CN107874755B - A kind of P wave extracting method based on rectangular projection and independent component analysis - Google Patents
A kind of P wave extracting method based on rectangular projection and independent component analysis Download PDFInfo
- Publication number
- CN107874755B CN107874755B CN201710816513.3A CN201710816513A CN107874755B CN 107874755 B CN107874755 B CN 107874755B CN 201710816513 A CN201710816513 A CN 201710816513A CN 107874755 B CN107874755 B CN 107874755B
- Authority
- CN
- China
- Prior art keywords
- wave
- signal
- electrocardiosignal
- qrs
- qrs complex
- 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
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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
Abstract
The P wave extracting method based on rectangular projection and independent component analysis that the invention discloses a kind of, it is used to be analyzed and processed electrocardiosignal, it includes carrying out Signal Pretreatment to the electrocardiosignal of acquisition, and QRS complex template is obtained by time domain average method, QRS complex ingredient is filtered out using the wave group template and by the method that orthogonal space projects, to improve the signal-to-noise ratio of P wave and T wave in remaining ingredient, separate the P wave and T wave, then using the P wave after separating as reference signal, using the method for half-blindness, P wave is extracted from original aliasing signal again.
Description
Technical field
The present invention relates to medical instruments fields, in particular to a kind of based on rectangular projection and isolated component point
Analyse P wave extracting method.
Background technique
The analysis of electrocardiosignal is widely used in clinical medicine, and the detection of its characteristic parameter is for cardiovascular disease
Clinics and Practices play key effect.QRS complex and P, T wave are the important components of electrocardiosignal, they reflect the heart
The different conditions of dirty movement.Electrocardiosignal characteristic parameter detection mainly in electrocardiosignal QRS complex, P wave, T wave and
The detection of the major parameters such as ST sections and identification wherein QRS complex be electrocardiosignal a main feature, be ECG signal sampling
In matter of utmost importance, same P wave and T wave also have great significance in the diagnosis about heart disease.P wave reflects left and right
Potential change in the process of depolarization in atrium.Most P wave waveform round blunts are smooth, last 0.05~0.10 second, and amplitude is
0.05~0.25mV is atrium conduction more reliably noninvasive index.Inhomogeneity and discontinuity in the broadening reflection atrium of P wave
Conduction is the important indicator of detection of occult hypertensive patient heart injury and prediction coronary heart disease atrial fibrillation risk.But P wave is weak
It is small, there is variform, and position is not fixed, P wave may be with T wave aliasing, when Ventricular Tachycardia, P when premature atrial
Wave may be covered by QRS wave.Therefore in order to correctly make clinical diagnosis, accurately and effectively extracting P wave signal seems necessary
And value.
In recent decades, have benefited from the development of signal processing technology, the extraction of P wave there are many theoretical results, such as poor
Threshold method, wavelet transformation, the P wave based on steady section identification is divided to detect and calculate, these all rely on greatly search window, inappropriate search
Window can cause P wave, stop position inaccurate, so having some limitations, and because P wave is small and weak, form is changeable, implements
Get up it is relatively difficult with it is inconvenient.
Summary of the invention
For overcome the deficiencies in the prior art, the invention proposes a kind of P based on rectangular projection and independent component analysis
Wave extracting method, this method effective solution electrocardiosignal P wave are difficult to the problem of extracting, and preferably serve clinical electrocardio prison
Shield.
The invention is achieved through the following technical solutions: a kind of P wave extraction side based on rectangular projection and independent component analysis
Method includes the following steps:
S1: Signal Pretreatment: pre-processing the multichannel electrocardiosignal of acquisition, and it is dry mainly to remove various noises, myoelectricity
Disturb, Hz noise and correction baseline drift.
S2: the detection of R wave is carried out to signal using threshold detection method, the multiple QRS complexes repeated is positioned, then passes through
Time domain average method obtains QRS complex template.
S3: to treated signal, filtering out QRS complex ingredient by the method that orthogonal space projects, to improve remaining
The signal-to-noise ratio of P wave and T wave in ingredient.
S4: the potential change in the process of depolarization of left atrium is reflected in view of P wave, T wave corresponds to the multipole mistake of ventricle
Journey, the two have certain independence, using independent component analysis, separate P wave and T wave.
S5: the detection of leakage value, when P wave and when QRS complex aliasing, QRS complex template that the method for average acquires can containing P wave at
Point, cause during filtering out for QRS groups, P wave can lose, and isolated component P wave component obtained after separation is simultaneously imperfect.Institute
With the P wave after separating as reference signal, using the method for half-blindness, P wave is extracted from original aliasing signal again, to realize
The purpose of P wave enhancing.
Detailed description of the invention
Fig. 1 is of the invention a kind of based on rectangular projection and independent component analysis P wave extracting method flow chart
Fig. 2 is the three tunnel source electrocardiosignals for being collected in abdomen
Fig. 3 is signal after pre-processing in the present invention
Fig. 4 is the QRS template that time domain average method of the present invention obtains
Fig. 5 is the signal after present invention removal QRS wave
Fig. 6 is the isolated P wave signal of isolated component of the present invention
Specific embodiment
Present invention will now be described in further detail with reference to the embodiments and the accompanying drawings, but embodiments of the present invention are unlimited
In this.
As shown in Figure 1, including the following steps: the present invention is based on rectangular projection and independent component analysis P wave extracting method
(1) to the three road electrocardiosignal S collected1、S2、S3Carry out comb filtering removal 0Hz, 50Hz Hz noise, with
And higher hamonic wave interference.Median filtering removes baseline drift.
(2) to treated, signal carries out threshold test, and threshold value is defined as follows:
Wherein p is threshold value, and μ takes 0.3 here,It is the average value of all extreme points.
(3) since the QRS complex time of normal person is 0.06~0.08 second, and the sample frequency of signal is fs=
1000Hz, so taking template length is 70.Remember that all QRS waves are respectively (1) x,
X (2), x (3) ..., x (n), then the template signal of QRS is as follows:
(4) it according to signal S1, the R wave position of S2, S3 and the template signal template of generation, is projected using orthogonal space
Method filters out QRS wave and obtains X1, X2, X3.
(5) independent component analysis is carried out to remaining signal, is mainly realized using FastICA algorithm, steps are as follows:
1) centralization is carried out to data and removes averaging operation, computation complexity can be effectively reduced, if input
Signal is that signal is x, output signal y.
Y=x-E { x }
2) albefaction is then carried out, z is obtained, meets E { zzT}=I
3) select the initialization with unit norm that vector w (can be randomly selected)
4) w is updated:
w←E{zg(wTz)}-E{g(wTz)}w
5) w is standardized:
w←w/||w||
6) judge whether to meet convergence | w (k+1)-w (k) | otherwise < ε, is completed, table if not yet convergence returns to step 4
Show Signal separator.
7) related coefficient is calculated, selects related coefficient maximum all the way as P wave signal.
(6) if there is leakage value situation, when being P wave and QRS complex aliasing, when filtering out QRS wave, P wave can be together
It filters out.P wave is separated from the signal of aliasing to mainly comprise the steps that
1) the starting place value of p wave is positioned.
If n is the cardiac electrical cycle of a p wave, m is the cardiac electrical cycle of a QRS, and the signal of overlapping position isWith separation
Obtained P wave is that reference signal calculates mean valueThe template signal of QRS isBy signal 5 'P 1×n
It is extended toMake itself and aliasing signalLength is identical,It indicates
In S 'P 1×nI 0 elements are added before signal,It indicates 5 '6 1×n(m-i-n) a 0 element is added behind signal,
To formation and mixed signalThe extension signal of equal length
By the extension signal of p wave:With the template signal of qrs:It is overlapped, obtains the template of a mixing
SignalSeek signalWith overlapped signalBetween error.Using European
Distance is used as objective function:
The as starting point of overlapping positions when reaching minimum.
2) separation matrix of P wave signal is sought from aliasing signal
It is calculated by step 1) and knows that the template signal of p wave is S 'P 1×n, extending signal isAnd the template letter being superimposed
Number it isAvailable following expression:
NoteForFor VP, then have
Because of AmIt is sequency spectrum matrix for column vector, so being acquired by descriptor matrix:
(Am)-1=(AHA)-1AH
It is available:The signal p of overlapping can be acquired according to separation matrix.
Technical solution provided by the present invention is a preferable implementation of the invention, but embodiments of the present invention
It is not limited thereto, other any changes made without departing from the spirit and principles of the present invention, modification, substitution, group
It closes, simplify, should be equivalent substitute mode, be included within the scope of the present invention.
Claims (1)
1. a kind of P wave extracting method based on rectangular projection and independent component analysis, is used to carry out at analysis electrocardiosignal
Reason, which comprises the steps of:
S1: carrying out Signal Pretreatment to the electrocardiosignal of acquisition, wherein the pretreatment includes the multichannel electrocardiosignal to acquisition
It is pre-processed, removes noise, myoelectricity interference, Hz noise and correction baseline drift therein；
S2: the detection of R wave is carried out to the electrocardiosignal using threshold detection method, positions the multiple QRS complexes repeated, then
QRS complex template is obtained by time domain average method；
S3: to pretreated signal, using the wave group template and by the method that orthogonal space projects filter out QRS complex at
Point, to improve the signal-to-noise ratio of P wave and T wave in remaining ingredient；
S4: the P wave and T wave are separated；
S5: the detection of leakage value, when P wave and QRS complex aliasing, the QRS complex template that the method for average acquires can contain P wave component, lead
It causes during filtering out for QRS groups, P wave can lose, and isolated component P wave component obtained after separation is simultaneously imperfect；So with
P wave after separation is as reference signal, using the method for half-blindness, extracts P wave from original aliasing signal again.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710816513.3A CN107874755B (en) | 2017-09-12 | 2017-09-12 | A kind of P wave extracting method based on rectangular projection and independent component analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710816513.3A CN107874755B (en) | 2017-09-12 | 2017-09-12 | A kind of P wave extracting method based on rectangular projection and independent component analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107874755A CN107874755A (en) | 2018-04-06 |
| CN107874755B true CN107874755B (en) | 2019-09-03 |
Family
ID=61780614
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710816513.3A Active CN107874755B (en) | 2017-09-12 | 2017-09-12 | A kind of P wave extracting method based on rectangular projection and independent component analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107874755B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101088456A (en) * | 2007-07-13 | 2007-12-19 | 天津大学 | Cardioelectric characteristic extracting process based on evolutive wavelet wiener deconvolution |
| CN101500480A (en) * | 2006-08-04 | 2009-08-05 | 詹姆麦德有限公司 | Method for processing cardioelectric signals and corresponding device |
| CN101766483A (en) * | 2008-12-26 | 2010-07-07 | 河南华南医电科技有限公司 | Novel T wave subtraction algorithm |
-
2017
- 2017-09-12 CN CN201710816513.3A patent/CN107874755B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101500480A (en) * | 2006-08-04 | 2009-08-05 | 詹姆麦德有限公司 | Method for processing cardioelectric signals and corresponding device |
| CN101088456A (en) * | 2007-07-13 | 2007-12-19 | 天津大学 | Cardioelectric characteristic extracting process based on evolutive wavelet wiener deconvolution |
| CN101766483A (en) * | 2008-12-26 | 2010-07-07 | 河南华南医电科技有限公司 | Novel T wave subtraction algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 胎儿心电信号的盲分离研究;蔡坤;《中国博士学位论文全文数据库 信息科技辑》;20120615(第2012年6期);摘要 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107874755A (en) | 2018-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109907752B (en) | Electrocardiogram diagnosis and monitoring system for removing motion artifact interference and electrocardio characteristic detection | |
| Sathyapriya et al. | Analysis and detection R-peak detection using Modified Pan-Tompkins algorithm | |
| CN106108889A (en) | Electrocardiogram classification method based on degree of depth learning algorithm | |
| Chanwimalueang et al. | Enabling R-peak detection in wearable ECG: Combining matched filtering and Hilbert transform | |
| Remya et al. | Classification of myocardial infarction using multi resolution wavelet analysis of ECG | |
| CN104783787B (en) | A kind of J wave detecting methods based on neutral net | |
| CN104161510A (en) | Multistage lead electrocardiograph signal QRS waveform identification method | |
| CN101919704B (en) | Heart sound signal positioning and segmenting method | |
| Balachandran et al. | Daubechies algorithm for highly accurate ECG feature extraction | |
| Malek et al. | Automated detection of premature ventricular contraction in ECG signals using enhanced template matching algorithm | |
| Pangerc et al. | Robust detection of heart beats in multimodal records using slope-and peak-sensitive band-pass filters | |
| CN105266800A (en) | Fetal electrocardiogram blind separation method based on low signal-to-noise ratio | |
| Talatov et al. | Algorithmic and software analysis and processing of ECG signals | |
| Luengo et al. | Blind analysis of atrial fibrillation electrograms: a sparsity-aware formulation | |
| Liu et al. | A novel P-QRS-T wave localization method in ECG signals based on hybrid neural networks | |
| CN103876731B (en) | A kind of Fetal ECG signal extracting device and method | |
| CN107874755B (en) | A kind of P wave extracting method based on rectangular projection and independent component analysis | |
| Vuksanovic et al. | Analysis of human electrocardiogram for biometric recognition using analytic and ar modeling extracted parameters | |
| CN109512405A (en) | A kind of pulse signal dividing method based on partial points detection | |
| CN111166325B (en) | Electrocardiosignal QRS complex wave detection method and system based on IPCMM algorithm | |
| Zhong et al. | Maternal ECG removal using short time Fourier transform and convolutional auto-encoder | |
| Chen et al. | Multi-window detection for P-wave in electrocardiograms based on bilateral accumulative area | |
| Khandait et al. | ECG signal processing using classifier to analyses cardiovascular disease | |
| Kaya et al. | Wavelet-based analysis method for heart rate detection of ecg signal using labview | |
| Adib et al. | ECG beat classification using discrete wavelet coefficients |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Xie Shengli Inventor after: Huang Chencuan Inventor after: Cai Kun Inventor after: Wang Xu Inventor after: Zhang Yuchang Inventor before: Huang Chencuan Inventor before: Xie Shengli Inventor before: Cai Kun Inventor before: Wang Xu Inventor before: Zhang Yuchang |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |