machine based heartbeat classification using higher order statistics and hermite basis function | |
| Costa et al. | Multiscale entropy to distinguish physiologic and synthetic RR time series | |
| CN109276241B (en) | Pressure identification method and equipment | |
| Javadi et al. | Classification of ECG arrhythmia by a modular neural network based on mixture of experts and negatively correlated learning | |
| Beritelli et al. | A novel training method to preserve generalization of RBPNN classifiers applied to ECG signals diagnosis | |
| CN107951485A (en) | Ambulatory ECG analysis method and apparatus based on artificial intelligence self study | |
| Anuradha et al. | CARDIAC ARRHYTHMIA CLASSIFICATION USING FUZZY CLASSIFIERS. | |
| CN107811626A (en) | A kind of arrhythmia classification method based on one-dimensional convolutional neural networks and S-transformation | |
| CN108056773A (en) | Based on the algorithms of QRS complexes detection in electrocardiogram signal for improving variation mode decomposition | |
| Zhang et al. | A multi-dimensional association information analysis approach to automated detection and localization of myocardial infarction | |
| CN109276242A (en) | The method and apparatus of electrocardiosignal type identification | |
| CN102626310A (en) | Electrocardiogram signal feature detection algorithm based on wavelet transformation lifting and approximate envelope improving | |
| CN106137185A (en) | A kind of epileptic chracter wave detecting method based on structure of transvers plate small echo | |
| CN105960643A (en) | Device for assessing autonomic nerve balancing and controlling ability, and method of controlling same | |
| Wan et al. | A neural network to identify human subjects with electrocardiogram signals | |
| CN103860167A (en) | J-wave extraction and analysis method based on non-negative matrix factorization | |
| CN107169424A (en) | Atrial fibrillation detection algorithm based on artificial neural network | |
| CN106137184A (en) | Electrocardiosignal QRS complex detection method based on wavelet transformation | |
| Billah et al. | A novel method to model ECG beats using Gaussian functions | |
| Garcia-Gonzalez et al. | Errors in the estimation of approximate entropy and other recurrence-plot-derived indices due to the finite resolution of RR time series | |
| Zhang et al. | Multi‐fractal detrended fluctuation half‐spectrum analysis of HRV |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |