CN109770859A - The treating method and apparatus of electrocardiosignal, storage medium, processor - Google Patents
The treating method and apparatus of electrocardiosignal, storage medium, processor Download PDFInfo
- Publication number
- CN109770859A CN109770859A CN201910244962.4A CN201910244962A CN109770859A CN 109770859 A CN109770859 A CN 109770859A CN 201910244962 A CN201910244962 A CN 201910244962A CN 109770859 A CN109770859 A CN 109770859A
- Authority
- CN
- China
- Prior art keywords
- electrocardiosignal
- poincare
- feature
- offset sequence
- offset
- 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.)
- Withdrawn
Links
Abstract
The invention discloses a kind for the treatment of method and apparatus of electrocardiosignal, storage medium, processors.Wherein, this method comprises: obtaining the first electrocardiosignal in the first preset time period, wherein the first electrocardiosignal is used to characterize the electrocardiosignal variation between sequences of ventricular depolarization and multipole；Obtain the offset sequence of the first electrocardiosignal, wherein for the multiple deviants for including in offset sequence for characterizing variation of first electrocardiosignal relative to reference voltage, reference voltage is extracted from electrocardiosignal；Feature extraction is carried out to the offset sequence of the first electrocardiosignal, obtains feature vector, wherein feature vector includes: feature, morphological feature and global characteristics based on Poincare section；Feature vector is identified, the offset type of the first electrocardiosignal is obtained.The processing method that the present invention solves electrocardiosignal in the prior art is big by a human specific and lead specific effect, leads to the technical problem that treatment effeciency is low.
Description
Technical field
The present invention relates to electrocardiogram fields, are situated between in particular to a kind for the treatment of method and apparatus of electrocardiosignal, storage
Matter, processor.
Background technique
Electrocardiogram (electrocardiogram, ECG) is that a kind of transthoracic electricity that heart is recorded as unit of the time is raw
Reason activity, shows due to cardiomotility and places the potential change between the skin different parts of electrode.ECG signal usually by
P wave, QRS structural body, T wave component reflect the process of heart each position depolarization and multipole.ST sections refer to sequences of ventricular depolarization and multipole
Between ECG Change, normally isoelectric period.When myocardial ischemia occurs, since ischemic and non-ischemic iuntercellular are deposited
In potential difference, lead to damaging electric current, so that the ST section of ECG signal is changed, lifts or move down compared on reference line.In ST section
Lift be common in transmural myocardial ischemia or variant angina pectoris patient, ST sections move down be common in endocardium cell ischemic or
Stable type and unstable angina pectoris.
Common ST field offset detection method has perceptron, PCA and KLT transformation and self-organizing network etc., these methods
It is maximum the disadvantage is that there is black box property, testing result interpretation is weaker.ECG signal is vulnerable to baseline drift, muscle twitches, work
The influence of noises such as frequency interference, and ST field offset is different in the performance of different leads, complex shape is changeable, and there are individual difference, ST
Field offset detection is difficult to accomplish precise and stable.
It is big by a human specific and lead specific effect for the processing method of electrocardiosignal in the prior art, cause to locate
The problem of managing low efficiency, currently no effective solution has been proposed.
Summary of the invention
The embodiment of the invention provides a kind for the treatment of method and apparatus of electrocardiosignal, storage medium, processors, at least
The processing method for solving electrocardiosignal in the prior art is big by a human specific and lead specific effect, causes treatment effeciency low
The technical issues of.
According to an aspect of an embodiment of the present invention, a kind of processing method of electrocardiosignal is provided, comprising: obtain first
The first electrocardiosignal in preset time period, wherein the first electrocardiosignal is used to characterize the electrocardio between sequences of ventricular depolarization and multipole
Signal intensity；Obtain the offset sequence of the first electrocardiosignal, wherein the multiple deviants for including in offset sequence are for characterizing the
Variation of one electrocardiosignal relative to reference voltage, reference voltage are extracted from electrocardiosignal；To the first electrocardiosignal
Offset sequence carry out feature extraction, obtain feature vector, wherein feature vector includes: feature, shape based on Poincare section
State feature and global characteristics；Feature vector is identified, the offset type of the first electrocardiosignal is obtained.
Further, morphological feature includes: deviant, slope, signal normalization slope and the intercept of the first electrocardiosignal；
Global characteristics include: the behavioral characteristics and average characteristics of the second preset time period.
Further, feature extraction is carried out to the offset sequence of the first electrocardiosignal, obtains the spy based on Poincare section
Sign, comprising: the offset sequence based on the first electrocardiosignal constructs Poincare figure, wherein Poincare figure is for characterizing the first electrocardio
Correlation in the offset sequence of signal between the deviant of adjacent moment；Obtain being averaged for all the points and target in Poincare figure
Distance obtains the feature based on Poincare section, wherein target includes at least one following: origin, the Poincare of Poincare figure
The diagonal line of second quadrant and fourth quadrant in figure.
Further, the offset sequence based on the first electrocardiosignal constructs Poincare figure, comprising: obtains the first electrocardio letter
Number offset sequence in adjacent moment two deviants；Two deviants based on adjacent moment construct cartesian coordinate system
In coordinate, wherein in two deviants of adjacent moment the deviant at previous moment as abscissa, the latter moment
Deviant is as ordinate；Based on the coordinate in cartesian coordinate system, Poincare figure is constructed.
Further, in the offset sequence based on the first electrocardiosignal, before constructing Poincare figure, the above method is also wrapped
It includes: obtaining the first electrocardiosignal that continuous multiple hearts are clapped in electrocardiosignal；To the inclined of the first electrocardiosignal that continuously multiple hearts are clapped
It moves sequence to be combined, the offset sequence after being combined；Based on the offset sequence after combination, Poincare figure is constructed.
Further, feature vector is identified, obtains the offset type of the first electrocardiosignal, comprising: using at random
Forest model identifies feature vector, obtains the offset type of the first electrocardiosignal.
Further, Random Forest model is to be trained to obtain by open PostgreSQL database.
According to another aspect of an embodiment of the present invention, a kind of processing unit of electrocardiosignal is additionally provided, comprising: first obtains
Modulus block, for obtaining the first electrocardiosignal in the first preset time period, wherein the first electrocardiosignal is removed for characterizing ventricle
Electrocardiosignal variation between pole and multipole；Second obtains module, for obtaining the offset sequence of the first electrocardiosignal, wherein
For characterizing variation of first electrocardiosignal relative to reference voltage, reference voltage is the multiple deviants for including in offset sequence
It is extracted from electrocardiosignal；Extraction module carries out feature extraction for the offset sequence to the first electrocardiosignal, obtains spy
Levy vector, wherein feature vector includes: feature, morphological feature and global characteristics based on Poincare section；Identification module is used
It is identified in feature vector, obtains the offset type of the first electrocardiosignal.
Further, morphological feature includes: deviant, slope, signal normalization slope and the intercept of the first electrocardiosignal；
Global characteristics include: the behavioral characteristics and average characteristics of the second preset time period.
Further, extraction module includes: building submodule, for the offset sequence based on the first electrocardiosignal, building
Poincare figure, wherein Poincare figure is used to characterize in the offset sequence of the first electrocardiosignal between the deviant of adjacent moment
Correlation；Acquisition submodule is obtained for obtaining the average distance of all the points and target in Poincare figure based on Poincare section
Feature, wherein target includes at least one following: the second quadrant and fourth quadrant in the origin of Poincare figure, Poincare figure
Diagonal line.
Further, building submodule includes: acquiring unit, adjacent in the offset sequence for obtaining the first electrocardiosignal
Two deviants at moment；First construction unit constructs in cartesian coordinate system for two deviants based on adjacent moment
Coordinate, wherein in two deviants of adjacent moment the deviant at previous moment as abscissa, the latter moment it is inclined
Shifting value is as ordinate；Second construction unit, for constructing Poincare figure based on the coordinate in cartesian coordinate system.
According to another aspect of an embodiment of the present invention, a kind of storage medium is additionally provided, storage medium includes the journey of storage
Sequence, wherein equipment where control storage medium executes the processing method of above-mentioned electrocardiosignal in program operation.
According to another aspect of an embodiment of the present invention, a kind of processor is additionally provided, processor is used to run program,
In, program executes the processing method of above-mentioned electrocardiosignal when running.
In embodiments of the present invention, after getting the first electrocardiosignal in the first preset time period, first is obtained
The offset sequence of electrocardiosignal, and feature extraction is carried out to the offset sequence of the first electrocardiosignal, it obtains including: based on Poincare
Feature, the feature vector of morphological feature and global characteristics in section, identify finally by feature vector, obtain final
Offset type.Compared with prior art, the feature of extraction is not merely based on morphologic ST field offset feature, it can also be ensured that more
The ST field offset variation of sample is added, and by extracting feature and global characteristics based on Poincare section, has reached raising Shandong
Stick avoids being influenced by individual difference, promotes the technical effect of accuracy in detection, and then solves prior art center telecommunications
Number processing method it is big by a human specific and lead specific effect, lead to the technical problem that treatment effeciency is low.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes part of this application, this hair
Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In the accompanying drawings:
Fig. 1 is a kind of flow chart of the processing method of electrocardiosignal according to an embodiment of the present invention；
Fig. 2 (a) is a kind of schematic diagram of the Poincare figure of optional ST field offset according to an embodiment of the present invention；
Fig. 2 (b) is a kind of schematic diagram of optional ST sections ECG sequence raised according to an embodiment of the present invention；
Fig. 2 (c) is a kind of schematic diagram of optional normal ECG sequence according to an embodiment of the present invention；
Fig. 2 (d) is a kind of schematic diagram of optional ST sections ECG sequence moved down according to an embodiment of the present invention；
Fig. 3 is a kind of flow chart of the processing method of optional electrocardiosignal according to an embodiment of the present invention；And
Fig. 4 is a kind of schematic diagram of the processing unit of electrocardiosignal according to an embodiment of the present invention.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people
The model that the present invention protects all should belong in member's every other embodiment obtained without making creative work
It encloses.
It should be noted that description and claims of this specification and term " first " in above-mentioned attached drawing, "
Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way
Data be interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein can in addition to illustrating herein or
Sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that cover
Cover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to
Step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, product
Or other step or units that equipment is intrinsic.
Embodiment 1
According to embodiments of the present invention, a kind of embodiment of the processing method of electrocardiosignal is provided, it should be noted that
The step of process of attached drawing illustrates can execute in a computer system such as a set of computer executable instructions, also,
It, in some cases, can be to be different from shown in sequence execution herein although logical order is shown in flow charts
The step of out or describing.
The processing method of electrocardiosignal provided by the embodiments of the present application can be executed by the processing equipment of electrocardiosignal, the heart
The processing equipment of electric signal can realize that the processing equipment of the electrocardiosignal can be two by way of software and/or hardware
A or multiple physical entities are constituted, and are also possible to a physical entity and are constituted, for example, the processing equipment of the electrocardiosignal can be
Computer, mobile phone, plate etc., the present embodiment is not especially limited this.Or the processing equipment of the electrocardiosignal can be storage
Program in the processor of the equipment such as computer, mobile phone, plate.
Fig. 1 is a kind of flow chart of the processing method of electrocardiosignal according to an embodiment of the present invention, as shown in Figure 1, the party
Method includes the following steps:
Step S102 obtains the first electrocardiosignal in the first preset time period, wherein the first electrocardiosignal is for characterizing
Electrocardiosignal variation between sequences of ventricular depolarization and multipole.
Specifically, in order to realize the purpose to the detection of ST field offset, the first preset time period in above-mentioned steps be can be
The ST sections of corresponding periods in electrocardiogram can extract ST sections of ECG signal from entire ECG signal, obtain in above-mentioned steps
The first electrocardiosignal, that is, obtaining ST sections.
It is alternatively possible to be achieved by the steps of ST sections of extraction: initial electrocardiosignal is obtained, for example, obtaining raw ECG
Signal.
Initial electrocardiosignal is pre-processed, the electrocardiosignal that obtains that treated, specifically, pretreatment successively can wrap
It includes multiple steps: according to preset sample frequency, frequency reducing being carried out to initial electrocardiosignal, the electrocardiosignal after obtaining frequency reducing is preset
Sample frequency can be unified for 250Hz；Denoising is carried out to the electrocardiosignal after frequency reducing based on Algorithms of Discrete Wavelet Transform, is obtained
To treated electrocardiosignal, wherein the kernel of Algorithms of Discrete Wavelet Transform and order are different, to eliminate original ECG signal
Noise, avoid electrocardiosignal vulnerable to myoelectricity interference, Hz noise or generate the interference such as baseline drift, influence the detection of ST field offset
Accuracy.
Obtain the first electrocardiosignal in treated electrocardiosignal in the first preset time period.Specifically, Ke Yitong
It crosses following steps and obtains ST sections: utilizing Pan-Thompkins algorithm detection QRS structural body position.Position based on QRS structural body
It sets, at the time of determining key point in treated electrocardiosignal, wherein key point includes at least: terminal (the i.e. J of QRS structural body
Point), key point can also include P wave terminal, Q wave starting point and T wave.Obtain the first preset time period in treated electrocardiosignal
The first interior electrocardiosignal, wherein to when being preset with interval at the time of key point at the time of the first preset time period is key point
Between at the time of.For example, J point can be extracted to the ECG signal in the interval J+0.08s as ST sections, that is, pre- in the present embodiment
If the time is 0.08s.
Step S104 obtains the offset sequence of the first electrocardiosignal, wherein the multiple deviants for including in offset sequence are used
In characterizing variation of first electrocardiosignal relative to reference voltage, reference voltage is extracted from electrocardiosignal.
Specifically, the reference voltage in above-mentioned steps can be the point voltage of the Q in ECG signal, opposite by extracting ST sections
In the variation of Q point voltage, available ST field offset.
Step S106 carries out feature extraction to the offset sequence of the first electrocardiosignal, obtains feature vector, wherein feature
Vector includes: feature, morphological feature and global characteristics based on Poincare section.
Specifically, morphological feature may include: the deviant of the first electrocardiosignal, slope, signal normalization slope and cut
Away from for example, each heart bat ST field offset value, slope, signal normalization ST field offset slope, intercept, may further clap each heart
Feature average value is intuitively demonstrated by the direction of ST field offset, degree as final morphological feature, morphological feature.Due to difference
The feature difference grade of Lead ST segment offset is larger, therefore, it is necessary to introduce lead information, that is, introducing global characteristics, global characteristics
It may include: the behavioral characteristics and average characteristics of the second preset time period, for example, dynamic RR interphase, average RR interphase feature.But
It is not limited only to this, the feature of other characterizations ST, equipotential line, T wave conversion can also be added.
It is alternatively possible to get the feature based on Poincare section of ST offset as follows: based on first heart
The offset sequence of electric signal constructs Poincare figure, wherein Poincare figure is used to characterize phase in the offset sequence of the first electrocardiosignal
Correlation between the deviant at adjacent moment；The average distance for obtaining all the points and target in Poincare figure obtains adding based on huge
The feature in Lay section, wherein target includes at least one following: the second quadrant and the in the origin of Poincare figure, Poincare figure
The diagonal line of four-quadrant.
Further, the building process of Poincare figure is as follows: obtaining adjacent moment in the offset sequence of the first electrocardiosignal
Two deviants；Two deviants based on adjacent moment construct the coordinate in cartesian coordinate system, wherein adjacent moment
Two deviants in the previous moment deviant as abscissa, the deviant at the latter moment is as ordinate；It is based on
Coordinate in cartesian coordinate system constructs Poincare figure.
It should be noted that in order to extract the feature based on Poincare section, and improve the robust of entire processing method
Property, the first electrocardiosignal that multiple continuous hearts are clapped in available electrocardiosignal is specifically, in the present embodiment, continuous with 5
The heart is illustrated for clapping, and the every 5 continuous hearts, which are clapped, is used as a detection section.Then the first electrocardiosignal continuous multiple hearts clapped
Offset sequence be combined, the offset sequence after being combined；Based on the offset sequence after combination, Poincare figure is constructed.
Specifically, the ST field offset sequence that each heart is clapped can be reconfigured, constructs its Poincare figure, Poincare figure indicates
Correlation in sequence between adjacent moment variable, building process are as follows: be equipped with sequence { xi(i=1,2 ..., n, then by sequence
Current value xi as the x-axis variable in cartesian coordinate system, subsequent time sequence xi+1 is as y-axis variable, in this way, by point
(xi,xi+1) (i=1,2 ..., n-1) sequence { x can be constructediPoincare figure.For example, Fig. 2 shows European ST-T
The typical Poincare figure of e0105 sample ST field offset curve in database, wherein circle is for indicating that ST sections are raised heart bat, " * "
For indicating that the normal heart is clapped, triangle is for indicating that ST moves down heart bat.
As shown in Figure 2, for the normal heart clap and ST section raise the heart bat for, Poincare scatterplot be concentrated mainly on one as
Limit, and move down heart bat for ST section, Poincare scatterplot is concentrated mainly on third quadrant, this definition and Fig. 2 with ST field offset
(b), (c), signal performance is consistent in (d).When ST field offset electric power reference line is remoter, corresponding Poincare scatterplot is from origin
It is remoter, therefore, Poincare figure scatterplot is extracted to the average distance d of origin as the feature for describing ST field offset degree.
In addition, ST sections usually present tiltedly or under ramp-like offset such as Fig. 2 (b), shown in (d), it is assumed that ST field offset { xi(i=
1,2 ..., n) linearly upper oblique or oblique offset, equation of change are as follows:
Y=kx+b, k ∈ R,
Then to its any Poincare point (yi, yi+1) i.e. (kxi+ b, kxi+1+ b), (yj, yj+1) i.e. (kxj+ b, kxj+1+ b), intend
Close Poincare figure point first order curve, curvilinear equation are as follows:
Substitute point (yi, yi+1), it obtains
S=k.
Therefore the intercept for extracting Poincare point single order fitting a straight line can reflect the degree of oblique type or oblique type ST field offset.
In addition, it can be seen from Fig. 2 (a) all Poincares o'clock to the cornerwise distance of two four-quadrants also with the degree of ST field offset and
Variation, therefore the average distance for extracting Poincare point to back-diagonal describes ST field offset degree.
It is extracted on the first electrocardiosignal that the single heart is clapped and it should also be noted that, morphological feature can be by continuous 5
The feature that a heart is clapped takes mean value；Global characteristics can be to be got on ECG signal record of a patient, and one
ECG signal record includes that thousands of a hearts are clapped.
Step S108, identifies feature vector, obtains the offset type of the first electrocardiosignal.
Optionally, feature vector is identified using Random Forest model, obtains the offset type of the first electrocardiosignal.
But it is not limited only to this, above-mentioned identifying purpose also may be implemented in other separation algorithms, such as support vector machines, xgboost etc..
Specifically, random forest, which helps, solves the problems, such as that single classifier generalization ability is insufficient, there is higher accuracy rate
And recall rate.Random Forest model is to be trained to obtain by open PostgreSQL database, wherein open PostgreSQL database can be with
It is in European ST-T database.European ST-T database is by including 90 length 2 hours, 2 lead ECG signals.
This research is under the premise of guaranteeing that training set, test set data do not derive from same individual, in the 27 signals composition randomly selected
Test set on 10 test results such as table 1, three classes ST field offset identify sensitivity it is higher, influenced by individual difference small；And
Variance is smaller, and algorithm is more stable.
Table 1
| It identifies sensitivity (average value ± variance) | |
| It is ST sections normal | 0.8515±0.0031 |
| ST sections are raised | 0.8694±0.0043 |
| ST sections move down | 0.8879±0.0043 |
It should be noted that in order to further increase ST field offset detection accuracy, available multiple detection sections, so
Multi-feature extraction is carried out to each detection section afterwards, extracts whole feature constitutive characteristic vectors and then the utilization of each detection section
Random forests algorithm identifies ST field offset.The feature vector calculating of each detection section only needs 0.031s.
Fig. 3 is a kind of flow chart of the processing method of optional electrocardiosignal according to an embodiment of the present invention, such as Fig. 3 institute
Show, after getting original ECG signal, can be pre-processed, after pretreatment, multiple features is carried out to each detection section and are mentioned
It takes, after whole features (including label and feature) the constitutive characteristic vector for extracting each detection section, with random forests algorithm
ST field offset is identified, estimation label is obtained, namely obtains the classification of ST field offset.
The above embodiments of the present application provide scheme, get the first electrocardiosignal in the first preset time period it
Afterwards, the offset sequence of the first electrocardiosignal is obtained, and feature extraction is carried out to the offset sequence of the first electrocardiosignal, is wrapped
Include: the feature vector of feature, morphological feature and global characteristics based on Poincare section is known finally by feature vector
Not, final offset type is obtained.Compared with prior art, it is special to be not merely based on morphologic ST field offset for the feature of extraction
Sign, it can also be ensured that more various ST field offset variation, and by extracting feature and overall situation spy based on Poincare section
Sign, has reached raising robustness, has avoided being influenced by individual difference, promoted the technical effect of accuracy in detection, and then solved
The processing method of electrocardiosignal is big by a human specific and lead specific effect in the prior art, leads to the skill that treatment effeciency is low
Art problem.
Embodiment 2
According to embodiments of the present invention, a kind of embodiment of the processing unit of electrocardiosignal is additionally provided.Fig. 4 is according to this hair
A kind of schematic diagram of the processing unit of electrocardiosignal of bright embodiment.The processing unit of electrocardiosignal provided in this embodiment can be with
It is integrated in the processing equipment of electrocardiosignal, as shown in figure 4, the device specifically includes: first, which obtains module 42, second, obtains mould
Block 44, extraction module 46 and identification module 48.
Wherein, first module 42 is obtained, for obtaining the first electrocardiosignal in the first preset time period, wherein first
Electrocardiosignal is used to characterize the electrocardiosignal variation between sequences of ventricular depolarization and multipole；Second obtains module 44, for obtaining first
The offset sequence of electrocardiosignal, wherein the multiple deviants for including in offset sequence for characterize the first electrocardiosignal relative to
The variation of reference voltage, reference voltage are extracted from electrocardiosignal；Extraction module 46, for the first electrocardiosignal
Offset sequence carries out feature extraction, obtains feature vector, wherein feature vector includes: feature, form based on Poincare section
Feature and global characteristics；Identification module 48 obtains the offset type of the first electrocardiosignal for identifying to feature vector.
Technical solution provided in this embodiment obtains after getting the first electrocardiosignal in the first preset time period
The offset sequence of the first electrocardiosignal is taken, and feature extraction is carried out to the offset sequence of the first electrocardiosignal, obtains including: to be based on
Feature, the feature vector of morphological feature and global characteristics of Poincare section, identify finally by feature vector, obtain
Final offset type.Compared with prior art, the feature of extraction is not merely based on morphologic ST field offset feature, can be with
Guarantee more various ST field offset variation, and by extracting feature and global characteristics based on Poincare section, reaches
Robustness is improved, avoids being influenced by individual difference, promotes the technical effect of accuracy in detection, and then solve in the prior art
The processing method of electrocardiosignal is big by a human specific and lead specific effect, leads to the technical problem that treatment effeciency is low.
Optionally, morphological feature includes: deviant, slope, signal normalization slope and the intercept of the first electrocardiosignal；Entirely
Office's feature includes: the behavioral characteristics and average characteristics of the second preset time period.
Optionally, extraction module includes: building submodule, and for the offset sequence based on the first electrocardiosignal, building is huge
Jia Laitu, wherein Poincare figure is used to characterize the phase in the offset sequence of the first electrocardiosignal between the deviant of adjacent moment
Guan Xing；First acquisition submodule obtains cutting based on Poincare for obtaining the average distance of all the points and target in Poincare figure
The feature in face, wherein target includes at least one following: the second quadrant and four-quadrant in the origin of Poincare figure, Poincare figure
The diagonal line of limit.
Optionally, building submodule includes: acquiring unit, when adjacent in the offset sequence for obtaining the first electrocardiosignal
Two deviants carved；First construction unit constructs in cartesian coordinate system for two deviants based on adjacent moment
Coordinate, wherein the deviant at previous moment is as abscissa, the offset at the latter moment in two deviants of adjacent moment
Value is used as ordinate；Second construction unit, for constructing Poincare figure based on the coordinate in cartesian coordinate system.
Optionally, extraction module further include: the second acquisition submodule, for obtaining multiple continuous hearts bats in electrocardiosignal
First electrocardiosignal；Submodule is combined, the offset sequence of the first electrocardiosignal for clapping continuous multiple hearts is combined, and is obtained
Offset sequence after to combination；Submodule is constructed, for constructing Poincare figure based on the offset sequence after combination.
Optionally, identification module is also used to identify feature vector using Random Forest model, obtains the first electrocardio
The offset type of signal.
Optionally, Random Forest model is to be trained to obtain by open PostgreSQL database.
Embodiment 3
According to embodiments of the present invention, a kind of embodiment of storage medium is additionally provided, storage medium includes the program of storage,
Wherein, equipment where controlling storage medium when program is run executes the processing method of the electrocardiosignal in above-described embodiment 1.
Embodiment 4
According to embodiments of the present invention, a kind of embodiment of processor is additionally provided, processor is for running program, wherein
Program executes the processing method of the electrocardiosignal in above-described embodiment 1 when running.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
In the above embodiment of the invention, it all emphasizes particularly on different fields to the description of each embodiment, does not have in some embodiment
The part of detailed description, reference can be made to the related descriptions of other embodiments.
In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others
Mode is realized.Wherein, the apparatus embodiments described above are merely exemplary, such as the division of the unit, Ke Yiwei
A kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or components can combine or
Person is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Between coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication link of unit or module
It connects, can be electrical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
On unit.It can some or all of the units may be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can for personal computer, server or network equipment etc.) execute each embodiment the method for the present invention whole or
Part steps.And storage medium above-mentioned includes: that USB flash disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic or disk etc. be various to can store program code
Medium.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (13)
1. a kind of processing method of electrocardiosignal characterized by comprising
Obtain the first electrocardiosignal in the first preset time period, wherein first electrocardiosignal is for characterizing sequences of ventricular depolarization
Electrocardiosignal variation between multipole；
Obtain the offset sequence of first electrocardiosignal, wherein the multiple deviants for including in the offset sequence are used for table
Variation of first electrocardiosignal relative to reference voltage is levied, the reference voltage is extracted from electrocardiosignal；
Feature extraction is carried out to the offset sequence of first electrocardiosignal, obtains feature vector, wherein described eigenvector packet
It includes: feature, morphological feature and global characteristics based on Poincare section；
Described eigenvector is identified, the offset type of first electrocardiosignal is obtained.
2. the method according to claim 1, wherein the morphological feature includes: first electrocardiosignal
Deviant, slope, signal normalization slope and intercept；The global characteristics include: the second preset time period behavioral characteristics and
Average characteristics.
3. according to the method described in claim 2, it is characterized in that, the offset sequence to first electrocardiosignal carries out feature
It extracts, obtains the feature based on Poincare section, comprising:
Based on the offset sequence of first electrocardiosignal, Poincare figure is constructed, wherein the Poincare figure is described for characterizing
Correlation in the offset sequence of first electrocardiosignal between the deviant of adjacent moment；
The average distance for obtaining all the points and target in the Poincare figure obtains the feature based on Poincare section,
In, the target includes at least one following: the second quadrant and four-quadrant in the origin of the Poincare figure, the Poincare figure
The diagonal line of limit.
4. according to the method described in claim 3, it is characterized in that, the offset sequence based on first electrocardiosignal,
Construct Poincare figure, comprising:
Obtain two deviants of adjacent moment in the offset sequence of first electrocardiosignal；
Two deviants based on the adjacent moment construct the coordinate in cartesian coordinate system, wherein the adjacent moment
The deviant at previous moment is as abscissa in two deviants, and the deviant at the latter moment is as ordinate；
Based on the coordinate in the cartesian coordinate system, the Poincare figure is constructed.
5. according to the method described in claim 3, it is characterized in that, in the offset sequence based on first electrocardiosignal,
Before constructing Poincare figure, the method also includes:
Obtain the first electrocardiosignal that continuous multiple hearts are clapped in electrocardiosignal；
The offset sequence for the first electrocardiosignal that continuous multiple hearts are clapped is combined, the offset sequence after being combined；
Based on the offset sequence after the combination, the Poincare figure is constructed.
6. according to the method described in claim 2, obtaining described first it is characterized in that, identify to described eigenvector
The offset type of electrocardiosignal, comprising:
Described eigenvector is identified using Random Forest model, obtains the offset type of first electrocardiosignal.
7. according to the method described in claim 6, it is characterized in that, the Random Forest model is by open PostgreSQL database
It is trained to obtain.
8. a kind of processing unit of electrocardiosignal characterized by comprising
First obtains module, for obtaining the first electrocardiosignal in the first preset time period, wherein first electrocardiosignal
For characterizing the variation of the electrocardiosignal between sequences of ventricular depolarization and multipole；
Second obtains module, for obtaining the offset sequence of first electrocardiosignal, wherein include in the offset sequence
Multiple deviants for characterizing variation of first electrocardiosignal relative to reference voltage, the reference voltage are believed from electrocardio
It is extracted in number；
Extraction module carries out feature extraction for the offset sequence to first electrocardiosignal, obtains feature vector, wherein
Described eigenvector includes: feature, morphological feature and global characteristics based on Poincare section；
Identification module obtains the offset type of first electrocardiosignal for identifying to described eigenvector.
9. device according to claim 8, which is characterized in that the morphological feature includes: first electrocardiosignal
Deviant, slope, signal normalization slope and intercept；The global characteristics include: the second preset time period behavioral characteristics and
Average characteristics.
10. device according to claim 9, which is characterized in that the extraction module includes:
Submodule is constructed, for the offset sequence based on first electrocardiosignal, constructs Poincare figure, wherein the Pang adds
Lay figure is used to characterize the correlation in the offset sequence of first electrocardiosignal between the deviant of adjacent moment；
Acquisition submodule, for obtaining the average distance of all the points and target in the Poincare figure, obtain it is described based on huge plus
The feature in Lay section, wherein the target includes at least one following: in the origin of the Poincare figure, the Poincare figure
The diagonal line of second quadrant and fourth quadrant.
11. device according to claim 10, which is characterized in that the building submodule includes:
Acquiring unit, two deviants of adjacent moment in the offset sequence for obtaining first electrocardiosignal；
First construction unit constructs the coordinate in cartesian coordinate system for two deviants based on the adjacent moment,
In, the deviant at previous moment is as abscissa, the deviant at the latter moment in two deviants of the adjacent moment
As ordinate；
Second construction unit, for constructing the Poincare figure based on the coordinate in the cartesian coordinate system.
12. a kind of storage medium, which is characterized in that the storage medium includes the program of storage, wherein run in described program
When control the storage medium where equipment perform claim require any one of 1 to 7 described in electrocardiosignal processing method.
13. a kind of processor, which is characterized in that the processor is for running program, wherein right of execution when described program is run
Benefit require any one of 1 to 7 described in electrocardiosignal processing method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910244962.4A CN109770859A (en) | 2019-03-28 | 2019-03-28 | The treating method and apparatus of electrocardiosignal, storage medium, processor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910244962.4A CN109770859A (en) | 2019-03-28 | 2019-03-28 | The treating method and apparatus of electrocardiosignal, storage medium, processor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109770859A true CN109770859A (en) | 2019-05-21 |
Family
ID=66491051
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910244962.4A Withdrawn CN109770859A (en) | 2019-03-28 | 2019-03-28 | The treating method and apparatus of electrocardiosignal, storage medium, processor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109770859A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110522440A (en) * | 2019-08-12 | 2019-12-03 | 广州视源电子科技股份有限公司 | Electrocardiosignal identification device based on grouping convolutional neural networks |
| CN111067508A (en) * | 2019-12-31 | 2020-04-28 | 深圳安视睿信息技术股份有限公司 | Non-intervention monitoring and evaluating method for hypertension in non-clinical environment |
| CN111568410A (en) * | 2020-05-18 | 2020-08-25 | 中国矿业大学 | Electrocardiogram data classification method based on 12-lead and convolutional neural network |
| CN111898694A (en) * | 2020-08-07 | 2020-11-06 | 广东电网有限责任公司计量中心 | Non-invasive load identification method and device based on random tree classification |
| CN116746900A (en) * | 2022-10-19 | 2023-09-15 | 荣耀终端有限公司 | Arrhythmia identification method and wearable device |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101991413A (en) * | 2008-08-21 | 2011-03-30 | 迈瑞控股（香港）有限公司 | Systems and methods for quantifying and providing indicia of ST-segment resolution in ECG signal |
| CN102085095A (en) * | 2009-12-07 | 2011-06-08 | 深圳市新元素医疗技术开发有限公司 | Method, system and electrocardioscanner for detecting ST segment in electrocardiogram |
| CN104783782A (en) * | 2015-04-13 | 2015-07-22 | 深圳市飞马与星月科技研究有限公司 | Automatic detection method and device for electrocardiosignals |
| CN105030233A (en) * | 2015-07-08 | 2015-11-11 | 上海师范大学 | Method for recognizing ST segment of electrocardiosignal |
| CN105496402A (en) * | 2015-11-20 | 2016-04-20 | 北京理工大学 | Electrocardio feature analyzing method based on point diagram and symbolic dynamics |
| CN103857330B (en) * | 2011-10-12 | 2016-05-18 | 皇家飞利浦有限公司 | The method and system of the Electrocardiographic ST-section of all automatic measurement level in ECG monitoring in real time |
| CN105997055A (en) * | 2016-07-11 | 2016-10-12 | 吉林大学 | Automatic classification method, system and device of electrocardiosignal ST band |
| WO2017050321A1 (en) * | 2015-09-23 | 2017-03-30 | Biosign Medical Ug | Method and device for quantifying a respiratory sinus arrhythmia and use of said type of method or said type of device |
| CN106815570A (en) * | 2016-12-30 | 2017-06-09 | 广东技术师范学院 | A kind of T sections of recognition methods of electrocardiosignal ST based on dynamic pattern recognition |
| CN104783787B (en) * | 2015-04-24 | 2017-06-16 | 太原理工大学 | A kind of J wave detecting methods based on neutral net |
| US9955890B2 (en) * | 2009-05-05 | 2018-05-01 | Siemens Healthcare Gmbh | Heart electrophysiological signal analysis system |
| CN109077714A (en) * | 2018-07-05 | 2018-12-25 | 广州视源电子科技股份有限公司 | Signal recognition method, device, equipment and storage medium |
-
2019
- 2019-03-28 CN CN201910244962.4A patent/CN109770859A/en not_active Withdrawn
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101991413A (en) * | 2008-08-21 | 2011-03-30 | 迈瑞控股（香港）有限公司 | Systems and methods for quantifying and providing indicia of ST-segment resolution in ECG signal |
| US9955890B2 (en) * | 2009-05-05 | 2018-05-01 | Siemens Healthcare Gmbh | Heart electrophysiological signal analysis system |
| CN102085095A (en) * | 2009-12-07 | 2011-06-08 | 深圳市新元素医疗技术开发有限公司 | Method, system and electrocardioscanner for detecting ST segment in electrocardiogram |
| CN103857330B (en) * | 2011-10-12 | 2016-05-18 | 皇家飞利浦有限公司 | The method and system of the Electrocardiographic ST-section of all automatic measurement level in ECG monitoring in real time |
| CN104783782A (en) * | 2015-04-13 | 2015-07-22 | 深圳市飞马与星月科技研究有限公司 | Automatic detection method and device for electrocardiosignals |
| CN104783787B (en) * | 2015-04-24 | 2017-06-16 | 太原理工大学 | A kind of J wave detecting methods based on neutral net |
| CN105030233A (en) * | 2015-07-08 | 2015-11-11 | 上海师范大学 | Method for recognizing ST segment of electrocardiosignal |
| WO2017050321A1 (en) * | 2015-09-23 | 2017-03-30 | Biosign Medical Ug | Method and device for quantifying a respiratory sinus arrhythmia and use of said type of method or said type of device |
| CN105496402A (en) * | 2015-11-20 | 2016-04-20 | 北京理工大学 | Electrocardio feature analyzing method based on point diagram and symbolic dynamics |
| CN105997055A (en) * | 2016-07-11 | 2016-10-12 | 吉林大学 | Automatic classification method, system and device of electrocardiosignal ST band |
| CN106815570A (en) * | 2016-12-30 | 2017-06-09 | 广东技术师范学院 | A kind of T sections of recognition methods of electrocardiosignal ST based on dynamic pattern recognition |
| CN109077714A (en) * | 2018-07-05 | 2018-12-25 | 广州视源电子科技股份有限公司 | Signal recognition method, device, equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| HONGMEI WANG ET AL: "ST Segment Change Classification Based on Multiple Feature Extraction Using ECG", 《COMPUTING IN CARDIOLOGY 2018》 * |
| 范姝琼: "心电信号ST段的提取及识别算法研究", 《中国硕士学位论文数据库》 * |
| 霍铖宇: "基于庞加莱散点图和递归图的心率变异性研究", 《中国博士学位论文全文数据库 医药卫生科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110522440A (en) * | 2019-08-12 | 2019-12-03 | 广州视源电子科技股份有限公司 | Electrocardiosignal identification device based on grouping convolutional neural networks |
| CN111067508A (en) * | 2019-12-31 | 2020-04-28 | 深圳安视睿信息技术股份有限公司 | Non-intervention monitoring and evaluating method for hypertension in non-clinical environment |
| CN111568410A (en) * | 2020-05-18 | 2020-08-25 | 中国矿业大学 | Electrocardiogram data classification method based on 12-lead and convolutional neural network |
| CN111898694A (en) * | 2020-08-07 | 2020-11-06 | 广东电网有限责任公司计量中心 | Non-invasive load identification method and device based on random tree classification |
| CN111898694B (en) * | 2020-08-07 | 2021-09-17 | 广东电网有限责任公司计量中心 | Non-invasive load identification method and device based on random tree classification |
| CN116746900A (en) * | 2022-10-19 | 2023-09-15 | 荣耀终端有限公司 | Arrhythmia identification method and wearable device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109770859A (en) | The treating method and apparatus of electrocardiosignal, storage medium, processor | |
| Afkhami et al. | Cardiac arrhythmia classification using statistical and mixture modeling features of ECG signals | |
| Qiu et al. | Selective encryption on ECG data in body sensor network based on supervised machine learning | |
| Sellami et al. | A robust deep convolutional neural network with batch-weighted loss for heartbeat classification | |
| Anwar et al. | Arrhythmia classification of ECG signals using hybrid features | |
| Alyasseri et al. | Person identification using EEG channel selection with hybrid flower pollination algorithm | |
| Fratini et al. | Individual identification via electrocardiogram analysis | |
| Javadi et al. | Classification of ECG arrhythmia by a modular neural network based on mixture of experts and negatively correlated learning | |
| Dar et al. | ECG biometric identification for general population using multiresolution analysis of DWT based features | |
| Chamatidis et al. | Using deep learning neural networks for ECG based authentication | |
| CN109077714B (en) | Signal identification method, device, equipment and storage medium | |
| CN106377247A (en) | Feature selection-based arrhythmia classification method | |
| Kang et al. | A method of denoising multi-channel EEG signals fast based on PCA and DEBSS algorithm | |
| CN109077720B (en) | Signal processing method, device, equipment and storage medium | |
| CN110367936B (en) | Electrocardiosignal detection method and device | |
| CN109009073A (en) | Atrial fibrillation detection device and storage medium | |
| Smíšek et al. | SVM based ECG classification using rhythm and morphology features, cluster analysis and multilevel noise estimation | |
| Monsy et al. | EEG‐based biometric identification using frequency‐weighted power feature | |
| Zhao et al. | Towards deep learning-based detection scheme with raw ECG signal for wearable telehealth systems | |
| Alcaraz et al. | Short-time regularity assessment of fibrillatory waves from the surface ECG in atrial fibrillation | |
| CN108597600A (en) | Singly lead heart patch data intelligence diagnosis cloud computing system and its processing method | |
| Celler et al. | Low computational cost classifiers for ECG diagnosis using neural networks | |
| CN111931578A (en) | Electroencephalogram identification method based on minimum spanning tree and regional double-layer network | |
| Zubair et al. | Removal of motion artifacts from ECG signals by combination of recurrent neural networks and deep neural networks | |
| CN113229798B (en) | Model migration training method, device, computer equipment and readable storage medium |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190521 |
|
| WW01 | Invention patent application withdrawn after publication |