CN105011931A - Method for detecting wavy boundary of electrocardiogram - Google Patents
Method for detecting wavy boundary of electrocardiogram Download PDFInfo
- Publication number
- CN105011931A CN105011931A CN201410174789.2A CN201410174789A CN105011931A CN 105011931 A CN105011931 A CN 105011931A CN 201410174789 A CN201410174789 A CN 201410174789A CN 105011931 A CN105011931 A CN 105011931A
- Authority
- CN
- China
- Prior art keywords
- point
- electrocardiogram
- boundary point
- boundary
- ripple
- 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
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention provides a method for detecting the wavy boundary of an electrocardiogram. The method comprises the following steps: (1) performing positive and negative sequence filtration preprocessing on the electrocardiogram by using a band-pass filter; (2) transforming the filtered electrocardiogram through a rain flow model; (3) detecting boundary points of the transformed electrocardiogram; and (4) correcting the deviation caused by interference or waveform diversity. According to the method, the positive and negative sequence filtration is performed on the electrocardiogram to strengthen the Gibbs effect, and the rain flow model transforms the electrocardiogram to assist in detecting the boundary of the electrocardiogram.
Description
Technical field
The present invention relates to image processing techniques, particularly relate to a kind of method of electrocardiographic wave border detection.
Background technology
Electrocardioscopy is a kind of effective method of diagnose arrhythmia, myocardial ischemia, and the method has the advantage of hurtless measure, low cost, has larger portfolio in hospital.Especially in mechanisms such as MEC, remote medical consultation with specialists centers, full-time electrocardiogram doctor every day need a large amount of Electrocardiographic of interpretation, for alleviating the work load of doctor, computer assisted electrocardiogram automatic classification recognition system more and more comes into one's own in recent years.
The general characteristics of the P-QRS-T wave group in electrocardiogram and the interval feature of derivation are indirectly the foundations of diagnosis.Peak point for various ripple is important feature, but boundary point is also very important information, therefore, how carry out Electrocardiographic wavy boundary by armarium auxiliary and be detected as one of trend in order to study.
Summary of the invention
In view of this, we provide a kind of method of electrocardiographic wave border detection, can locate fast and accurately and auxiliary detection peak point.
The method of electrocardiographic wave border detection of the present invention, comprises the steps: that (1) uses band filter to carry out positive-negative sequence filter preprocessing to electrocardiogram; (2) by rain flow model, filtered electrocardiogram is converted, obtain the boundary point of crest, trough; (3) to described boundary points detection; (4) deviation that interference or waveform multiformity cause is corrected.
Preferably, in step (1), the band passband section of described band filter is 1 ~ 20Hz, and Ripple is 0.5dB.
Preferably, in step (2), described rain flow model, be the dot chart of a width sinusoidal sequence, be the model of the situation of rainwater flowing after raining in simulation hillside, rainwater drops on hillside and can flow to lower along hillside, and then a certain low-lying place converges accumulation, forms trough boundary point.
Preferably, in step (2), after being inverted negate, try to achieve Electrocardiographic crest boundary point.
Preferably, in step (4), the step of correction comprises, and judges that whether the position of described boundary point has with real border point and departs from.
Preferably, in step (4), concrete steps comprise: amplitude and this boundary point difference of choosing down any are reference, are assumed to be diff; Continue mobile backward, until the amplitude of certain point and the difference in magnitude of this boundary point are greater than 3*diff; Adjust to this position as the boundary point after correction.
The present invention, by carrying out positive-negative sequence filtering to electrocardiogram, is strengthened Gibbs' effect, and is converted by rain flow model, come the Electrocardiographic border of auxiliary detection.
Accompanying drawing explanation
Fig. 1 is the method flow diagram of electrocardiographic wave border detection of the present invention.
Fig. 2 is the exemplary plot of Gibbs' effect.
Fig. 3 is rain flow model exemplary plot.
Fig. 4 to Fig. 6 is the schematic diagram of the embodiment checking in the present invention.
Detailed description of the invention
In order to make object of the present invention, technical scheme and advantage more clear, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
Refer to Fig. 1, be depicted as the method flow diagram of electrocardiographic wave border detection of the present invention.
In step S101, band filter is used to carry out positive-negative sequence filter preprocessing to electrocardiogram.
The band passband section of band filter chooses 1 ~ 20Hz, passband ripple 0.5dB.Herein, the effect of band filter, one is filter to remove baseline drift and high-frequency noise, and two is the Gibbs' effects after will utilizing bandpass filtering.3 ~ 40Hz is about because QRS wave frequency is higher, P, T ripple is about 0.7 ~ 10Hz, therefore this band filter can produce Gibbs' effect to P ripple, QRS ripple and T ripple border, and set 0.5dB passband ripple, make Gibbs' effect obvious and minimum on original waveform impact at boundary.Further, bandpass filtering adopts positive-negative sequence twice filtering, not only can offset the phase shift that filtering produces, and the waveform end point Gibbs' effect of positive sequence filtering is obvious, and after inverted sequence, the Gibbs' effect of the starting point of waveform is also obvious.Finally the data point reuse of positive-negative sequence twice filtering to normal sequence.
Refer to Fig. 2, be depicted as the exemplary plot of Gibbs' effect.Gibbs' effect, after the periodic function (as rectangular pulse) with discontinuity point is carried out fourier progression expanding method, chooses finite term and synthesize.When the item number chosen is more, the peak occurred in synthesized waveform plays the discontinuity point the closer to original signal.When the item number chosen is very large, this peak plays value and is tending towards a constant, approximates greatly 9% of total hop value.This phenomenon is called Gibbs phenomenon.
And Gibbs' effect is obvious after use band filter.Because the boundary of electrocardiographic wave is a trip point relative to peak point, thus under Gibbs' effect, there will be a little fluctuation, and this is just being exaggerated the place, position on border, create a little extreme point (crest or trough), thus be beneficial to the location of lower surface model opposite side circle.
In step s 102, by rain flow model, filtered electrocardiogram is converted, obtain the boundary point of crest, trough.
Described rain flow model, refer to the model of the situation of rainwater flowing after raining in simulation hillside, it is equally distributed that prerequisite comprises rainwater, and rainwater can not lose in flowing, under the rule that Water seeks its own level, rainwater drops on hillside and can flow to lower along hillside, then converges accumulation at a certain low-lying place, forms trough; And after being inverted negate, try to achieve crest.
Refer to Fig. 3, be depicted as rain flow model exemplary plot.Rain flow model is the dot chart of a width sinusoidal sequence, and suppose that this dot chart is a mountain, and A point is peak point, B point is the lowest point point.Suppose that in the sky being about to is rained, as ' o ' in figure, and raindrop is uniform, and namely each point of sinusoidal sequence can the rain of equal number below.When rain drops on this seat, suppose that hillside can not sponge rain, so will diffluence downwards, and constantly converge, such as, when reaching the lowest point, such as B point, then rainfall can converge at this.In addition, when having projection at certain point of summit and the lowest point, can consider the inertia of rain stream, if the width of this projection is less than certain threshold value TH, (due to the smoothing effect of filtering, general burr type projection all can be filtered, so, the hillside having obvious waveform generally there will not be projection, therefore this threshold value TH=2), just can continue toward dirty, otherwise the raindrop of accumulation above just converges to this place.
Use rain flow model above to process to filtered data, discovery raindrop is converged in a series of the lowest point place.And the corresponding point of this series of the lowest point can accumulate certain rainfall (raindrop number), we are set as Srain.This trough raindrop Number Sequence is for identifying that the border of forward crest is more effective.Such as, suppose that the form of QRS wave group is Rs type, then because R ripple is the crest of forward, so, from R ripple position, the non-zero points (rainfall convergent point) of first Srain forward, is namely the starting point of R ripple.
Conveniently the boundary point of waveform is inverted in location, filtered data is carried out negate simultaneously, and also carries out the process of rain flow model, obtain trough raindrop Number Sequence SrainR.The same with the boundary alignment of forward waveform, inverted waveform just can be positioned by SrainR as the border of S ripple etc.
Certainly, the detection for peak point also has certain effect, because the extreme point of rainfall convergent point inherently certain waveform.
In step s 103, boundary points detection is carried out.
As, suppose that the form of QRS wave group is Rs type, then because R ripple is the crest of forward, so, from R ripple position, the non-zero points (rainfall convergent point) of first Srain forward, is namely the starting point of R ripple.For another example the T ripple of forward, generally also can produce a convergent point at its boundary point.
Conveniently the boundary point of waveform is inverted in location, filtered data is carried out negate simultaneously, and also carries out the process of rain flow model, obtain trough raindrop Number Sequence SrainR.The same with the boundary alignment of forward waveform, inverted waveform just can be positioned by SrainR as the border of Q ripple etc.
In addition, the scope generally due to T ripple is wider, and the rainfall pointing out accumulation on the border of T ripple is general namely maximum, also can confirm separately the position of T ripple, certainly, by known R crest value point, can more accurately and rapidly identify with this.Such as, after R ripple position, open a window, the scope of T ripple will be included.Then search the maximum of Srain and SrainR in this window, and choose that both are maximum, then the peak value of T ripple generally just determines, and T ripple is forward or inversion also determines.Such as, maximum is in SrainR, then T ripple is forward, then with this forward seek first non-zero points in Srain, is then T ripple starting point; Searching first non-zero points backward, is then T ripple terminal.
This method also can detect peak point.The peak value of the T ripple of such as forward, we just can use the rainfall convergent point of SrainR to detect; And for the crest of inverted T ripple, we just can use the rainfall convergent point of Srain to detect.By that analogy, other crest uses or complementaryly to detect.
In step S104, the deviation that interference or waveform multiformity cause is corrected.
It may be a class value of peak point by the 3rd step, because waveform is various and noise reason, the P ripple that particularly range of waveforms and amplitude are all smaller, can not ensure to be exactly necessarily the boundary point that we want, therefore final step exactly to may be above boundary point carry out correction confirm.
Correction principle is: judge that whether the position of described boundary point has with real border point and depart from.
Concrete grammar comprises: (1) chooses down any amplitude and this boundary point difference is reference, is assumed to be diff; Continue mobile backward, until the amplitude of certain point and the difference in magnitude of this boundary point are greater than 3*diff; Adjust to this position as the boundary point after correction.
Or (2) use other clinical experiences to correct, as whether range estimation has boundary point substantial deviation.
Experimental verification embodiment 1
Refer to Fig. 4, the arrhythmia data base of the ARR data base of the research provided for Massachusetts Institute Technology (MIT-BIH), the effectiveness of checking this method.Below to choose the 1900-2300 sampled point of the first lead signals of No. 101 records, application this method is carried out detecting its effect and is respectively Fig. 4.Figure mid point horizontal line is primary signal, and solid line is added some points as data after bandpass filtering, and solid line is the data of filtered data after the process of rain flow model and Srain, and dotted line is data after filtered data negate again after the process of rain flow model and SrainR.As can be seen from the figure, there is rainfall convergent point at the boundary point of waveform and peak point place.For the waveform of forward, such as P ripple, R ripple and T ripple, now with reference to solid line Srain, can find out, their border is just at rainfall convergent point place.For another example their peak point, now with reference to dotted line SrainR, can determine that peak point is exactly rainfall convergence place.And can find out, the cumulative precipitation of wider T ripple is maximum.
Embodiment 2
Refer to Fig. 5, be depicted as Chinese cardiovascular disease data base (Chinese Cardiovascular DiseaseDatabase, CCDD) No. 2 records: the partial data that the V2 of record leads, use the design sketch after the method.Originally the turning point of S ripple and T ripple is obvious, but respective border is fuzzyyer, and after this method, the end point of S ripple has had an obvious forward extreme point, and similarly, an obvious negative sense extreme point has also appearred in the starting point of T ripple.For T ripple, in figure, left triangle represents starting point, and equilateral triangle represents peak point, and right triangle represents end point.
Embodiment 3
Refer to Fig. 6, the partial data that the aVR being depicted as CCDD data base No. 9 record leads uses the design sketch after the method.Located P and involve border and T involves border.This example is inverted P ripple and negative T wave, and in figure, left triangle represents starting point, and equilateral triangle represents peak point, and right triangle represents end point.
Beneficial effect
(1) can detect P-QRS-T border, because of the generally easy identification of QRS wave group peak value, its border re-uses the method, more fast with accurate.In addition, wide T ripple is the most obvious to the method effect.
(2) certain auxiliaring effect is also had to the peakvalue's checking of waveform.
(3) by positive-negative sequence bandpass filtering, not only can offset the phase shift that filtering produces, and the waveform end point Gibbs' effect of positive sequence filtering is obvious, after inverted sequence, the Gibbs' effect of the starting point of waveform is also obvious.Finally the data point reuse of positive-negative sequence twice filtering to normal sequence.
(4) other documents are all eliminate Gibbs' effect as far as possible, and in order to prevent the leakage effect of wave filter, and this patent utilizes Gibbs' effect just, as the method for border detection or peakvalue's checking.
(5) generated the boundary point of crest, trough by rain flow model, and detect by being inverted negate.
The above is only the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (6)
1. a method for electrocardiographic wave border detection, is characterized in that, comprises the steps:
(1) band filter is used to carry out positive-negative sequence filter preprocessing to electrocardiogram;
(2) by rain flow model, filtered electrocardiogram is converted, obtain the boundary point of crest, trough;
(3) to described boundary points detection;
(4) deviation that interference or waveform multiformity cause is corrected.
2. the method for electrocardiographic wave border detection as claimed in claim 1, it is characterized in that, in step (1), the band passband section of described band filter is 1 ~ 20Hz, and Ripple is 0.5dB.
3. the method for electrocardiographic wave border detection as claimed in claim 1, it is characterized in that, in step (2), described rain flow model, be the dot chart of a width sinusoidal sequence, be the model of the situation of rainwater flowing after raining in simulation hillside, rainwater drops on hillside and can flow to lower along hillside, then converge accumulation at a certain low-lying place, form trough boundary point.
4. the method for electrocardiographic wave border detection as claimed in claim 3, is characterized in that, in step (2), after being inverted negate, tries to achieve Electrocardiographic crest boundary point.
5. the method for electrocardiographic wave border detection as claimed in claim 1, it is characterized in that, in step (4), the step of correction comprises, and judges that whether the position of described boundary point has with real border point and departs from.
6. the method for electrocardiographic wave border detection as claimed in claim 5, it is characterized in that, in step (4), concrete steps comprise:
Amplitude and this boundary point difference of choosing down any are reference, are assumed to be diff;
Continue mobile backward, until the amplitude of certain point and the difference in magnitude of this boundary point are greater than 3*diff;
Adjust to this position as the boundary point after correction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410174789.2A CN105011931B (en) | 2014-04-28 | 2014-04-28 | A kind of method of electrocardiographic wave border detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410174789.2A CN105011931B (en) | 2014-04-28 | 2014-04-28 | A kind of method of electrocardiographic wave border detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105011931A true CN105011931A (en) | 2015-11-04 |
| CN105011931B CN105011931B (en) | 2018-01-09 |
Family
ID=54402509
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410174789.2A Active CN105011931B (en) | 2014-04-28 | 2014-04-28 | A kind of method of electrocardiographic wave border detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105011931B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000004824A1 (en) * | 1998-07-23 | 2000-02-03 | Cardio Technologies, Inc. | Digital ecg detection system |
| WO2003005879A2 (en) * | 2001-07-13 | 2003-01-23 | Cardiac Science, Inc. | Method and apparatus for monitoring cardiac patients for t-wave alternans |
| CN101766484A (en) * | 2010-01-18 | 2010-07-07 | 董军 | Method and equipment for identification and classification of electrocardiogram |
| KR20100124409A (en) * | 2009-05-19 | 2010-11-29 | 동서대학교산학협력단 | Measuring system of pulse transit time and method thereof |
| US20110105923A1 (en) * | 2009-10-30 | 2011-05-05 | Bliss David E | Communications link health monitoring |
| CN102188240A (en) * | 2010-03-05 | 2011-09-21 | 华东师范大学 | Electrocardiographic data sampling method and device |
| CN102379694A (en) * | 2011-10-12 | 2012-03-21 | 中国科学院苏州纳米技术与纳米仿生研究所 | Electrocardiogram R wave detection method |
| CN103110417A (en) * | 2013-02-28 | 2013-05-22 | 华东师范大学 | Automatic electrocardiogram recognition system |
| US20130253356A1 (en) * | 2001-11-21 | 2013-09-26 | Cameron Health, Inc. | Apparatus and method for identifying atrial arrhythmia by far-field sensing |
| CN103417209A (en) * | 2013-08-29 | 2013-12-04 | 中国科学院苏州纳米技术与纳米仿生研究所 | Electrocardiogram characteristic selecting method |
| KR101366101B1 (en) * | 2012-12-31 | 2014-02-26 | 부산대학교 산학협력단 | System and method for classificating normal signal of personalized ecg |
-
2014
- 2014-04-28 CN CN201410174789.2A patent/CN105011931B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000004824A1 (en) * | 1998-07-23 | 2000-02-03 | Cardio Technologies, Inc. | Digital ecg detection system |
| WO2003005879A2 (en) * | 2001-07-13 | 2003-01-23 | Cardiac Science, Inc. | Method and apparatus for monitoring cardiac patients for t-wave alternans |
| US20130253356A1 (en) * | 2001-11-21 | 2013-09-26 | Cameron Health, Inc. | Apparatus and method for identifying atrial arrhythmia by far-field sensing |
| KR20100124409A (en) * | 2009-05-19 | 2010-11-29 | 동서대학교산학협력단 | Measuring system of pulse transit time and method thereof |
| US20110105923A1 (en) * | 2009-10-30 | 2011-05-05 | Bliss David E | Communications link health monitoring |
| CN101766484A (en) * | 2010-01-18 | 2010-07-07 | 董军 | Method and equipment for identification and classification of electrocardiogram |
| CN102188240A (en) * | 2010-03-05 | 2011-09-21 | 华东师范大学 | Electrocardiographic data sampling method and device |
| CN102379694A (en) * | 2011-10-12 | 2012-03-21 | 中国科学院苏州纳米技术与纳米仿生研究所 | Electrocardiogram R wave detection method |
| KR101366101B1 (en) * | 2012-12-31 | 2014-02-26 | 부산대학교 산학협력단 | System and method for classificating normal signal of personalized ecg |
| CN103110417A (en) * | 2013-02-28 | 2013-05-22 | 华东师范大学 | Automatic electrocardiogram recognition system |
| CN103417209A (en) * | 2013-08-29 | 2013-12-04 | 中国科学院苏州纳米技术与纳米仿生研究所 | Electrocardiogram characteristic selecting method |
Non-Patent Citations (9)
| Title |
|---|
| DONG JUNE ET AL: "Experiences-based Intelligence Simulation in ECG Recognition", 《INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE FOR MODELLING CONTROL & AUTOMATION》 * |
| IK DASKALOV ET AL: "Electrocardiogram signal preprocessing for automatic detection of QRS boundaries", 《ELECTROCARDIOGRAM SIGNAL PREPROCESSING FOR AUTOMATIC DETECTION OF QRS BOUNDARIES》 * |
| MA ARAFAT ET AL: "Automatic detection of ECG wave boundaries using empirical mode decomposition", 《IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS》 * |
| P LAGUNA ET AL: "Automatic detection of wave boundaries in multilead ECG signals:validation with the CSE database", 《COMPUTERS & BIOMEDICAL RESEARCH》 * |
| 朱侃杰: "心电图特征参数获取技术及其应用", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
| 王丽苹: "心电图模式分类方法研究进展及分析", 《中国生物医学工程学报》 * |
| 罗小刚等: "ECG信号小波变换与峰谷检测算法的研究", 《北京生物医学工程》 * |
| 胡晓等: "用三角形检测QRS复合波起止点", 《应用科学学报》 * |
| 黄进文等: "一种基于LabVIEW8.2提取ECG特征点的新方法", 《虚拟仪器技术》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105011931B (en) | 2018-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103169456A (en) | Processing method and processing system for pulse wave signals | |
| CN102379694B (en) | Electrocardiogram R wave detection method | |
| CN103584854B (en) | Extraction method of electrocardiosignal R waves | |
| CN105054926A (en) | Electrocardiosignal feature information extraction method and device | |
| CN103705234B (en) | Demodulation method in ambulatory ecg signal data and device | |
| CN106462957A (en) | Method and system for removing stripe noise in infrared image | |
| JP2015531643A5 (en) | ||
| TW201505606A (en) | Electrocardiography signal extraction method | |
| KR20190019668A (en) | An apparatus for detecting r-peak in electrocardiogram signal by using wavelet transform and modified shannon energy envelope and the method thereof | |
| CN112190268A (en) | Physiological signal processing device | |
| CN105030233A (en) | Method for recognizing ST segment of electrocardiosignal | |
| CN103927048B (en) | Touch-control scan method, touch-control display panel and touch control display apparatus | |
| CN112244835A (en) | Signal processing method and related device | |
| KR102007011B1 (en) | Device and method for detecting electrocardiogram signal from important point | |
| US9547891B2 (en) | Image processing method, system using the same, and recording medium for performing the same | |
| US20070232947A1 (en) | Method of extracting low-frequency or high-frequency component from a signal with slope tracing waves | |
| CN103083011A (en) | Electrocardio R wave crest real-time locating method assisted by chest impedance second order difference image | |
| CN109425897B (en) | Method and system for eliminating seismic data outlier interference | |
| CN105011931A (en) | Method for detecting wavy boundary of electrocardiogram | |
| CN106419899A (en) | Method and device for quantum resonance spectrometer (QRS) wave group extraction | |
| KR102155206B1 (en) | R wave peak detection method using periodicity of ECG signal | |
| KR102210468B1 (en) | Method and computer program for removing baseline fluctuation noise of ECG signal | |
| Rankawat et al. | ECG artifacts detection in noncardiovascular signals using Slope Sum Function and Teager Kaiser Energy | |
| CN103006205B (en) | Device and method for electrocardiosignal baseline recognition | |
| CN114580475A (en) | Electrocardio noise removing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |